<?xmlversion='1.0' encoding='utf-8'?>version="1.0" encoding="UTF-8"?> <!DOCTYPE rfc [ <!ENTITY nbsp " "> <!ENTITY zwsp "​"> <!ENTITY nbhy "‑"> <!ENTITY wj "⁠"> ]><!-- This template is for creating an Internet Draft using xml2rfc, which is available here: http://xml.resource.org. --> <?xml-stylesheet type='text/xsl' href='rfc2629.xslt' ?> <!-- used by XSLT processors --> <!-- For a complete list and description of processing instructions (PIs), please see http://xml.resource.org/authoring/README.html. --> <!-- Below are generally applicable Processing Instructions (PIs) that most I-Ds might want to use. (Here they are set differently than their defaults in xml2rfc v1.32) --> <?rfc strict="yes" ?> <!-- give errors regarding ID-nits and DTD validation --> <!-- control the table of contents (ToC) --> <?rfc toc="yes"?> <!-- generate a ToC --> <?rfc tocdepth="4"?> <!-- the number of levels of subsections in ToC. default: 3 --> <!-- control references --> <?rfc symrefs="yes"?> <!-- use symbolic references tags, i.e, [RFC2119] instead of [1] --> <?rfc sortrefs="yes" ?> <!-- sort the reference entries alphabetically --> <!-- control vertical white space (using these PIs as follows is recommended by the RFC Editor) --> <?rfc compact="yes" ?> <!-- do not start each main section on a new page --> <?rfc subcompact="no" ?> <!-- keep one blank line between list items --> <!-- end of list of popular I-D processing instructions --><rfc xmlns:xi="http://www.w3.org/2001/XInclude" submissionType="IETF" category="exp" consensus="true" docName="draft-ietf-tsvwg-ecn-l4s-id-29" number="9331" ipr="trust200902" obsoletes="" updates=""submissionType="IETF"xml:lang="en" tocInclude="true" tocDepth="4" symRefs="true" sortRefs="true" version="3"><!-- xml2rfc v2v3 conversion 3.14.1 --> <!-- category values: std, bcp, info, exp, and historic ipr values: trust200902, noModificationTrust200902, noDerivativesTrust200902, or pre5378Trust200902 you can add the attributes updates="NNNN" and obsoletes="NNNN" they will automatically be output with "(if approved)" --> <!-- ***** FRONT MATTER ***** --><front><!-- The abbreviated title is used in the page header - it is only necessary if the full title is longer than 39 characters --><titleabbrev="L4S ECNabbrev="ECN Protocol forV Low Queuing Delay">ExplicitL4S"> The Explicit Congestion Notification (ECN) Protocol forVeryLowQueuing DelayLatency, Low Loss, and Scalable Throughput (L4S)</title> <seriesInfoname="Internet-Draft" value="draft-ietf-tsvwg-ecn-l4s-id-29"/>name="RFC" value="9331"/> <author fullname="Koen De Schepper" initials="K." surname="De Schepper"> <organization>Nokia Bell Labs</organization> <address> <postal> <street/> <city>Antwerp</city> <country>Belgium</country> </postal> <email>koen.de_schepper@nokia.com</email> <uri>https://www.bell-labs.com/about/researcher-profiles/koende_schepper/</uri> </address> </author> <author fullname="Bob Briscoe" initials="B."role="editor" surname="Briscoe">surname="Briscoe" role="editor"> <organization>Independent</organization> <address> <postal> <street/><country>UK</country><country>United Kingdom</country> </postal> <email>ietf@bobbriscoe.net</email> <uri>https://bobbriscoe.net/</uri> </address> </author><!-- <author fullname="Olga Bondarenko" initials="O." surname="Bondarenko"> <organization>Simula Research Lab</organization> <address> <postal> <street/> <city>Lysaker</city> <country>Norway</country> </postal> <email>olgabnd@gmail.com</email> <uri>https://www.simula.no/people/olgabo</uri> </address> </author> --> <!-- <author fullname="Ing-jyh Tsang" initials="I." surname="Tsang"> <organization>Nokia Bell Labs</organization> <address> <postal> <street/> <city>Antwerp</city> <country>Belgium</country> </postal> <email>ing-jyh.tsang@nokia.com</email> </address> </author> --><dateyear=""/> <area>Transport</area> <workgroup>Transport Services (tsv)</workgroup> <keyword>Internet-Draft</keyword> <keyword>I-D</keyword>year="2023" month="January" /> <area>tsv</area> <workgroup>tsvwg</workgroup> <keyword>Performance</keyword> <keyword>Queuing Delay</keyword> <keyword>One Way Delay</keyword> <keyword>Round-Trip Time</keyword> <keyword>RTT</keyword> <keyword>Jitter</keyword> <keyword>Congestion Control</keyword> <keyword>Congestion Avoidance</keyword> <keyword>Quality of Service</keyword> <keyword>QoS</keyword> <keyword>Quality of Experience</keyword> <keyword>QoE</keyword> <keyword>Active Queue Management</keyword> <keyword>AQM</keyword> <keyword>Explicit Congestion Notification</keyword> <keyword>ECN</keyword> <keyword>Burstiness</keyword> <abstract> <t>This specification defines the protocol to be used for a new network service calledlow latency, low lossLow Latency, Low Loss, andscalableScalable throughput (L4S). L4S uses an Explicit Congestion Notification (ECN) scheme at the IP layer that is similar to the original (or 'Classic') ECN approach, except as specified within. L4S uses'scalable''Scalable' congestion control, which induces much more frequent control signals from thenetworknetwork, and it responds to them with much more fine-grainedadjustments,adjustments so that very low (typically sub-millisecond on average) and consistently low queuing delay becomes possible for L4S traffic without compromising link utilization.ThusThus, even capacity-seeking (TCP-like) traffic can have high bandwidth and very low delay at the same time, even during periods of high traffic load.</t> <t>The L4S identifier defined in this document distinguishes L4S from 'Classic'(e.g. TCP-Reno-friendly)(e.g., TCP-Reno-friendly) traffic. Then, network bottlenecks can be incrementally modified to distinguish and isolate existing traffic that still follows the Classic behaviour, to prevent it from degrading the low queuing delay and low loss of L4S traffic. Thisexperimental trackExperimental specification defines the rules that L4S transports and network elements need to follow, with the intention that L4S flows neither harm each other's performance nor that of Classic traffic. It also suggests open questions to be investigated during experimentation. Examples of newactive queue managementActive Queue Management (AQM) marking algorithms andexamples ofnew transports (whether TCP-like orreal-time)real time) are specified separately.</t> </abstract> </front> <middle> <section anchor="l4sid_intro" numbered="true" toc="default"> <name>Introduction</name> <t>Thisexperimental trackExperimental specification defines the protocol to be used for a new network service calledlow latency, low lossLow Latency, Low Loss, andscalableScalable throughput (L4S). L4S uses an Explicit Congestion Notification (ECN) scheme at the IP layer with the same set of codepoint transitions as the original (or 'Classic')Explicit Congestion Notification (ECN <xrefECN <xref target="RFC3168"format="default"/>). RFC 3168 requiredformat="default"/>. <xref target="RFC3168"/> requires an ECN mark to be equivalent to a drop, both when applied in the network and when responded to by a transport. Unlike Classic ECNmarking:marking, i) the network applies L4S marking more immediately and more frequently thandrop;drop and ii) the transport response to each mark is reduced and smoothed relative to that for drop. The two changes counterbalance each other so that the throughput of an L4S flow will be roughly the same as a comparable non-L4S flow under the same conditions. Nonetheless, the much more frequent ECN control signals and the finer responses to these signals result in very low queuing delay without compromising link utilization, and this low delay can be maintained during high load. For instance, queuing delay under heavy and highly varying load with the example DCTCP/DualQ solution described below on a DSL or Ethernet link is sub-millisecond on average and roughly 1 to 2 milliseconds at the 99th percentile without losing linkutilization <xref target="DualPI2Linux" format="default"/>,utilization <xreftarget="DCttH19"target="L4Seval22" format="default"/> <xref target="DualPI2Linux" format="default"/>. Note that the queuing delay while waiting to acquire a shared medium such as wireless has to be added to the above. It is a different issue that needs to be addressed, but separately (seesection 6.3Section <xref target="RFC9330" section="6.3" sectionFormat="bare"/> of the L4Sarchitecture <xref target="I-D.ietf-tsvwg-l4s-arch"architecture <xref target="RFC9330" format="default"/>).</t> <t>L4S relies on'scalable''Scalable' congestion controls for these delay properties and for preserving low delay as flow rate scales, hence the name. The congestion control used in Data Center TCP (DCTCP) is an example of ascalableScalable congestion control, but DCTCP is applicable solely to controlled environments like datacentres <xrefcentres <xref target="RFC8257" format="default"/>, because it is too aggressive toco-existcoexist with existing TCP-Reno-friendly traffic.The DualQDual-Queue Coupled AQM, which is defined in a complementaryexperimental specification <xref target="I-D.ietf-tsvwg-aqm-dualq-coupled"Experimental specification <xref target="RFC9332" format="default"/>, is an AQM framework that enablesscalableScalable congestion controls derived from DCTCP toco-existcoexist with existing traffic, each getting roughly the same flow rate when they compete under similar conditions. Note that ascalableScalable congestion control is still not safe to deploy on the Internet unless it satisfies the requirements listed in <xref target="l4sid_transport_req" format="default"/>.</t> <t>L4S is not only for elastic (TCP-like) traffic--- there arescalableScalable congestion controls for real-time media, such as the L4S variant <xref target="SCReAM-L4S" format="default"/> of theSCReAM <xrefSCReAM <xref target="RFC8298" format="default"/>real-time media congestion avoidance techniqueRTP Media Congestion Avoidance Techniques (RMCAT). The factor that distinguishes L4S from Classic traffic is its behaviour in response to congestion. The transport wire protocol,e.g. TCP,e.g., TCP, QUIC,SCTP, DCCP,the Stream Control Transmission Protocol (SCTP), the Datagram Congestion Control Protocol (DCCP), or RTP/RTCP, is orthogonal (and therefore not suitable for distinguishing L4S from Classic packets).</t> <t>The L4S identifier defined in this document is the key piece that distinguishes L4S from 'Classic'(e.g. Reno-friendly)(e.g., Reno-friendly) traffic. Then, network bottlenecks can be incrementally modified to distinguish and isolate existing Classic traffic from L4S traffic, to prevent the former from degrading the very low queuing delay and loss of the newscalableScalable transports, without harming Classic performance at these bottlenecks. Although both sender and network deployment are required before any benefit, initial implementations of the separate parts of the system have been motivated by the potential performance benefits.</t> <section anchor="l4sid_problem" numbered="true" toc="default"> <name>Latency,LossLoss, and Scaling Problems</name> <t>Latency is becoming the critical performance factor for many(most?)(perhaps most) Internet applications,e.g. interactivee.g., interactive web, web services, voice, conversational video, interactive video, interactive remote presence, instant messaging, online gaming, remote desktop, cloud-based applications & services, and remote control of machinery and industrial processes. In many parts of the world, further increases in access networkbit ratebitrate offer diminishing returns <xref target="Dukkipati06" format="default"/>, whereas latency is still a multi-faceted problem. As a result, much has been done to reduce propagation time by placing caches or servers closer to users. However, queuing remains a major, albeit intermittent, component of latency.</t> <t>The Diffserv architecture provides ExpeditedForwarding <xrefForwarding (EF) <xref target="RFC3246"format="default"/>,format="default"/> so thatlow latencylow-latency traffic can jump the queue of other traffic. If growth in latency-sensitive applications continues, periods with solely latency-sensitive traffic will become increasingly common on links where traffic aggregation is low. During these periods, if all the traffic were marked for the same treatment, Diffserv would make no difference. The links with low aggregation also tend to become the path bottleneck under load, for instance, the access links dedicated to individual sites (homes, smallenterprisesenterprises, or mobile devices). So, instead of differentiation, it becomes imperative to remove the underlying causes of any unnecessary delay.</t> <t>The Bufferbloat project has shown thatexcessively-largeexcessively large buffering ('bufferbloat') has been introducing significantly more delay than the underlying propagation time <xref target="Bufferbloat" format="default"/>. These delays appear only intermittently -- only when a capacity-seeking(e.g. TCP)(e.g., TCP) flow is long enough for the queue to fill the buffer, causing every packet in other flows sharing the buffer to have to work its way through the queue.</t><t>Active queue management (AQM)<t>AQM was originally developed to solve this problem (and others). Unlike Diffserv, which gives low latency to some traffic at the expense of others, AQM controls latency for <em>all</em> traffic in a class. In general, AQM methods introduce an increasing level of discard from the buffer, the longer the queue persists above a shallow threshold. This gives sufficient signals to capacity-seeking(aka. greedy)(a.k.a. greedy) flows to keep the buffer empty for its intended purpose: absorbing bursts. However,RED <xref target="RFC2309" format="default"/>Random Early Detection (RED) and other algorithms from the 1990s were sensitive to their configuration and hard to setcorrectly. So,correctly <xref target="RFC7567" format="default"/>. So this form of AQM was not widely deployed.</t> <t>More recent state-of-the-art AQM methods, suchas FQ-CoDel <xrefas Flow Queue CoDel <xref target="RFC8290" format="default"/>,PIE <xrefProportional Integral controller Enhanced (PIE) <xref target="RFC8033"format="default"/>format="default"/>, or AdaptiveRED <xrefRED <xref target="ARED01" format="default"/>, are easier to configure, because they define the queuing threshold in time not bytes, so configuration is invariant whatever the link rate. However, the sawtoothing window of a Classic congestion control creates a dilemma for the operator:i)either i) configure a shallow AQM operatingpoint,point so the tips of the sawteeth cause minimal queue delay, but then the troughs underutilize thelink;link, or ii) configure the operating point deeper into thebuffer,buffer so the troughs utilize the link better, but then the tips cause more delay variation. Even with a perfectly tuned AQM, the additional queuing delay at the tips of the sawteeth will be of the same order as the underlying baseround tripround-trip time (RTT), thereby roughly doubling the totalround-trip time.</t>RTT.</t> <t>If a sender's own behaviour is introducing queuing delay variation, no AQM in the network can 'un-vary' the delay without significantly compromising link utilization. Evenflow-queuing (e.g. <xrefflow queuing (e.g., <xref target="RFC8290" format="default"/>), which isolates one flow from another, cannot isolate a flow from the delay variations it inflicts on itself. Therefore, those applications that need to seek out high bandwidth but also need low latency will have to migrate toscalableScalable congestion control, which uses much smaller sawtooth variations.</t> <t>Altering host behaviour is not enough on its own though. Even if hosts adoptlow latency scalablelow-latency Scalable congestion controls, they need to be isolated from the large queue variations induced by existing Classic congestion controls. L4S AQMs provide that latency isolation in thenetworknetwork, and the L4S identifier enables the AQMs to distinguish the two types ofpacketpackets that need to be isolated: L4S and Classic. L4S isolation can be achieved with a queue per flow(e.g. <xref(e.g., <xref target="RFC8290"format="default"/>)format="default"/>), but aDualQ <xref target="I-D.ietf-tsvwg-aqm-dualq-coupled"DualQ <xref target="RFC9332" format="default"/> issufficient,sufficient and actually gives better taillatency <xreflatency <xref target="DCttH19" format="default"/>. Both approaches are addressed in this document.</t> <t>The DualQ solution was developed to make very low latency available without requiring per-flow queues at every bottleneck. This was useful becauseper-flow-queuingper-flow queuing (FQ) has well-known downsides--- not least the need to inspecttransport layertransport-layer headers in the network, which makes it incompatible with privacy approaches such asIPSec VPN tunnels,IPsec Virtual Private Network (VPN) tunnels and incompatible withlink layerlink-layer queue management, wheretransport layertransport-layer headers can be hidden,e.g. 5G.</t>e.g., 5G.</t> <t>Latency is not the only concern addressed by L4S. It was known when TCP congestion avoidance was first developed that it would not scale to high bandwidth-delay products(footnote(see footnote 6 of Jacobson andKarels <xrefKarels <xref target="TCP-CA" format="default"/>). Given that Reno congestion control is already beyond its scaling range at regular broadbandbit-ratesbitrates over WAN distancesare already <xref<xref target="RFC3649"format="default"/> beyond the scaling range of Reno congestion control,format="default"/>, 'less unscalable'Cubic <xrefCUBIC <xref target="RFC8312" format="default"/> andCompound <xrefCompound <xref target="I-D.sridharan-tcpm-ctcp" format="default"/> variants of TCP have been successfully deployed. However, these are now approaching their scaling limits. Unfortunately, fullyscalableScalable congestion controls such asDCTCP <xrefDCTCP <xref target="RFC8257" format="default"/> outcompete Classic ECN congestion controls sharing the same queue, which is why they have been confined to private data centres or research testbeds.</t> <t>It turns out that thesescalableScalable congestion control algorithms that solve the latency problem can also solve the scalability problem of Classic congestion controls. The finer sawteeth in the congestion window (cwnd) have low amplitude, so they cause very little queuing delayvariationvariation, and the average time to recover from one congestion signal to the next (the average duration of each sawtooth) remains invariant, which maintains constant tight control asflow-rateflow rate scales. A backgroundpaper <xref target="DCttH19"paper <xref target="L4Seval22" format="default"/> gives the full explanation of why the design solves both the latency and the scaling problems, both in plain English and in more precise mathematical form. The explanation is summarized without the mathematics in Section4<xref target="RFC9330" section="4" sectionFormat="bare"/> of the L4Sarchitecture <xref target="I-D.ietf-tsvwg-l4s-arch"architecture <xref target="RFC9330" format="default"/>.</t> </section> <section anchor="l4sid_Terminology" numbered="true" toc="default"> <name>Terminology</name><t>The<t> The key words"MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY","<bcp14>MUST</bcp14>", "<bcp14>MUST NOT</bcp14>", "<bcp14>REQUIRED</bcp14>", "<bcp14>SHALL</bcp14>", "<bcp14>SHALL NOT</bcp14>", "<bcp14>SHOULD</bcp14>", "<bcp14>SHOULD NOT</bcp14>", "<bcp14>RECOMMENDED</bcp14>", "<bcp14>NOT RECOMMENDED</bcp14>", "<bcp14>MAY</bcp14>", and"OPTIONAL""<bcp14>OPTIONAL</bcp14>" in this document are to be interpreted as described inBCP 14BCP 14 <xreftarget="RFC2119" format="default"/>target="RFC2119"/> <xreftarget="RFC8174" format="default"/>target="RFC8174"/> when, and only when, they appear in all capitals, as shownhere.</t> <t>[Note to the RFC Editor (to be removed before publication as an RFC): The L4S architecture <xref target="I-D.ietf-tsvwg-l4s-arch" format="default"/> repeats the following definitions, which should be identical, except in the architecture Classic CC and Scalable CC are condensed because they refer to a section later in the architecture.]</t>here. </t> <dl newline="false" spacing="normal"> <dt>Classic Congestion Control:</dt> <dd>A congestion control behaviour that canco-existcoexist with standardReno <xrefReno <xref target="RFC5681" format="default"/> without causing significantly negative impact on its flowrate <xrefrate <xref target="RFC5033" format="default"/>. With Classic congestion controls, such as Reno orCubic,CUBIC, because flow rate has scaled since TCP congestion control was first designed in 1988, it now takes hundreds of round trips (and growing) to recover after a congestion signal (whether a loss or an ECN mark) as shown in the examples insection 5.1Section <xref target="RFC9330" section="5.1" sectionFormat="bare"/> of the L4Sarchitecture <xref target="I-D.ietf-tsvwg-l4s-arch"architecture <xref target="RFC9330" format="default"/> and in <xref target="RFC3649" format="default"/>. Therefore, control of queuing and utilization becomes very slack, and the slightest disturbances(e.g. from(e.g., from new flows starting) prevent a high rate from being attained.</dd> <dt>Scalable Congestion Control:</dt> <dd>A congestion control where the average time from one congestion signal to the next (the recovery time) remains invariant astheflow rate scales, all other factors being equal. This maintains the same degree of control overqueueingqueuing and utilization whatever the flow rate, as well as ensuring that high throughput is robust to disturbances. For instance, DCTCP averages 2 congestion signals perround-tripround trip, whatever the flow rate, as do other recently developedscalableScalable congestion controls,e.g. Relentless TCP <xrefe.g., Relentless TCP <xref target="I-D.mathis-iccrg-relentless-tcp" format="default"/>,TCPPrague for TCP and QUIC <xref target="I-D.briscoe-iccrg-prague-congestion-control"format="default"/>,format="default"/> <xref target="PragueLinux" format="default"/>,BBRv2the L4S ECN part of Bottleneck Bandwidth and Round-trip propagation time (BBRv2) <xref target="BBRv2"format="default"/>,format="default"/> <xref target="I-D.cardwell-iccrg-bbr-congestion-control"format="default"/>format="default"/>, and the L4S variant ofSCREAMSCReAM for real-timemedia <xrefmedia <xref target="SCReAM-L4S"format="default"/>, <xrefformat="default"/> <xref target="RFC8298" format="default"/>. See <xref target="l4sid_congestion_response" format="default"/> for more explanation.</dd> <dt>Classicservice:</dt>Service:</dt> <dd>The Classic service is intended for all the congestion control behaviours thatco-existcoexist withReno <xrefReno <xref target="RFC5681" format="default"/>(e.g. Reno(e.g., Reno itself,Cubic <xrefCUBIC <xref target="RFC8312" format="default"/>,Compound <xrefCompound <xref target="I-D.sridharan-tcpm-ctcp" format="default"/>,TFRC <xrefand TFRC <xref target="RFC5348" format="default"/>). The term 'Classic queue' means a queue providing the Classic service.</dd><dt>Low-Latency, Low-Loss<dt>Low Latency, Low Loss, and Scalable throughput (L4S) service:</dt> <dd> <t>The 'L4S' service is intended for traffic fromscalableScalable congestion control algorithms, such as the Prague congestion control <xref target="I-D.briscoe-iccrg-prague-congestion-control" format="default"/>, which was derived fromDCTCP <xrefDCTCP <xref target="RFC8257" format="default"/>. The L4S service is for more general traffic than just Prague -- it allows the set of congestion controls with similar scaling properties to Prague to evolve, such as the examples listed above (Relentless, SCReAM, etc.). The term 'L4S queue' means a queue providing the L4S service.</t> <t>The terms Classic or L4S can also qualify other nouns, such as 'queue', 'codepoint', 'identifier', 'classification', 'packet', and 'flow'. Forexample:example, an L4S packet means a packet with an L4S identifier sent from an L4S congestion control.</t> <t>Both Classic and L4S services can cope with a proportion of unresponsive or less-responsive traffic aswell, butwell but, in the L4Scasecase, its rate has to be smooth enough or low enoughnotto not build a queue(e.g. DNS, VoIP,(e.g., DNS, Voice over IP (VoIP), game sync datagrams, etc.).</t> </dd> <dt>Reno-friendly:</dt> <dd>The subset of Classic traffic that is friendly to the standard Reno congestion control defined for TCP in <xref target="RFC5681" format="default"/>. The TFRCspec <xrefspec <xref target="RFC5348" format="default"/> indirectly implies that 'friendly' is defined as "generally within a factor of two of the sending rate of a TCP flow under the same conditions". Reno-friendly is used here in place of 'TCP-friendly', given the latter has become imprecise, because the TCP protocol is now used with so many different congestion control behaviours, and Renocan beis used in non-TCPtransportstransports, such asQUIC <xrefQUIC <xref target="RFC9000" format="default"/>.</dd> <dt>Classic ECN:</dt><dd>The<dd><t>The original Explicit Congestion Notification (ECN)protocol <xrefprotocol <xref target="RFC3168"format="default"/>, whichformat="default"/> that requires ECN signals to be treatedthe sameas equivalent to drops, both when generated in the network and when responded to by thesender. Forsender.</t> <t>For L4S, the names used for the four codepoints of the 2-bit IP-ECN field are unchanged from those defined in the ECN spec <xref target="RFC3168"format="default"/>: Not ECT,format="default"/>, i.e., Not-ECT, ECT(0),ECT(1)ECT(1), and CE, where ECT stands for ECN-Capable Transport and CE stands for Congestion Experienced. A packet marked with the CE codepoint is termed 'ECN-marked' or sometimes just 'marked' where the context makes ECNobvious.</dd>obvious.</t></dd> <dt>Site:</dt> <dd>A home, mobile device, smallenterpriseenterprise, orcampus,campus where the network bottleneck is typically the access link to the site. Not all network arrangements fit thismodelmodel, but it is a useful, widely applicable generalization.</dd> </dl> </section> <section numbered="true" toc="default"> <name>Scope</name> <t>The new L4S identifier defined in this specification is applicable for IPv4 and IPv6 packets (asforis ClassicECN <xrefECN <xref target="RFC3168" format="default"/>). It is applicable for the unicast,multicastmulticast, and anycast forwarding modes.</t> <t>The L4S identifier is an orthogonal packet classification to the Differentiated Services Code Point(DSCP) <xref(DSCP) <xref target="RFC2474" format="default"/>. <xref target="l4sid_other_IDs" format="default"/> explains what this means in practice.</t> <t>This document isintended for experimental status,Experimental, so it does not update anystandards trackStandards Track RFCs. Therefore, it depends on <xref target="RFC8311" format="default"/>, which is astandards trackStandards Track specification that:</t> <ul spacing="normal"> <li>updates the ECNproposed standard <xrefProposed Standard <xref target="RFC3168" format="default"/> to allowexperimental trackExperimental RFCs to relax the requirement that an ECN mark must be equivalent to a drop (when the network applies markings and/or when the sender responds to them). For instance, in theABE experiment <xrefAlternative Backoff with ECN (ABE) experiment <xref target="RFC8511"format="default"/>format="default"/>, this relaxation permits a sender to respond less to ECN marks than to drops;</li> <li>changes the status of theexperimentalExperimental ECNnonce <xrefnonce spec <xref target="RFC3540" format="default"/> tohistoric;</li>Historic; and</li> <li> <t>makes consequent updates to the following additionalproposed standardProposed Standard RFCs to reflect the above two bullets:</t> <ul spacing="normal"> <li>ECN forRTP <xrefRTP <xref target="RFC6679"format="default"/>;</li>format="default"/> and </li> <li>the congestion control specifications of various DCCPcongestion control identifierCongestion Control Identifier (CCID)profiles <xrefprofiles <xref target="RFC4341"format="default"/>,format="default"/> <xref target="RFC4342"format="default"/>,format="default"/> <xref target="RFC5622" format="default"/>.</li> </ul> </li> </ul> <t>This document is about identifiers that are used for interoperation between hosts and networks. So the audience is broad, covering developers of host transports and network AQMs, as well as covering how operators might wish to combine various identifiers, which would require flexibility from equipment developers.</t> </section> </section> <section anchor="l4sid_identification" numbered="true" toc="default"> <name>L4S Packet Identification: Document Roadmap</name> <t>The L4S ECN marking treatment is an experimentaltrackalternativepacket marking treatmentto the Classic ECN treatment in <xref target="RFC3168" format="default"/>, which has been updated by <xref target="RFC8311" format="default"/> to allow experiments such as the one defined in the present specification. <xref target="RFC4774" format="default"/> discusses some of the issues and evaluation criteria when defining alternative ECN semantics, which are further discussed in <xref target="l4sid_congestion_response_rfcs" format="default"/>.</t> <t>The L4S architecture <xreftarget="I-D.ietf-tsvwg-l4s-arch"target="RFC9330" format="default"/> describes the three main components of L4S: the sending host behaviour, the marking behaviour in thenetworknetwork, and the L4S ECN protocol that identifies L4S packets as they flow between the two.</t><t>The next section<t><xref target="l4sid_reqs" format="default"/> ofthe presentthis document(<xref target="l4sid_reqs" format="default"/>)records the requirements that informed the choice of L4S identifier.ThenThen, subsequent sections specify the L4S ECN protocol, which i) identifies packets that have been sent from hosts that are expected to comply with a broad type of sendingbehaviour;behaviour and ii) identifies the marking treatment that network nodes are expected to apply to L4S packets.</t> <t>For a packet to receive L4S treatment as it is forwarded, the sender sets the ECN field in the IP header to the ECT(1) codepoint. See <xref target="l4sid_transport_req" format="default"/> for fulltransport layertransport-layer behaviour requirements, including feedback and congestion response.</t> <t>A network node that implements the L4S service always classifies arriving ECT(1) packets for L4S treatment and by default classifies CE packets for L4S treatment unless the heuristics described in <xref target="l4sid_identification_transport_aware" format="default"/> are employed. See <xref target="l4sid_network_req" format="default"/> for full network element behaviour requirements, including classification,ECN-markingECN marking, and interaction of the L4S identifier with other identifiers and per-hop behaviours.</t> <t>L4S ECN works with ECN tunnelling and encapsulation behaviour as is, except there is one known case where careful attention to configuration is required, which is detailed in <xref target="l4sid_encaps" format="default"/>.</t><t>L4S<t>This specification of L4S ECNiscurrentlyon the experimental track.has Experimental status. So <xref target="l4sid_expts" format="default"/> collectstogetherthe general questions and issues that remain open for investigation during L4S experimentation. Open issues or questions specific to particular components are called out in the specifications of each component part, such as the DualQ <xreftarget="I-D.ietf-tsvwg-aqm-dualq-coupled"target="RFC9332" format="default"/>.</t> <t>The IANA assignment of the L4S identifier is specified in <xref target="l4sid_IANA" format="default"/>. And <xref target="l4sid_Security_Considerations" format="default"/> covers security considerations specific to the L4S identifier. System security aspects, such as policing and privacy, are covered in the L4S architecture <xreftarget="I-D.ietf-tsvwg-l4s-arch"target="RFC9330" format="default"/>.</t> </section> <section anchor="l4sid_reqs" numbered="true" toc="default"> <name>Choice of L4S Packet Identifier: Requirements</name> <t>This subsection briefly records the process that led to the chosen L4S identifier.</t> <t>The identifier for packets using theLow Latency, Low Loss, Scalable throughput (L4S)L4S service needs to meet the following requirements:</t> <ul spacing="normal"> <li>itSHOULD<bcp14>SHOULD</bcp14> surviveend-to-endend to end between source and destinationend-points:endpoints: across the boundary between host and network, between interconnected networks, and through middleboxes;</li> <li>itSHOULD<bcp14>SHOULD</bcp14> be visible at the IP layer;</li> <li>itSHOULD<bcp14>SHOULD</bcp14> be common to IPv4 and IPv6 and be transport-agnostic;</li> <li>itSHOULD<bcp14>SHOULD</bcp14> be incrementally deployable;</li> <li>itSHOULD<bcp14>SHOULD</bcp14> enable an AQM to classify packets encapsulated by outer IP or lower-layer headers;</li> <li>itSHOULD<bcp14>SHOULD</bcp14> consume minimal extracodepoints;</li>codepoints; and</li> <li>itSHOULD<bcp14>SHOULD</bcp14> be consistent on all the packets of atransport layertransport-layer flow, so that some packets of a flow are not served by a different queue to others.</li> </ul> <t>Whether the identifier would be recoverable if the experiment failed is a factor that could be taken into account. However, this has not been made a requirement, because that would favour schemes that would be easier tofail,fail rather thanthosemore likely to succeed.</t> <t>It is recognized that any choice of identifier is unlikely to satisfy all these requirements, particularly given the limited space left in the IP header. Therefore, a compromise will always be necessary, which is why all the above requirements are expressed with the word'SHOULD'"<bcp14>SHOULD</bcp14>" not'MUST'.</t>"<bcp14>MUST</bcp14>".</t> <t>After extensive assessment of alternative schemes, "ECT(1) and CE codepoints" was chosen as the best compromise. Therefore, this scheme is defined in detail in the following sections, while <xref target="l4sid_ECT1_CE" format="default"/> records its pros and cons against the above requirements.</t> </section> <section anchor="l4sid_transport_req" numbered="true" toc="default"><name>Transport Layer<name>Transport-Layer Behaviour (the 'Prague L4S Requirements')</name> <t>This section defines L4S behaviour at the transport layer, also known as thePrague'Prague L4SRequirementsRequirements' (see <xref target="l4sps_tcp-prague-reqs" format="default"/> for the origin of the name).</t> <section anchor="l4sid_codepoint" numbered="true" toc="default"> <name>Codepoint Setting</name> <t>A sender that wishes a packet to receive L4S treatment as it isforwarded, MUSTforwarded <bcp14>MUST</bcp14> set the ECN field in the IP header (v4 or v6) to the ECT(1) codepoint.</t> </section> <section anchor="l4sid_feedback" numbered="true" toc="default"> <name>Prerequisite Transport Feedback</name> <t>For a transport protocol to providescalableScalable congestion control (<xref target="l4sid_congestion_response"format="default"/>)format="default"/>), itMUST<bcp14>MUST</bcp14> provide feedback of the extent of CE marking on the forward path. When ECN was added toTCP <xrefTCP <xref target="RFC3168" format="default"/>, the feedback method reported no more than one CE mark per round trip. Some transport protocols derived from TCP mimic this behaviour while others report the accurate extent of ECN marking. This means that some transport protocols will need to be updated as a prerequisite forscalableScalable congestion control. The position for a few well-known transport protocols is given below.</t> <dl newline="false" spacing="normal"> <dt>TCP:</dt> <dd>Support for the accurate ECN feedbackrequirements <xrefrequirements <xref target="RFC7560" format="default"/> (such as that provided byAccECN <xrefAccECN <xref target="I-D.ietf-tcpm-accurate-ecn" format="default"/>) by both ends is a prerequisite forscalableScalable congestion control in TCP. Therefore, the presence of ECT(1) in the IP headers even in one direction of a TCP connection will imply that both ends support accurate ECN feedback. However, the converse does not apply. So even if both ends support AccECN, either of the two ends can choose not to use ascalableScalable congestion control, whatever the other end'schoice.</dd>choice is.</dd> <dt>SCTP:</dt> <dd>A suitable ECN feedback mechanism for SCTP could add a chunk to report the number of received CE marks (as described in a long-expireddraft <xrefdocument <xref target="I-D.stewart-tsvwg-sctpecn" format="default"/> or as sketched out in AppendixA<xref target="RFC4960" section="A" sectionFormat="bare"/> of thenow obsoletenow-obsolete secondstandards trackStandards Track specification ofSCTP <xrefSCTP <xref target="RFC4960" format="default"/>).</dd> <dt>RTP over UDP:</dt> <dd>A prerequisite forscalableScalable congestion control is for both (all) ends of one media-level hop to signal ECNsupport <xrefsupport <xref target="RFC6679" format="default"/> and use the new generic RTCP feedback format of <xref target="RFC8888" format="default"/>. The presence of ECT(1) implies that both (all) ends of that media-level hop support ECN. However, the converse does not apply. So each end of a media-level hop can independently choose not to use ascalableScalable congestion control, even if both ends support ECN.</dd> <dt>QUIC:</dt> <dd>Support for sufficiently fine-grained ECN feedback is provided bythe v1IETF QUICtransport <xreftransport v1 <xref target="RFC9000" format="default"/>.</dd> <dt>DCCP:</dt> <dd>TheACKAcknowledgement (ACK) vector inDCCP <xrefDCCP <xref target="RFC4340" format="default"/> is already sufficient to report the extent of CE marking as needed by ascalableScalable congestion control.</dd> </dl> </section> <section anchor="l4sid_congestion_response" numbered="true" toc="default"> <name>Prerequisite Congestion Response</name> <t>As a condition for a host to send packets with the L4S identifier (ECT(1)), itSHOULD<bcp14>SHOULD</bcp14> implement a congestion control behaviour that ensures that, in steady state, the average duration between induced ECN marks does not increase as flow rate scales up, all other factors being equal. This is termed ascalableScalable congestion control. This invariant duration ensures that, as flow rate scales, the average period with no feedback information about capacity does not become excessive. It also ensures that queue variations remain small, without having to sacrifice utilization.</t> <t>With a congestion control that sawtooths to probe capacity, this duration is called the recovery time, because each time the sawtooth yields, on average it takes this time to recover to its previous high point. AscalableScalable congestion control does not have to sawtooth, but it has to coexist withscalableScalable congestion controls that do.</t> <t>For instance, forDCTCP <xrefDCTCP <xref target="RFC8257" format="default"/>, TCP Prague <xref target="I-D.briscoe-iccrg-prague-congestion-control"format="default"/>, <xref target="PragueLinux"format="default"/> <xref target="PragueLinux" format="default"/>, and the L4S variant ofSCReAM <xrefSCReAM <xref target="SCReAM-L4S"format="default"/>,format="default"/> <xref target="RFC8298" format="default"/>, the average recovery time is always half a round trip (or half a reference round trip), whatever the flow rate.</t> <t>As with all transport behaviours, a detailed specification (probably anexperimentalExperimental RFC) is expected for each congestion control, following the guidelines for specifying new congestion control algorithmsin <xrefin <xref target="RFC5033" format="default"/>. In addition, it is expectedto documentthat these L4S-specificmatters,matters will be documented, specifically the timescale over which the proportionality isaveraged,averaged and the control of burstiness. The recovery time requirement above is worded as a'SHOULD'"<bcp14>SHOULD</bcp14>" rather than a'MUST'"<bcp14>MUST</bcp14>" to allow reasonable flexibility for such implementations.</t> <t>The condition 'all other factors beingequal',equal' allows the recovery time to be different for differentround tripround-trip times, as long as it does not increase with flow rate for any particular RTT.</t> <t>Saying that the recovery time remains roughly invariant is equivalent to saying that the number of ECN CE marks per round trip remains invariant as flow rate scales, all other factors being equal. For instance, an average recovery time of half of 1 RTT is equivalent to 2 ECN marks per round trip. For those familiar with steady-state congestion response functions, it is also equivalent to say that the congestion window is inversely proportional to the proportion of bytes in packets marked with the CE codepoint (seesectionSection 2 of <xref target="PI2" format="default"/>).</t><!--For example, the timescale over which this rough proportionality applies will depend on the type of application, ranging from per packet adjustment through smooth adjustment of encoding over perhaps tens of seconds. Low bit-rate flows might even respond at the timescale of self-admission control solely at the start of each flow. As with any congestion control, the sender SHOULD also avoid excessive bursts, but this is particularly important with the L4S service.--><t>In order to coexist safely with other Internet traffic, ascalableScalable congestion control is not allowed to tag its packets with the ECT(1) codepoint unless it complies with the following numbered requirements and recommendations:</t> <ol spacing="normal"type="1"><li>A scalabletype="1"><li anchor="l4sid_Prague_req-replaceable">A Scalable congestion controlMUST<bcp14>MUST</bcp14> be capable of being replaced by a Classic congestion control (by application and/or by administrative control). If a Classic congestion control is activated, it will not tag its packets with the ECT(1) codepoint (see <xref target="l4sid_sec_replaceable" format="default"/> for rationale).</li><li>As<li anchor="l4sid_Prague_req-loss-response">As well as responding to ECN markings, ascalableScalable congestion controlMUST<bcp14>MUST</bcp14> react to packet loss in a way that will coexist safely with Classic congestion controls such as standardReno <xrefReno <xref target="RFC5681" format="default"/>, as required by <xref target="RFC5033" format="default"/> (see <xref target="l4sid_sec_fallback_on_loss" format="default"/> for rationale).</li><li><li anchor="l4sid_Prague_req-classic-ecn-response"> <t>In uncontrolled environments, monitoringMUST<bcp14>MUST</bcp14> be implemented to support detection of problems with an ECN-capable AQM at the path bottleneck that appears not to support L4S and that might be in a shared queue. Such monitoringSHOULD<bcp14>SHOULD</bcp14> be applied to live traffic that is using Scalable congestion control. Alternatively, monitoring need not be applied to live traffic, if monitoring with test traffic has been arranged to cover the paths that live traffic takes through uncontrolled environments. </t> <t>A function to detect the above problems with an ECN-capable AQMMUST<bcp14>MUST</bcp14> also be implemented and used. The detection functionSHOULD<bcp14>SHOULD</bcp14> be capable of making the congestion control adapt its ECN-marking response inreal-timereal time to coexist safely with Classic congestion controls such as standardReno <xrefReno <xref target="RFC5681" format="default"/>, as required by <xref target="RFC5033" format="default"/>. This could be complemented by more detailed offline detection of potential problems. If only offline detection is used and potential problems with such an AQM are detected on certain paths, thescalableScalable congestion controlMUST<bcp14>MUST</bcp14> be replaced by a Classic congestion control, at least for the problem paths. </t> <t>See <xref target="l4sid_congestion_response_rfcs" format="default"/>, <xref target="l4sid_sec_fallback_on_classic_CE"format="default"/>format="default"/>, and the L4S operationalguidance <xrefguidance <xref target="I-D.ietf-tsvwg-l4sops" format="default"/> for rationale and explanation.</t><t>Note<t indent="3">Note that ascalableScalable congestion control is not expected to change to setting ECT(0) while it transiently adapts to coexist with Classic congestion controls, whereas a replacement congestion control that solely behaves in the Classic way will set ECT(0).</t> </li><li>In<li anchor="l4sid_Prague_req-rtt-independence">In the range between the minimum likely RTT and typical RTTs expected in the intended deployment scenario, ascalableScalable congestion controlMUST<bcp14>MUST</bcp14> converge towards a rate that is as independent of RTT as is possible without compromising stability or utilization (see <xref target="l4sid_sec_RTT_dependence" format="default"/> for rationale).</li><li>A scalable<li anchor="l4sid_Prague_req-fractional-window">A Scalable congestion controlSHOULD<bcp14>SHOULD</bcp14> remain responsive to congestion when typical RTTs over the public Internet are significantly smaller because they are no longer inflated by queuing delay. It would be preferable for the minimum window of ascalableScalable congestion control to be lower than 1 segment rather than use the timeout approach described for TCP inS.6.1.2 of the ECN spec <xref<xref target="RFC3168"format="default"/>sectionFormat="of" section="6.1.2">the ECN spec</xref> (or an equivalent for other transports). However, a lower minimum is not set as a formal requirement for L4S experiments (see <xref target="l4sid_sec_min_cwnd" format="default"/> for rationale).</li><li>A scalable<li anchor="l4sid_Prague_req-reordering">A Scalable congestion control's loss detectionSHOULD<bcp14>SHOULD</bcp14> be resilient to reordering over an adaptive time interval that scales with throughput and adapts to reordering (as inRACK <xrefRecent ACK (RACK) <xref target="RFC8985" format="default"/>), as opposed to counting only in fixed units of packets (as in the 3DupACKDuplicate ACK (DupACK) rule ofNew Reno <xrefNewReno <xref target="RFC5681" format="default"/>and<xref target="RFC6675" format="default"/>, which is not scalable). As data rates increase (e.g., due to new and/or improved technology), congestion controls that detect loss by counting in units of packets become more likely to incorrectly treat reordering events as congestion-caused loss events (see <xref target="l4sid_sec_reordering_tolerance" format="default"/> for further rationale). This requirement does not apply to congestion controls that are solely used in controlled environments where the network introduces hardly any reordering.</li><li>A scalable<li anchor="l4sid_Prague_req-burstiness">A Scalable congestion control is expected to limit the queue caused by bursts of packets. It would not seem necessary to set the limit any lower than 10% of the minimum RTT expected in a typical deployment(e.g. additional(e.g., additional queuing of roughly 250 us for the public Internet). This would be converted to a number of packets by multiplying by the current average packet rate. Then, the queue caused by each burst at the bottleneck link would not exceed250us250 us (under the worst-case assumption that the flow is filling the capacity). No normative requirement to limit bursts is givenhere and,here, and until there is more industry experience from the L4S experiment, it is not even known whether one is needed--- it seems to be in an L4S sender's self-interest to limit bursts.</li> </ol> <t>Each sender in a session can use ascalableScalable congestion control independently of the congestion control used by the receiver(s) when they send data. Therefore, there might be ECT(1) packets in one direction and ECT(0) or Not-ECT in the other.</t><t>Later (<xref target="l4sid_inclusion_dualq" format="default"/>)<t>Later, this document discusses the conditions for mixing other "'Safe' Unresponsive Traffic"(e.g. DNS, LDAP,(e.g., DNS, Lightweight Directory Access Protocol (LDAP), NTP, voice, and game sync packets) with L4Straffic.traffic; see <xref target="l4sid_inclusion_dualq" format="default"/>. To be clear, although such traffic can share the same queue as L4S traffic, it is not appropriate for the sender to tag it as ECT(1), except in the (unlikely) case that it satisfies the above conditions.</t> <section anchor="l4sid_congestion_response_rfcs" numbered="true" toc="default"> <name>Guidance on Congestion Response in the RFC Series</name><t>RFC 3168<t><xref target="RFC3168"/> requires the congestion responses to a CE-marked packet and a dropped packet to be the same.RFC 8311<xref target="RFC8311"/> is astandards-trackStandards Track update toRFC 3168<xref target="RFC3168"/> that is intended to enable experimentation with ECN, including the L4S experiment.RFC 8311<xref target="RFC8311"/> allows an experimental congestion control's response to a CE-marked packet to differ from the response to a dropped packet, provided that the differences are documented in anexperimentalExperimental RFC, such as the present document.</t><t>BCP 124 <xref<t>BCP 124 <xref target="RFC4774" format="default"/> gives guidance to protocol designers, when specifying alternative semantics for theECNIP-ECN field.RFC 8311<xref target="RFC8311"/> explained that it did not need to update the best current practice inBCP 124BCP 124 in order to relax the 'equivalence with drop' requirement because, althoughBCP 124BCP 124 quotes the same requirement fromRFC 3168,<xref target="RFC3168"/>, the BCP does not impose requirements based on it.BCP 124BCP 124 <xref target="RFC4774" format="default"/> describes three options for incremental deployment, with Option 3 (inSection 4.3 of BCP 124)<xref target="RFC4774" section="4.3" sectionFormat="of">BCP 124</xref>) best matching the L4S case. Option 3's requirement for end-nodes is that they respond to CE marks "in a way that is friendly to flows using IETF-conformant congestion control." This echoes other general congestion control requirements in the RFCseries,Series, forexampleexample, <xref target="RFC5033"format="default"/>, which saysformat="default"/> states that "...congestion controllers that have a significantly negative impact on traffic using standard congestion control may besuspect", orsuspect" and <xref target="RFC8085"format="default"/> concerningformat="default"/>, which concerns UDP congestioncontrol sayscontrol, states that "Bulk-transfer applications that choose not to implement TFRC or TCP-like windowingSHOULD<bcp14>SHOULD</bcp14> implement a congestion control scheme that results in bandwidth (capacity) use that competes fairly with TCP within an order of magnitude."</t> <t>ThethirdnormativebulletItem <xref target="l4sid_Prague_req-classic-ecn-response" format="counter"/> in <xref target="l4sid_congestion_response" format="default"/> above (which concerns L4S response to congestion from a Classic ECN AQM) aims to ensure that these 'coexistence' requirements are satisfied, but it makes some compromises. This subsection highlights and justifies thosecompromisescompromises, and <xref target="l4sid_sec_fallback_on_classic_CE" format="default"/> and the L4S operationalguidance <xrefguidance <xref target="I-D.ietf-tsvwg-l4sops" format="default"/> give detailed analysis,examplesexamples, and references (the normative text in that bullet takes precedence if any informative elaboration leads to ambiguity). The approach is based on an assessment of the risk of harm, which is a combination of the prevalence of the conditions necessary for harm to occur, and the potential severity of the harm if they do. </t> <dl newline="false" spacing="normal"> <dt>Prevalence:</dt> <dd> <t>There are three cases:</t> <ul spacing="normal"> <li>Drop Tail: Coexistence between L4S and Classic flows is not in doubt where the bottleneck does not support any form of ECN, which has remained by far the most prevalent case since the ECNRFCspec <xref target="RFC3168" format="default"/> was published in 2001.</li> <li>L4S: Coexistence is not in doubt if the bottleneck supports L4S.</li> <li> <t>ClassicECN <xref target="RFC3168" format="default"/>:ECN: The compromises centre around cases where the bottleneck supports Classic ECN <xref target="RFC3168" format="default"/> but not L4S. But it depends on which sub-case:</t> <ul spacing="normal"> <li>Shared Queue with Classic ECN: At the time of writing, the members of the Transport WorkinggroupGroup are not aware of any current deployments of single-queue Classic ECN bottlenecks in the Internet. Nonetheless, at the scale of the Internet, rarity need not imply smallnumbers,numbers nor that there will be rarity in the future.</li> <li><t>Per-Flow-queues<t>Per-Flow Queues with Classic ECN: Most AQMs withper-flow-queuing (FQ)per-flow queuing deployed from 2012 onwards had Classic ECN enabled by default, specificallyFQ-CoDel <xrefFQ-CoDel <xref target="RFC8290" format="default"/> andCOBALT <xrefCOBALT <xref target="COBALT" format="default"/>. But the compromises only apply to the second of two further sub-cases:</t> <ul spacing="normal"> <li>Withper-flow-queuing, co-existenceper-flow queuing, coexistence between Classic and L4S flows is not normally a problem, because different flows are not meant to be in the same queue(BCP 124 <xref(BCP 124 <xref target="RFC4774" format="default"/> did not foresee the introduction ofper-flow-queuing,per-flow queuing, which appeared as a potential isolation technique some eight years after the BCP was published).</li> <li>However, the isolation between L4S and Classic flows is not perfect in cases where the hashes of flowIDsidentifiers (IDs) collide or where multiple flows within alayer-3Layer 3 VPN are encapsulated within one flow ID.</li> </ul> </li> </ul> </li> </ul> <t>To summarize, the coexistence problem is confined to cases of imperfect flow isolation in anFQ,FQ or in potential cases where a Classic ECN AQM has been deployed in a shared queue (see the L4S operationalguidance <xrefguidance <xref target="I-D.ietf-tsvwg-l4sops" format="default"/> for further details including recent surveys attempting to quantify prevalence). Further, if one of these cases does occur, the coexistence problem does not arise unless sources of Classic and L4S flows are simultaneously sharing the same bottleneck queue(e.g. different(e.g., different applications in the samehousehold)household), and flows of each type have to be large enough to coincide for long enough for any throughput imbalance to have developed. Therefore, how often the coexistence problem arises in practice is listed in <xref target="l4sid_expts" format="default"/> as an open question that L4S experiments will need to answer.</t> </dd> <dt>Severity:</dt> <dd>Where long-running L4S and Classic flows coincide in a shared queue, testing of one L4S congestion control (TCP Prague) has found that the imbalance in average throughput between an L4S and a Classic flow can reach 25:1 in favour of L4S in the worstcase <xrefcase <xref target="ecn-fallback" format="default"/>. However, when capacity is most scarce, the Classic flow gets a higher proportion of the link, forinstanceinstance, over a 4 Mb/s link the throughput ratio is below ~10:1 over paths with a base RTT below 100 ms, and it falls below ~5:1 for base RTTs below20ms.</dd>20 ms.</dd> </dl> <t>These throughput ratios can clearly fall well outside current RFC guidance on coexistence. However, the tendency towards leaving a greater share for Classic flows at lower link rate and the very limited prevalence of the conditions necessary for harm to occur led to the possibility of allowing the RFC requirements to be compromised, albeit briefly:</t> <ul spacing="normal"> <li>The recommended approach is still to detect and adapt to a Classic ECN AQM inreal-time,real time, which is fully consistent with all the RFCs on coexistence. In other words, the"SHOULD"s"<bcp14>SHOULD</bcp14>"s inthe third bulletItem <xref target="l4sid_Prague_req-classic-ecn-response" format="counter"/> of <xref target="l4sid_congestion_response" format="default"/> above expect the sender to implement something similar to theproof of conceptproof-of-concept code that detects the presence of a Classic ECN AQM and falls back to a Classic congestion response within a few roundtrips <xreftrips <xref target="ecn-fallback" format="default"/>. However, although this code reliably detects a Classic ECN AQM, the current code can also wrongly categorize an L4S AQM as Classic, most often in cases when link rate is low or RTT is high. Although this is the safe way round, and although implementers are expected to be able to improve on this proof of concept, concerns have been raised that implementers might lose faith in such detection and disable it.</li><li>Therefore the third bullet<li>Item <xref target="l4sid_Prague_req-classic-ecn-response" format="counter"/> in <xref target="l4sid_congestion_response" format="default"/> above therefore allows a compromise where coexistence could briefly diverge from the requirements in the RFCSeries briefly,Series, but mandatory monitoring isrequired,required in order to detect such cases and trigger remedial action. This approach tolerates a brief divergence from the RFCs given the likely low prevalence and given harm here means a flow progresses more slowly than it would otherwise, but it does progress. The L4S operationalguidance <xrefguidance <xref target="I-D.ietf-tsvwg-l4sops" format="default"/> outlines a range of example remedial actions that include alterationseitherto either the sender ortothe network. However, the final normative requirement inthe third bulletItem <xref target="l4sid_Prague_req-classic-ecn-response" format="counter"/> of <xref target="l4sid_congestion_response" format="default"/> above places ultimate responsibility for remedial action on the sender. If coexistence problems with a Classic ECN AQM are detected (implying they have not been resolved by the network), itsaysstates that the sender"MUST""<bcp14>MUST</bcp14>" revert to a Classic congestioncontrol."</li>control.</li> </ul> <t><xref target="I-D.ietf-tsvwg-l4sops" format="default"/> also gives example ways in which L4S congestion controls can be rolled out initially inlower risklower-risk scenarios.</t> </section> </section> <section numbered="true" toc="default"> <name>Filtering or Smoothing of ECN Feedback</name> <t><xref target="l4sid_Semantics" format="default"/> below specifies that an L4S AQM is expected to signal L4S ECN immediately, to avoid introducing delay due to filtering or smoothing. This contrasts with a Classic AQM, which filters out variations in the queue before signalling ECN marking or drop. In the L4Sarchitecture <xref target="I-D.ietf-tsvwg-l4s-arch"architecture <xref target="RFC9330" format="default"/>, responsibility for smoothing out these variations shifts to the sender's congestion control.</t> <t>This shift of responsibility has the advantage that each sender can smooth variations over a timescale proportionate to its own RTT. Whereas, in the Classic approach, the network doesn't know the RTTs of any of the flows, so it has to smooth out variations for a worst-case RTT to ensure stability. For all the typical flows with shorterRTTRTTs than the worst-case, this makes congestion control unnecessarily sluggish.</t> <t>This also gives an L4S sender the choice not to smooth, depending on its context (start-up, congestion avoidance, etc.). Therefore, this document places no requirement on an L4S congestion control to smooth out variations in any particular way. Implementers are encouraged to openly publish the approach they take tosmoothing, and thesmoothing as well as results and experience they gain during the L4S experiment.</t> </section> </section> <section anchor="l4sid_network_req" numbered="true" toc="default"> <name>Network Node Behaviour</name><t/><section numbered="true" toc="default"> <name>Classification and Re-Marking Behaviour</name> <t>A network node that implements the L4S service:</t> <ul spacing="normal"><li>MUST<li><bcp14>MUST</bcp14> classify arriving ECT(1) packets for L4S treatment, unless overridden by another classifier(e.g., see(e.g., see <xref target="l4sid_exclusion_dualq"format="default"/>);</li>format="default"/>).</li> <li><t>MUST<t><bcp14>MUST</bcp14> classify arriving CE packets for L4S treatment as well, unless overridden by another classifier or unless the exception referred to nextapplies;</t>applies.</t> <t>CE packets might have originated as ECT(1) or ECT(0), but the above rule to classify them as if they originated as ECT(1) is the safe choice (see <xref target="l4sid_ECT1_CE" format="default"/> for rationale). The exception is where some flow-aware in-network mechanism happens to be available for distinguishing CE packets that originated as ECT(0), as described in <xref target="l4sid_identification_transport_aware" format="default"/>, but there is no implication that such a mechanism is necessary.</t> </li> </ul> <t>An L4S AQM treatment follows similar codepoint transition rules to those inRFC 3168.<xref target="RFC3168"/>. Specifically, the ECT(1) codepointMUST NOT<bcp14>MUST NOT</bcp14> be changed to anyothercodepoint other than CE, and CEMUST NOT<bcp14>MUST NOT</bcp14> be changed to any other codepoint. An ECT(1) packet is classified asECN-capable and,'ECN-capable', and if congestion increases, an L4S AQM algorithm will increasingly mark theECNIP-ECN field as CE, otherwise forwarding packets unchanged as ECT(1). Necessary conditions for an L4S marking treatment are defined in <xref target="l4sid_Semantics" format="default"/>.</t> <t>Under persistentoverloadoverload, an L4S marking treatmentMUST<bcp14>MUST</bcp14> begin applying drop to L4S traffic until the overload episode has subsided, as recommended for all AQM methods in <xref target="RFC7567"format="default"/> (Section 4.2.1),sectionFormat="of" section="4.2.1"/>, which follows the similar advice inRFC 3168 (Section 7).<xref target="RFC3168" sectionFormat="of" section="7"/>. During overload, itMUST<bcp14>MUST</bcp14> apply the same drop probability to L4S traffic as it would to Classic traffic.</t> <t>Where an L4S AQM is transport-aware, this requirement could be satisfied by using drop in only the most overloaded individual per-flow AQMs. In a DualQ with flow-aware queue protection(e.g. <xref(e.g., <xref target="I-D.briscoe-docsis-q-protection" format="default"/>), this could be achieved by redirecting packets in those flows contributing most to the overload out of the L4S queue so that they are subjected to drop in the Classic queue.</t><!--KDS:Do we want to propose this? In the paper I proposed to limit the probabilities to a max value, and to use delay to control overload (until tail drop). BB: I've allowed what you do and made it sound compatible with RFC3168 --><t>For backward compatibility in uncontrolled environments, a network node that implements the L4S treatmentMUST<bcp14>MUST</bcp14> also implement an AQM treatment for the Classic service as defined in <xref target="l4sid_Terminology" format="default"/>. This Classic AQM treatment need not mark ECT(0) packets, but if it does, see <xref target="l4sid_Semantics" format="default"/> for the strengths of the markings relative to drop. ItMUST<bcp14>MUST</bcp14> classify arriving ECT(0) and Not-ECT packets for treatment by this Classic AQM (for the DualQ CoupledAQM,AQM; see the extensive discussion on classification in Sections2.3<xref target="RFC9332" section="2.3" sectionFormat="bare"/> and2.5.1.1<xref target="RFC9332" section="2.5.1.1" sectionFormat="bare"/> of <xreftarget="I-D.ietf-tsvwg-aqm-dualq-coupled"target="RFC9332" format="default"/>).</t> <t>In case unforeseen problems arise with the L4S experiment, itMUST<bcp14>MUST</bcp14> be possible to configure an L4S implementation to disable the L4S treatment. Once disabled,all packets of all ECN codepoints will receive Classic treatment andECT(1) packetsMUST<bcp14>MUST</bcp14> be treated as if they were Not-ECT.</t> </section> <section anchor="l4sid_Semantics" numbered="true" toc="default"> <name>The Strength of L4S CE Marking Relative to Drop</name> <t>The relative strengths of L4S CE and drop are irrelevant where AQMs are implemented in separate queuesper-application-flow,per application-flow, which are then explicitly scheduled(e.g. with(e.g., with an FQ scheduler as inFQ-CoDel <xrefFQ-CoDel <xref target="RFC8290" format="default"/>). Nonetheless, the relationship between them needs to be defined for the coupling between L4S and Classic congestion signals in a DualQ CoupledAQM <xref target="I-D.ietf-tsvwg-aqm-dualq-coupled"AQM <xref target="RFC9332" format="default"/>, as indicated below.</t> <t>Unless an AQM node schedules application flows explicitly, the likelihood that the AQM drops a Not-ECT Classic packet (p_C)MUST<bcp14>MUST</bcp14> be roughly proportional to the square of the likelihood that it would have marked it if it had been an L4S packet (p_L). Thatis</t> <ul empty="true" spacing="normal"> <li>p_Cis:</t> <t indent="3">p_C ~= (p_L /k)^2</li> </ul>k)<sup>2</sup></t> <t>The constant of proportionality (k) does not have to be standardized for interoperability, but a value of 2 isRECOMMENDED.<bcp14>RECOMMENDED</bcp14>. The term 'likelihood' is used above to allow for marking and dropping to be either probabilistic or deterministic.</t> <t>This formula ensures that Scalable and Classic flows will converge to roughly equal congestion windows, for the worst case of Reno congestion control. This is because the congestion windows of Scalable and Classic congestion controls are inversely proportional to p_L andsqrt(p_C)sqrt(p_C), respectively. So squaring p_C in the above formula counterbalances the square root that characterizes Reno-friendly flows.</t><t>Note<aside><t>Note that, contrary toRFC 3168,<xref target="RFC3168" format="default"/>, an AQM implementing the L4S and Classic treatments does not mark an ECT(1) packet under the same conditions that it would have dropped a Not-ECT packet, as allowed by <xref target="RFC8311" format="default"/>, which updatesRFC 3168.<xref target="RFC3168" format="default"/>. However, if it marks ECT(0) packets, it does so under the same conditions that it would have dropped a Not-ECTpacket <xref target="RFC3168" format="default"/>.<!--KDS: replace "a Not-ECT packet" by "a packet", as any drop should be classic, and also ECT(0),(1) or CE packets can be dropped BB: But that's not the intended meaning. Yes, apacketwith any ECN field can be dropped, but this rule is just about what /would/ have been done /if/ the ECN field had been one specific value (Not-ECT). --> </t><xref target="RFC3168" format="default"/>. </t></aside> <t>Also,Inin the L4Sarchitecture <xref target="I-D.ietf-tsvwg-l4s-arch"architecture <xref target="RFC9330" format="default"/>, the sender, not the network, is responsible for smoothing out variations in the queue.So,So an L4S AQMMUST<bcp14>MUST</bcp14> signal congestion as soon as possible. Then, an L4S sender generally interprets CE marking as an unsmoothed signal.</t> <t>This requirement does not prevent an L4S AQM from mixing in additional congestion signals that are smoothed, such as the signals from a Classic smoothed AQM that are coupled with unsmoothed L4S signals in the coupledDualQ <xref target="I-D.ietf-tsvwg-aqm-dualq-coupled" format="default"/>. ButDualQ <xref target="RFC9332" format="default"/>, but only as long as the onset of congestion can be signalledimmediately,immediately and can be interpreted by the sender as if it has been signalled immediately, which is important for interoperability</t> </section> <section anchor="l4sid_identification_transport_aware" numbered="true" toc="default"> <name>Exception for L4S Packet Identification by Network Nodes with Transport-Layer Awareness</name><!--This section doesn't fit very well here. It would be better as an Appendix, then it would need the following introductory para: Section 2.3 recognizes that CE packets might have originally been L4S or Classic. It argues that this ambiguity is not serious, and therefore, for simplicity, it recommends that a packet classifer in the network "SHOULD" classify all CE packets as L4S. Even though it is probably unnecessary to resolve the ambiguity, the following text provides a way to do so using per-flow processing in the network. Per-flow processing has other detrimental side-effects, so this text is controversial, but worth recording, particularly for those cases where per-flow processing is already being used for other purposes.--><t>To implement L4S packet classification, a network node does not need to identify transport-layer flows. Nonetheless, if an L4S network node classifies packets by their transport-layer flow ID and their ECN field, and if all the ECT packets in a flow have been ECT(0), the nodeMAY<bcp14>MAY</bcp14> classify any CE packets in the same flow as if they were Classic ECT(0) packets. In all other cases, a network nodeMUST<bcp14>MUST</bcp14> classify all CE packets as if they were ECT(1) packets. Examples of such other cases are: i) if no ECT packets have yet been identified in a flow; ii) if it is not desirable for a network node to identify transport-layer flows; or iii) if some ECT packets in a flow have been ECT(1) (this advice will need to be verified as part of L4S experiments).</t> </section> <section anchor="l4sid_other_IDs" numbered="true" toc="default"> <name>Interaction of the L4S Identifier withotherOther Identifiers</name> <t>The examples in this section concern how additional identifiers might complement the L4S identifier to classify packets between class-based queues.FirstlyFirstly, <xref target="l4sid_iother_IDs_dualq" format="default"/> considers two queues, L4S and Classic, as in theCoupledDualQAQM <xref target="I-D.ietf-tsvwg-aqm-dualq-coupled"Coupled AQM <xref target="RFC9332" format="default"/>, either alone (<xref target="l4sid_inclusion_dualq" format="default"/>) or within a larger queuing hierarchy (<xref target="l4sid_exclusion_dualq" format="default"/>).ThenThen, <xref target="l4sid_iother_IDs_fq" format="default"/> considers schemes that might combine per-flow 5-tuples with other identifiers.</t> <section anchor="l4sid_iother_IDs_dualq" numbered="true" toc="default"> <name>DualQ Examples of Other Identifiers Complementing L4S Identifiers</name><t/><section anchor="l4sid_inclusion_dualq" numbered="true" toc="default"> <name>Inclusion of Additional Traffic with L4S</name> <t>In a typical case for the publicInternetInternet, a network element that implements L4S in a shared queue might want to classify some low-rate but unresponsive traffic(e.g. DNS,(e.g., DNS, LDAP, NTP, voice, and game sync packets) into thelow latencylow-latency queue to mix with L4S traffic. In thiscasecase, it would not be appropriate to call the queue an L4S queue, because it is shared by L4S and non-L4S traffic. Instead, it will be called thelow latencylow-latency or L queue. The L queue then offers two different treatments:</t> <ul spacing="normal"><li>The<li>the L4S treatment, which is a combination of the L4S AQM treatment and a priority schedulingtreatment;</li> <li>The low latencytreatment, and</li> <li>the low-latency treatment, which is solely the priority scheduling treatment, withoutECN-markingECN marking by the AQM.</li> </ul> <t>To identify packets for just the scheduling treatment, it would be inappropriate to use the L4S ECT(1) identifier, because such traffic is unresponsive to ECN marking. Examples of relevant non-ECN identifiers are:</t> <ul spacing="normal"> <li>address ranges of specific applications or hosts configured to be, or known to be, safe,e.g. hard-coded IoTe.g., hard-coded Internet of Things (IoT) devices sendinglow intensitylow-intensity traffic;</li> <li>certain low data-volume applications or protocols(e.g. ARP, DNS);</li>(e.g., ARP and DNS); and</li> <li>specific Diffserv codepoints that indicate traffic with limited burstiness such as the EF(Expedited Forwarding <xref<xref target="RFC3246"format="default"/>), Voice-Admit <xrefformat="default"/>, VOICE-ADMIT <xref target="RFC5865"format="default"/>format="default"/>, or proposedNQB (Non-Queue-Building <xrefNon-Queue-Building (NQB) <xref target="I-D.ietf-tsvwg-nqb"format="default"/>)format="default"/> service classes or equivalentlocal-useLocal-use DSCPs (see <xref target="I-D.briscoe-tsvwg-l4s-diffserv" format="default"/>).</li> </ul> <t>To be clear, classifying into the L queue based onapplication layerapplication-layer identification(e.g.(e.g., DNS) is an example of a local optimization, not a recommendation. Applications will not be able to rely on such unsolicited optimization. A more reliable approach would be for the sender to set an appropriateIP layerIP-layer identifier, such as one of the above Diffserv codepoints.</t> <t>In summary, a network element that implements L4S in a shared queueMAY<bcp14>MAY</bcp14> classify additional types of packets into the L queue based on identifiers other than theECNIP-ECN field, but the typesSHOULD<bcp14>SHOULD</bcp14> be 'safe' to mix with L4S traffic, where 'safe' is explained in <xref target="l4sid_safe_unresponsive" format="default"/>.</t> <t>A packet that carries one of these non-ECN identifiers to classify it into the L queue would not be subject to the L4SECN markingECN-marking treatment, unless it also carried an ECT(1) or CE codepoint. The specification of an L4S AQMMUST<bcp14>MUST</bcp14> define the behaviour for packets with unexpected combinations of codepoints,e.g. ae.g., a non-ECN-based classifier for the Lqueue,queue but with ECT(0) in theECNIP-ECN field (for examples with appropriate behaviours, seesection 2.5.1.1Section <xref target="RFC9332" section="2.5.1.1" sectionFormat="bare"/> of the DualQspec <xref target="I-D.ietf-tsvwg-aqm-dualq-coupled"spec <xref target="RFC9332" format="default"/>).</t> <t>For clarity, non-ECN identifiers, such as the examples itemized above, might be used by some network operators who believe they identify non-L4S traffic that would be safe to mix with L4S traffic. They are not alternative ways for a host to indicate that it is sending L4S packets. Only the ECT(1) ECN codepoint indicates to a network element that a host is sending L4S packets (and CE indicates that it could have originated as ECT(1)).SpecificallySpecifically, ECT(1) indicates that the host claims its behaviour satisfies the prerequisite transport requirements in <xref target="l4sid_transport_req" format="default"/>.</t> <t>In order to include non-L4S packets in the L queue, a network nodeMUST NOT alter<bcp14>MUST NOT</bcp14> change Not-ECT or ECT(0) in the IP-ECN fieldtointo an L4S identifier. This ensures that these codepoints survive for any potential use later on the network path. If a non-compliant or malicious network node did swap ECT(0) to ECT(1), the packet could subsequently be ECN-marked by a downstream L4S AQM, but the sender would respond to congestion indications thinking it had sent a Classic packet. This could result in the flow yielding excessively to other L4S flows sharing the downstream bottleneck.</t> <section anchor="l4sid_safe_unresponsive" numbered="true" toc="default"> <name>'Safe' Unresponsive Traffic</name> <t>The above section requires unresponsive traffic to be 'safe' to mix with L4S traffic.IdeallyIdeally, this means that the sender never sends any sequence of packets at a rate that exceeds the available capacity of the bottleneck link. However, typically an unresponsive transport does not even know the bottleneck capacity of the path, let alone its available capacity. Nonetheless, an application can be considered safe enough if it paces packets out (not necessarilycompletely regularly)with absolute regularity) such that its maximum instantaneous rate from packet to packet stays well below a typical broadband access rate.</t> <t>This is a vague but useful definition, because manylow latencylow-latency applications of interest, such as DNS, voice, game sync packets, RPC, ACKs, and keep-alives, could match this description.</t><t>Low rate streams<t>Low-rate streams, such as voice and game sync packets, might not use continuously adapting ECN-based congestion control, but they ought to at least use a 'circuit-breaker' style of congestionresponse <xrefresponse <xref target="RFC8083" format="default"/>. If the volume of traffic from unresponsive applications is high enough to overload the link, this will at least protect the capacity available to responsive applications. However, queuing delay in the L queuewillwould probably then rise tothat controlled bythe typically higher level targeted by a Classic (drop-based) AQM. If a network operator considers that such self-restraint is not enough, it might want to police the L queue (see Section8.2<xref target="RFC9330" section="8.2" sectionFormat="bare"/> of the L4Sarchitecture <xref target="I-D.ietf-tsvwg-l4s-arch"architecture <xref target="RFC9330" format="default"/>).</t> </section> </section> <section anchor="l4sid_exclusion_dualq" numbered="true" toc="default"> <name>Exclusion of TrafficFromfrom L4S Treatment</name> <t>To extend the above example, an operator might want to exclude some traffic from the L4S treatment for a policy reason,e.g. securitye.g., security (traffic from malicious sources) or commercial(e.g. initially(e.g., the operator may wish to initially confine the benefits of L4S to business customers).</t> <t>In this exclusion case, the classifierMUST<bcp14>MUST</bcp14> classify on the relevantlocally-usedlocally used identifiers(e.g. source(e.g., source addresses) before classifying the non-matching traffic on the end-to-end L4S ECN identifier.</t> <t>A network nodeMUST NOT<bcp14>MUST NOT</bcp14> alter the end-to-end L4S ECN identifier from L4S to Classic, because an operator decision to exclude certain traffic from L4S treatment is local-only. The end-to-end L4S identifier then survives for other operators to use, or indeed, they can apply their own policy, independently based on their own choice oflocally-usedlocally used identifiers. This approach also allows any operator to remove itslocally-appliedlocally applied exclusions in future,e.g. ife.g., if it wishes to widen the benefit of the L4S treatment to all its customers. If a non-compliant or malicious network node did swap ECT(1) to ECT(0), the packet could subsequently be ECN-marked by a downstream Classic ECN AQM. L4S senders are required to detect and handle such treatment(<xref(see Item <xref target="l4sid_Prague_req-classic-ecn-response" format="counter"/> in <xref target="l4sid_congestion_response"format="default"/> item 3),format="default"/>), but that does not make this swap OK, because such detection is not known to be perfect or immediate.</t> <t>A network node that supports L4S but excludes certain packets carrying the L4S identifier from L4S treatmentMUST<bcp14>MUST</bcp14> still apply marking or dropping that is compatible with an L4S congestion response. For instance, it could either drop such packets with the same likelihood as Classic packets orit couldECN-mark them with a likelihood appropriate to L4S traffic(e.g. the(e.g., the coupled probability in a DualQcoupledCoupled AQM) but aiming for the Classic delay target. ItMUST NOT<bcp14>MUST NOT</bcp14> ECN-mark such packets with a Classic marking probability, which could confuse the sender.</t> </section> <section numbered="true" toc="default"> <name>Generalized Combination of L4S and Other Identifiers</name> <t>L4S concerns low latency, which it can provide for all traffic without differentiation and without <em>necessarily</em> affecting bandwidth allocation. Diffserv provides for differentiation of both bandwidth and low latency, but its control of latency depends on its control of bandwidth.The twoL4S and Diffserv can be combined if a network operator wants to control bandwidth allocation butitalso wants to provide lowlatency -latency, i.e., for any amount of traffic within one of these allocations of bandwidth (rather than only providing low latency by limiting bandwidth) <xref target="I-D.briscoe-tsvwg-l4s-diffserv" format="default"/>.</t> <t>The DualQ examples so far have been framed in the context of providing the default BestEffortsEffort Per-Hop Behaviour (PHB) using two queues--- aLow Latencylow-latency (L) queue and a Classic (C)Queue.queue. This single DualQ structure is expected to be the most common and useful arrangement. But, more generally, an operator might choose to control bandwidth allocation through a hierarchy of Diffserv PHBs at anode,node and to offer one (or more) of these PHBs using a pair of queues for a low latency and a Classic variant of the PHB.</t> <t>In the first case, if we assume that a network element provides no PHBs except the DualQ, if a packet carries ECT(1) or CE, the network element would classify it for the L4S treatment irrespective of its DSCP. And, if a packet carried(say)(for example) the EF DSCP, the network element could classify it into the L queue irrespective of its ECN codepoint. However, where the DualQ is in a hierarchy of other PHBs, the classifier would classify some traffic into other PHBs based on DSCP before classifying between thelow latencylow-latency and Classic queues (based on ECT(1),CECE, and perhaps also the EF DSCP or other identifiers as in the above example). <xref target="I-D.briscoe-tsvwg-l4s-diffserv" format="default"/> gives a number of examples of such arrangements to address various requirements.</t> <t><xref target="I-D.briscoe-tsvwg-l4s-diffserv" format="default"/> describes how an operator might use L4S to offer low latency as well asusingDiffserv for bandwidth differentiation. It identifies two main types of approach, which can be combined: the operator might split certain Diffserv PHBs between L4S and a corresponding Classic service. Or it might split the L4S and/or the Classic service into multiple Diffserv PHBs. In either of these cases, a packet would have to be classified on its Diffserv and ECN codepoints.</t> <t>In summary, there are numerous ways in which the L4S ECN identifier (ECT(1) and CE) could be combined with other identifiers to achieve particular objectives. The following categorization articulates those that are valid, but it is not necessarily exhaustive. Those tagged as 'Recommended-standard-use' could be set by the sending host or a network. Those tagged as 'Local-use' would only be set by a network:</t> <ol spacing="normal" type="1"><li> <t>Identifiers Complementing the L4S Identifier</t> <ol spacing="normal" type="a"><li> <t>Including More Traffic in the L Queue</t><t>(Could<t>(could use Recommended-standard-use or Local-use identifiers)</t> </li> <li> <t>Excluding Certain Traffic from the L Queue</t> <t>(Local-use only)</t> </li> </ol> </li> <li> <t>Identifiers toplacePlace L4SclassificationClassification in a PHB Hierarchy</t><t>(Could<t>(could use Recommended-standard-use or Local-use identifiers)</t> <ol spacing="normal"type="a"><li>PHBs Beforetype="A"><li>PHBs before L4S ECN Classification</li> <li>PHBsAfterafter L4S ECN Classification</li> </ol> </li> </ol> </section> </section> <section anchor="l4sid_iother_IDs_fq" numbered="true" toc="default"><name>Per-Flow<name>Per-flow Queuing Examples of Other Identifiers Complementing L4S Identifiers</name> <t>At a node with per-flowqueueing (e.g. FQ-CoDel <xrefqueuing (e.g., FQ-CoDel <xref target="RFC8290" format="default"/>), the L4S identifier could complement theLayer-4transport-layer flow ID as a further level of flow granularity(i.e. Not-ECT(i.e., Not-ECT and ECT(0) queued separately from ECT(1) and CE packets). In <xref target="l4sid_ECT1_CE" format="default"/>, the "Risk of reordering Classic CEpackets" in <xref target="l4sid_ECT1_CE" format="default"/>packets within a flow" discusses the resulting ambiguity if packets originallymarkedset to ECT(0) are marked CE by an upstream AQM before they arrive at a node that classifies CE as L4S. It argues that the risk of reordering is vanishinglysmallsmall, and the consequence of such a low level of reordering is minimal.</t> <t>Alternatively, it could be assumed that it is not in a flow's own interest to mix Classic and L4S identifiers.ThenThen, the AQM could use theECNIP-ECN field to switch itself between a Classic and an L4S AQM behaviour within one per-flow queue. For instance, for ECN-capable packets, the AQM might consist of a simple markingthresholdthreshold, and an L4S ECN identifier might simply select a shallower threshold than a Classic ECN identifier would.</t> </section> </section> <section anchor="l4sid_bursts_links" numbered="true" toc="default"> <name>Limiting Packet Bursts from Links</name> <t>As well as senders needing to limit packet bursts (<xref target="l4sid_congestion_response" format="default"/>), links need to limit the degree of burstiness they introduce. In both cases (senders andlinks)links), this is atradeoff,trade-off, because batch-handling of packets is done for good reason,e.g. processinge.g., for processing efficiency or to make efficient use of medium acquisition delay. Some take the attitude that there is no point reducing burst delay at the sender below that introduced by links (or vice versa). However, delay reduction proceeds by cutting down 'the longest pole in the tent', which turns the spotlight on the next longest, and so on.</t> <t>This document does not set any quantified requirements for links to limit burst delay, primarily because link technologies are outside the remit of L4S specifications. Nonetheless, the following two subsections outline opportunities for addressing bursty links in the process of L4S implementation and deployment.</t> <section anchor="l4sid_bursts_links_l4s" numbered="true" toc="default"> <name>Limiting Packet Bursts from Links Fed by an L4S AQM</name> <t>It would not make sense to implement an L4S AQM that feeds into a particular link technology without also reviewing opportunities to reduce any form of burst delay introduced by that link technology. This would at least limit the bursts that the link would otherwise introduce into the onward traffic, which would cause jumpy feedback to the sender as well as potential extra queuing delay downstream. This document does not presume to even give guidance on an appropriate target for such burst delay until there is more industry experience of L4S. However, as suggested in <xref target="l4sid_congestion_response"format="default"/>format="default"/>, it would not seem necessary to limit bursts lower than roughly 10% of the minimum base RTT expected in the typical deployment scenario(e.g. 250(e.g., 250 us burst duration for links within the public Internet).</t> </section> <section anchor="l4sid_bursts_links_l4s_upstream" numbered="true" toc="default"> <name>Limiting Packet Bursts from Links Upstream of an L4S AQM</name> <t>The initial scope of the L4S experiment is to deploy L4S AQMs at bottlenecks and L4S congestion controls at senders. This is expected to highlight interactions with the most bursty upstream links and lead operators to tune down the burstiness of those links in theirnetworknetworks that areconfigurable, orconfigurable or, failing that, to have to compromise on the delay target of some L4S AQMs. It might also require specific redesign work relevant to the most problematic link types. Such knock-on effects of initial L4S deployment would all be a part of the learning from the L4S experiment.</t> <t>The details of such link changes are beyond the scope of the present document. Nonetheless, where L4S technology is being implemented on an outgoing interface of a device, it would make sense to consider opportunities for reducing bursts arriving at other incominginterface(s).interfaces. For instance, where an L4S AQM is implemented to feed into the upstream WAN interface of a home gateway, there would be opportunities to alter the Wi-Fi profiles sent out of any Wi-Fi interfaces from the same device, in order to mitigate incoming bursts of aggregated Wi-Fi frames from other Wi-Fi stations.</t> </section> </section> </section> <section anchor="l4sid_encaps" numbered="true" toc="default"> <name>Behaviour of Tunnels and Encapsulations</name> <section anchor="l4sid_encaps_no_change" numbered="true" toc="default"> <name>No Change to ECN Tunnels and Encapsulations in General</name> <t>The L4S identifier is expected to work through and within any tunnel without modification, as long as the tunnel propagates the ECN field in any of the ways that have been defined since the first variant in the year2001 <xref2001 <xref target="RFC3168" format="default"/>. L4S will also work with (but does not rely on) any of the more recent updates to ECN propagation in <xref target="RFC4301" format="default"/>, <xref target="RFC6040"format="default"/>format="default"/>, or <xref target="I-D.ietf-tsvwg-rfc6040update-shim" format="default"/>. However, it is likely that some tunnels still do not implement ECN propagation at all. In these cases, L4S will work through such tunnels, but within them the outer header of L4S traffic will appear as Classic.</t> <t>AQMs are typically implemented where an IP-layer buffer feeds into a lower layer, so they are agnostic tolink layerlink-layer encapsulations. Where a bottleneck link is not IP-aware, the L4S identifier is still expected to work within anylower layerlower-layer encapsulation without modification, as long it propagates theECNIP-ECN field as defined for the link technology, forexampleexample, forMPLS <xrefMPLS <xref target="RFC5129" format="default"/> orTRILL <xrefTransparent Interconnection of Lots of Links (TRILL) <xref target="I-D.ietf-trill-ecn-support" format="default"/>. In some of these cases,e.g. layer-3e.g., Layer 3 Ethernet switches, the AQM accesses theIP layerIP-layer header within the outer encapsulation, so again the L4S identifier is expected to work without modification. Nonetheless, the programme to define ECN for other lower layers is still inprogress <xrefprogress <xref target="I-D.ietf-tsvwg-ecn-encap-guidelines" format="default"/>.</t> </section> <section anchor="l4sid_VPN_anti-replay" numbered="true" toc="default"> <name>VPN Behaviour to Avoid Limitations of Anti-Replay</name> <t>If a mix of L4S and Classic packets is sent into the same security association (SA) of avirtual private network (VPN),VPN, and if the VPN egress is employing the optional anti-replay feature, it could inappropriately discard Classic packets (or discard the records in Classic packets) by mistaking their greater queuing delay for a replay attack (see "Dropped Packets for Tunnels with Replay Protection Enabled" in <xref target="Heist21" format="default"/> for the potential performance impact). This known problem is common to bothIPsec <xrefIPsec <xref target="RFC4301" format="default"/> andDTLS <xrefDTLS <xref target="RFC9147" format="default"/> VPNs, given they use similar anti-replay window mechanisms. The mechanism used can only check for replay within its window, so if the window is smaller than the degree of reordering, it can only assume there might be a replay attack and discard all the packets behind the trailing edge of the window. The specifications of IPsecAH <xrefAuthentication Header (AH) <xref target="RFC4302" format="default"/> andESP <xrefEncapsulating Security Payload (ESP) <xref target="RFC4303" format="default"/> suggest that an implementer scales the size of the anti-replay window with interface speed, and DTLSv1.3 <xrefv1.3 <xref target="RFC9147" format="default"/>saysstates that "The receiverSHOULD<bcp14>SHOULD</bcp14> pick a window large enough to handle any plausible reordering, which depends on the data rate." However, in practice, the size of a VPN's anti-replay window is not always scaled appropriately.</t> <t>If a VPN carrying traffic participating in the L4S experiment experiences inappropriate replay detection, the foremost remedy would be to ensure that the egress is configured to comply with the above window-sizing requirements.</t> <t>If an implementation of a VPN egress does not support a sufficiently large anti-replay window,e.g. duee.g., due to hardware limitations, one of the temporary alternatives listed in order of preference below might be feasible instead:</t> <ul spacing="normal"> <li>If the VPN can be configured to classify packets into different SAs indexed by DSCP, apply the appropriate locally defined DSCPs to Classic and L4S packets. The DSCPs could be applied by the network (based on theleast significantleast-significant bit of theECNIP-ECN field), or by the sending host. Such DSCPs would only need to survive as far as the VPN ingress.</li> <li> <t>If the above is not possible and it is necessary to use L4S, either of the following might be appropriate as a last resort:</t> <ul spacing="normal"> <li>disable anti-replay protection at the VPN egress, after considering the security implications (it is mandatory to allow the anti-replay facility to be disabled in both IPsec andDTLS);</li>DTLS), or</li> <li>configure the tunnel ingress not to propagate ECN to the outer, which would lose the benefits of L4S and Classic ECN over the VPN.</li> </ul> </li><!--ToDo: Perhaps better to delete this whole third bullet.--></ul> <t>Modification to VPN implementations is outside the present scope, which is why this section has so far focused on reconfiguration. Although this document does not define any requirements for VPN implementations, determining whether there is a need for such requirements could be one aspect of L4S experimentation.</t> </section> </section> <section anchor="l4sid_expts" numbered="true" toc="default"> <name>L4S Experiments</name> <t>This section describes open questions that L4SExperimentsexperiments ought to focus on. This section also documents outstanding open issues that will need to be investigated as part of L4S experimentation, given they could not be fully resolved during theWGworking group phase. It also lists metrics that will need to be monitored during experiments (summarizing text elsewhere in L4S documents) and finally lists some potential future directions that researchers might wish to investigate.</t> <t>In addition to this section, i) the DualQspec <xref target="I-D.ietf-tsvwg-aqm-dualq-coupled"spec <xref target="RFC9332" format="default"/> sets operational and management requirements for experiments with DualQ CoupledAQMs;AQMs, andGeneralii) general operational and management requirements for experiments with L4S congestion controls are given in Sections <xref target="l4sid_transport_req"format="default"/>format="counter"/> and <xref target="l4sid_network_req"format="default"/>format="counter"/> above,e.g. co-existencee.g., coexistence and scalingrequirements,requirements and incremental deployment arrangements.</t> <t>The specification of eachscalableScalable congestion control will need to include protocol-specific requirements for configuration and monitoring performance during experiments.Appendix A of the guidelines in<xref target="RFC5706"format="default"/>sectionFormat="of" section="A"/> provides a helpful checklist.</t> <section numbered="true" toc="default"> <name>Open Questions</name> <t>L4S experiments would be expected to answer the following questions:</t> <ul spacing="normal"> <li> <t>Have all the parts of L4S been deployed, and if so, what proportion of paths support it?</t> <ul spacing="normal"> <li>What types of L4S AQMs were deployed,e.g. FQ,e.g., FQ, coupled DualQ, uncoupled DualQ, other? And how prevalent was each?</li> <li>Are the signalling patterns emitted by the deployed AQMs in any way different from those expected when the Prague requirements for endpoints were written?</li> </ul> </li> <li>Does use of L4S over the Internet result in a significantly improved user experience?</li> <li>Has L4S enabled novel interactive applications?</li> <li> <t>Did use of L4S over the Internet result in improvements to the following metrics:</t> <ul spacing="normal"> <li>queue delay (mean and 99th percentile) under various loads;</li> <li>utilization;</li> <li>starvation /fairness;</li>fairness; and</li> <li>scaling range of flow rates and RTTs?</li> </ul> </li> <li>How dependent was the performance of L4S service on the bottleneck bandwidth or the path RTT?</li> <li>How much do bursty links in the Internet affect L4S performance (see "Underutilization with Bursty Links" in <xref target="Heist21" format="default"/>) and how prevalent are they? How much limitation of burstiness from upstream links was needed and/or was realized--- both at senders and at links, especially radio links -- or how much did L4S target delay have to be increased to accommodate the bursts (seebullet #7Item <xref target="l4sid_Prague_req-burstiness" format="counter"/> in <xreftarget="l4sid_congestion_response" format="default"/>target="l4sid_congestion_response"/> and see <xreftarget="l4sid_bursts_links_l4s_upstream" format="default"/>)?</li>target="l4sid_bursts_links_l4s_upstream"/>)?</li> <li>Is the initial experiment withmis-markedmis-identified bursty traffic at high RTT (see "Underutilization with Bursty Traffic" in <xref target="Heist21" format="default"/>) indicative of similar problems at lowerRTTs and,RTTs, and if so, how effective is the suggested remedy inAppendix A.1 of the<xref target="RFC9332" format="default" section="A.1" sectionFormat="of">the DualQspec <xref target="I-D.ietf-tsvwg-aqm-dualq-coupled" format="default"/>spec</xref> (or possible other remedies)?</li> <li> <t>Was per-flow queue protection typically (un)necessary? </t> <ul spacing="normal"> <li>How well did overload protection or queue protection work?</li> </ul> </li><li>How<li><t>How well did L4S flows coexist with Classic flows when sharing abottleneck?</li> <li>bottleneck?</t> <ul spacing="normal"> <li>How frequently did problems arise?</li> <li>What caused any coexistence problems, and were any problems due to single-queue Classic ECN AQMs (this assumes single-queue Classic ECN AQMs can be distinguished from FQ ones)?</li> </ul> </li> <li>How prevalent were problems with the L4S service due totunnels / encapsulationstunnels/encapsulations that do not support ECN decapsulation?</li> <li>How easy was it to implement a fully compliant L4S congestion control, over various different transport protocols (TCP, QUIC, RMCAT, etc.)?</li> </ul> <t>Monitoring for harm to other traffic, specifically bandwidth starvation or excess queuing delay, will need to be conducted alongside all early L4S experiments. It is hard, if not impossible, for an individual flow to measure its impact on other traffic. So such monitoring will need to be conducted using bespoke monitoring across flows and/or across classes of traffic.</t> </section> <section numbered="true" toc="default"> <name>Open Issues</name> <ul spacing="normal"> <li>What is the best way forward to deal with L4S over single-queue Classic ECN AQM bottlenecks, given current problems with misdetecting L4S AQMs as Classic ECN AQMs? See the L4S operationalguidance <xrefguidance <xref target="I-D.ietf-tsvwg-l4sops" format="default"/>.</li> <li>Fixing the poorInteractioninteraction between current L4S congestion controls and CoDel with only Classic ECN support during flow startup. Originally, this was due to a bug in the initialization of the congestionEWMAaverage in the Linux implementation of TCP Prague. That was quickly fixed, which removed the main performance impact, but further improvement would be useful(either by(by modifyingCoDel,either CoDel or Scalable congestion controls, or both).</li> </ul> </section> <section numbered="true" toc="default"> <name>Future Potential</name> <t>Researchers might find that L4S opens up the following interesting areas for investigation:</t> <ul spacing="normal"><li>Potential<li>potential for faster convergence time and tracking of available capacity;</li><li>Potential<li>potential for improvements to particular linktechnologies,technologies and cross-layer interactions with them;</li><li>Potential<li>potential for using virtual queues,e.g. toe.g., to further reduce latencyjitter,jitter or to leave headroom for capacity variation in radio networks;</li><li>Development<li>development and specification of reverse path congestion control using L4S building blocks(e.g. AccECN,(e.g., AccECN or QUIC);</li><li>Once<li>once queuing delay is cut down, what becomes the 'second-longest pole in the tent' (other than the speed of light)?</li><li>Novel<li>novel alternatives to the existing set of L4SAQMs;</li> <li>NovelAQMs; and</li> <li>novel applications enabled by L4S.</li> </ul> </section> </section> <section anchor="l4sid_IANA" numbered="true" toc="default"> <name>IANA Considerations</name> <t>The semantics of the 01 codepoint of the ECNFieldfield of the IP headerisare specified bythe presentthis Experimental RFC. The process for anexperimentalExperimental RFC to assign this codepoint in the IP header (v4 and v6) is documented in Proposed Standard <xref target="RFC8311" format="default"/>, which updates the Proposed Standard <xref target="RFC3168" format="default"/>.</t><t>When the present document is published as an RFC, IANA is asked to update<t>IANA has updated the 01 entry in theregistry,"ECN Field (Bits 6-7)"to the followingregistry (seehttps://www.iana.org/assignments/dscp-registry/dscp-registry.xhtml#ecn-field ):</t><eref target="https://www.iana.org/assignments/dscp-registry/" brackets="angle"/>) as follows:</t> <table align="center"> <name>ECN Field (Bits 6-7) Registry</name> <thead> <tr> <th align="left">Binary</th> <th align="left">Keyword</th> <thalign="left">References</th>align="left">Reference</th> </tr> </thead> <tbody> <tr> <td align="left">01</td> <td align="left">ECT(1) (ECN-Capable Transport(1))[1]</td> <td align="left"> <xref target="RFC8311" format="default"/>[RFC Errata 5399] [RFCXXXX]</td>[RFC Errata 5399] RFC 9331</td> </tr> </tbody> </table><t>[XXXX<t>[1] ECT(1) isthe number that the RFC Editor assigns to the present document (this sentence to be removed by the RFC Editor)].</t>for experimental use only <xref target="RFC8311" section="4.2" sectionFormat="comma"/></t> </section> <section anchor="l4sid_Security_Considerations" numbered="true" toc="default"> <name>Security Considerations</name> <t>Approaches to assure the integrity of signals using the new identifier are introducedin <xrefin <xref target="l4sid_competing_integrity" format="default"/>. See the security considerations in the L4Sarchitecture <xref target="I-D.ietf-tsvwg-l4s-arch"architecture <xref target="RFC9330" format="default"/> for further discussion ofmis-usemisuse of the identifier, as well as extensive discussion of policing rate and latency in regard to L4S.</t> <t>Defining ECT(1) as the L4S identifier introduces a difference between the effects ofECT0ECT(0) andECT1ECT(1) thatRFC 3168<xref target="RFC3168" format="default"/> previously defined as distinct but with equivalent effect. For L4S ECN, a network node is still required not to swap one to the other, even if the network operator chooses to locally apply the treatment associated with the opposite codepoint (see Sections <xref format="counter" target="l4sid_inclusion_dualq"/> and <xref format="counter" target="l4sid_exclusion_dualq"/>). These sections also describe the potential effects if a non-compliant or malicious network node does swap one to the other. The present specification does not change the effects of other unexpected transitions of theECNIP-ECN field, which are still as described inSection 18 of RFC 3168.</t><xref target="RFC3168" sectionFormat="of" section="18"/>.</t> <t>If the anti-replay window of a VPN egress is too small, it will mistake deliberate delay differences as a replayattack,attack and discardhigher delayhigher-delay packets(e.g. Classic)(e.g., Classic) carried within the same security association (SA) aslow delaylow-delay packets(e.g. L4S).(e.g., L4S). <xref target="l4sid_VPN_anti-replay" format="default"/> recommends that VPNs used in L4S experiments are configured with a sufficiently large anti-replay window, as required by the relevant specifications. It also discusses other alternatives.</t> <t>If a user taking part in the L4S experiment sets up a VPN without being aware of the above advice, and if the user allows anyone to send traffic into their VPN, they would open up a DoS vulnerability in which an attacker could induce the VPN's anti-replay mechanism to discard enough of the user's Classic (C) traffic (if they are receiving any) to cause a significant rate reduction. While the user is actively downloading C traffic, the attacker sends C traffic into the VPN to fill the remainder of the bottleneck link, then sends intermittent L4S packets to maximize the chance of exceeding the VPN's replay window. The user can prevent this attack by following the recommendations in <xref target="l4sid_VPN_anti-replay"format="default"/>.<!--ToDo: I'm not sure this para is warranted. It's a bit lame to say there's an attack if you don't follow advice, and the solution to the attack is to follow the advice.-->format="default"/>. </t> <t>The recommendation to detect loss in time units prevents the ACK-splitting attacks described in <xref target="Savage-TCP" format="default"/>.</t> </section> </middle><!-- *****BACK MATTER ***** --><back> <displayreference target="I-D.ietf-tcpm-accurate-ecn" to="ACCECN"/> <displayreference target="I-D.ietf-tsvwg-ecn-encap-guidelines" to="ECN-ENCAP"/> <displayreference target="I-D.ietf-tsvwg-rfc6040update-shim" to="ECN-SHIM"/> <displayreference target="I-D.ietf-trill-ecn-support" to="TRILL-ECN-SUPPORT"/> <displayreference target="I-D.stewart-tsvwg-sctpecn" to="SCTP-ECN"/> <displayreference target="I-D.briscoe-tsvwg-l4s-diffserv" to="L4S-DIFFSERV"/> <displayreference target="I-D.ietf-tsvwg-nqb" to="NQB-PHB"/> <displayreference target="I-D.ietf-tsvwg-l4sops" to="L4SOPS"/> <displayreference target="I-D.ietf-tcpm-generalized-ecn" to="ECN++"/> <displayreference target="I-D.sridharan-tcpm-ctcp" to="CTCP"/> <displayreference target="I-D.briscoe-docsis-q-protection" to="DOCSIS-QPROT"/> <displayreference target="I-D.briscoe-iccrg-prague-congestion-control" to="PRAGUE-CC"/> <displayreference target="I-D.cardwell-iccrg-bbr-congestion-control" to="BBR-CC"/> <displayreference target="I-D.mathis-iccrg-relentless-tcp" to="RELENTLESS"/> <references> <name>References</name> <references> <name>Normative References</name> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2119.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3168.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4774.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6679.xml"/> </references> <references> <name>Informative References</name> <!-- [I-D.ietf-tsvwg-aqm-dualq-coupled]; companion document 9332 - title matches as of 1/17/23 --> <referenceanchor="RFC2119" target="https://www.rfc-editor.org/info/rfc2119" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.2119.xml">anchor='RFC9332' target='https://www.rfc-editor.org/info/rfc9332'> <front><title>Key words<title>Dual-Queue Coupled Active Queue Management (AQM) foruse in RFCs to Indicate Requirement Levels</title>Low Latency, Low Loss, and Scalable Throughput (L4S)</title> <authorinitials="S." surname="Bradner" fullname="S. Bradner"> <organization/>initials='K' surname='De Schepper' fullname='Koen De Schepper'> <organization /> </author> <author initials='B' surname='Briscoe' fullname='Bob Briscoe' role="editor"> <organization /> </author> <author initials='G' surname='White' fullname='Greg White'> <organization /> </author> <dateyear="1997" month="March"/> <abstract> <t>In many standards track documents several words are used to signify the requirements in the specification. These words are often capitalized. Thismonth="January" year="2023"/> </front> <seriesInfo name="RFC" value="9332"/> <seriesInfo name="DOI" value="10.17487/RFC9332"/> </reference> <!-- [I-D.ietf-tsvwg-l4s-arch]; companion documentdefines these words9330 - title matches asthey should be interpreted in IETF documents. This document specifies an Internet Best Current Practices for the Internet Community, and requests discussionof 1/17/23 --> <reference anchor='RFC9330' target='https://www.rfc-editor.org/info/rfc9330'> <front> <title>Low Latency, Low Loss, andsuggestions for improvements.</t> </abstract>Scalable Throughput (L4S) Internet Service: Architecture</title> <author initials='B' surname='Briscoe' fullname='Bob Briscoe' role='editor' /> <author initials='K' surname='De Schepper' fullname='Koen De Schepper' /> <author initials='M' surname='Bagnulo' fullname='Marcelo Bagnulo' /> <author initials='G' surname='White' fullname='Greg White' /> <date month="January" year="2023"/> </front> <seriesInfoname="BCP" value="14"/> <seriesInfoname="RFC"value="2119"/>value="9330"/> <seriesInfo name="DOI"value="10.17487/RFC2119"/>value="10.17487/RFC9330"/> </reference> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2474.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3540.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3246.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3649.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4301.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4302.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4303.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4340.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4341.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4342.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5033.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5129.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5348.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5622.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5865.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4960.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5562.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5681.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5706.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6040.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6077.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6660.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6675.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7560.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7567.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7713.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5925.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8033.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8083.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8085.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8174.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8290.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8298.xml"/> <!-- [I-D.ietf-tcpm-accurate-ecn] IESG state I-D Exists as of 1/17/23--> <xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.ietf-tcpm-accurate-ecn.xml"/> <!-- [I-D.ietf-tsvwg-ecn-encap-guidelines] Expired as of 1/17/23 --> <xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.ietf-tsvwg-ecn-encap-guidelines.xml"/> <!-- [I-D.ietf-tsvwg-rfc6040update-shim] Expired as of 1/17/23 --> <xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.ietf-tsvwg-rfc6040update-shim.xml"/> <!-- [I-D.ietf-trill-ecn-support] in MISSREF state as of 1/17/23. xi:include incorrectly shows Eastlake's name as "D. E. Eastlake" but author name preferences show he wants "D. Eastlake 3rd" --> <referenceanchor="RFC3168" target="https://www.rfc-editor.org/info/rfc3168" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.3168.xml">anchor="I-D.ietf-trill-ecn-support"> <front><title>The Addition<title>TRILL (TRansparent Interconnection ofExplicitLots of Links): ECN (Explicit CongestionNotification (ECN) to IP</title> <author initials="K." surname="Ramakrishnan" fullname="K. Ramakrishnan"> <organization/> </author>Notification) Support</title> <authorinitials="S." surname="Floyd" fullname="S. Floyd"> <organization/>initials="D." surname="Eastlake 3rd" fullname="Donald Eastlake 3rd"> <organization>Huawei</organization> </author> <authorinitials="D." surname="Black" fullname="D. Black"> <organization/>initials="B." surname="Briscoe" fullname="Bob Briscoe"> <organization>CableLabs</organization> </author> <dateyear="2001" month="September"/> <abstract> <t>This memo specifies the incorporation of ECN (Explicit Congestion Notification) to TCP and IP, including ECN's usemonth="February" day="25" year="2018"/> </front> <seriesInfo name="Internet-Draft" value="draft-ietf-trill-ecn-support-07"/> </reference> <!-- [I-D.stewart-tsvwg-sctpecn] IESG state Expired as oftwo bits in1/17/23. xi:include incorrectly shows Stewart's name as "R. R. Stewart" when theIP header. [STANDARDS-TRACK]</t> </abstract>draft only has "R. Stewart" --> <reference anchor="I-D.stewart-tsvwg-sctpecn"> <front> <title>ECN for Stream Control Transmission Protocol (SCTP)</title> <author initials="R." surname="Stewart" fullname="Randall R. Stewart"> <organization>Adara Networks</organization> </author> <author initials="M." surname="Tüxen" fullname="Michael Tüxen"> <organization>Muenster Univ. of Appl. Sciences</organization> </author> <author initials="X." surname="Dong" fullname="Xuesong Dong"> <organization>Huawei</organization> </author> <date month="January" day="15" year="2014"/> </front> <seriesInfoname="RFC" value="3168"/> <seriesInfo name="DOI" value="10.17487/RFC3168"/>name="Internet-Draft" value="draft-stewart-tsvwg-sctpecn-05"/> </reference> <!-- [I-D.briscoe-tsvwg-l4s-diffserv] IESG state Expired as of 1/17/23. xi:include wrong date: --> <referenceanchor="RFC4774" target="https://www.rfc-editor.org/info/rfc4774" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.4774.xml">anchor="I-D.briscoe-tsvwg-l4s-diffserv"> <front><title>Specifying Alternate Semantics for the Explicit Congestion Notification (ECN) Field</title><title>Interactions between Low Latency, Low Loss, Scalable Throughput (L4S) and Differentiated Services</title> <authorinitials="S." surname="Floyd" fullname="S. Floyd"> <organization/>fullname="Bob Briscoe" surname="Briscoe" initials="B"> <organization>CableLabs</organization> </author> <dateyear="2006" month="November"/> <abstract> <t>There have been a numbermonth="November" day="1" year="2018"/> </front> <seriesInfo name="Internet-Draft" value="draft-briscoe-tsvwg-l4s-diffserv-02"/> </reference> <!-- [I-D.ietf-tsvwg-nqb] IESG state I-D Exists.--> <xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.ietf-tsvwg-nqb.xml"/> <!-- [I-D.ietf-tsvwg-l4sops] IESG state I-D Exists as ofproposals for alternate semantics1/17/23. xi:include missing editor role--> <reference anchor="I-D.ietf-tsvwg-l4sops"> <front> <title>Operational Guidance forthe Explicit Congestion Notification (ECN) field in the IP header RFC 3168. This document discusses someDeployment ofthe issuesL4S indefining alternate semantics fortheECN field, and specifies requirements for a safe coexistence in anInternetthat could include routers that do not understand the defined alternate semantics. This document evolved</title> <author fullname="Greg White" surname="White" initials="G" role="editor"> <organization>CableLabs</organization> </author> <date month="April" day="28" year="2022"/> </front> <seriesInfo name="Internet-Draft" value="draft-ietf-tsvwg-l4sops-03"/> </reference> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8311.xml"/> <!-- [I-D.ietf-tcpm-generalized-ecn] IESG state I-D Exists asa result of discussions with the authorsofone recent proposal for such alternate semantics. This document specifies an Internet Best Current Practices1/1/7/23 --> <xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.ietf-tcpm-generalized-ecn.xml"/> <reference anchor="ARED01" target="https://www.icsi.berkeley.edu/icsi/node/2032"> <front> <title>Adaptive RED: An Algorithm for Increasing theInternet Community, and requests discussion and suggestions for improvements.</t> </abstract>Robustness of RED's Active Queue Management</title> <author fullname="Sally Floyd" initials="S." surname="Floyd"> <organization>ACIRI</organization> </author> <author fullname="Ramakrishna Gummadi" initials="R." surname="Gummadi"> <organization>ACIRI</organization> </author> <author fullname="S. Shenker" initials="S." surname="Shenker"> <organization>ACIRI</organization> </author> <date month="August" year="2001"/> </front><seriesInfo name="BCP" value="124"/> <seriesInfo name="RFC" value="4774"/> <seriesInfo name="DOI" value="10.17487/RFC4774"/><refcontent>ACIRI Technical Report 301</refcontent> </reference> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8257.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8312.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8511.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8985.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8888.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.9000.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.9001.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.9147.xml"/> <!-- [I-D.sridharan-tcpm-ctcp] IESG state Expired as of 1/1/7/23. xi:include Wrong date.--> <referenceanchor="RFC6679" target="https://www.rfc-editor.org/info/rfc6679" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.6679.xml">anchor="I-D.sridharan-tcpm-ctcp"> <front><title>Explicit<title>Compound TCP: A New TCP CongestionNotification (ECN)Control forRTP over UDP</title>High-Speed and Long Distance Networks </title> <author initials="M."surname="Westerlund" fullname="M. Westerlund"> <organization/> </author> <author initials="I." surname="Johansson" fullname="I. Johansson"> <organization/>surname="Sridharan" fullname="Murari Sridharan"> <organization>Microsoft</organization> </author> <authorinitials="C." surname="Perkins" fullname="C. Perkins"> <organization/>initials="K." surname="Tan" fullname="Kun Tan"> <organization>Microsoft Research</organization> </author> <authorinitials="P." surname="O'Hanlon" fullname="P. O'Hanlon"> <organization/>initials="D." surname="Bansal" fullname="Deepak Bansal"> <organization>Microsoft</organization> </author> <authorinitials="K." surname="Carlberg" fullname="K. Carlberg"> <organization/>initials="D." surname="Thaler" fullname="Dave Thaler"> <organization>Microsoft</organization> </author> <dateyear="2012" month="August"/> <abstract> <t>This memo specifies how Explicit Congestion Notification (ECN) can be used with the Real-time Transport Protocol (RTP) running over UDP, using the RTP Control Protocol (RTCP) as a feedback mechanism. It defines a new RTCP Extended Report (XR) block for periodic ECN feedback, a new RTCP transport feedback message for timely reporting of congestion events, and a Session Traversal Utilities for NAT (STUN) extension used in the optional initialisation method using Interactive Connectivity Establishment (ICE). Signalling and procedures for negotiation of capabilities and initialisation methods are also defined. [STANDARDS-TRACK]</t> </abstract>month="November" day="3" year="2008"/> </front> <seriesInfoname="RFC" value="6679"/> <seriesInfo name="DOI" value="10.17487/RFC6679"/>name="Internet-Draft" value="draft-sridharan-tcpm-ctcp-02"/> </reference></references> <references> <name>Informative References</name><!-- [I-D.briscoe-docsis-q-protection] in MISSREF state as of 1/17/23. xi:include is missing editor role --> <referenceanchor="RFC2474" target="https://www.rfc-editor.org/info/rfc2474" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.2474.xml">anchor="I-D.briscoe-docsis-q-protection"> <front><title>Definition of the Differentiated Services Field (DS Field) in the IPv4 and IPv6 Headers</title> <author initials="K." surname="Nichols" fullname="K. Nichols"> <organization/> </author> <author initials="S." surname="Blake" fullname="S. Blake"> <organization/> </author><title>The DOCSIS® Queue Protection Algorithm to Preserve Low Latency</title> <authorinitials="F." surname="Baker" fullname="F. Baker"> <organization/>fullname="Bob Briscoe" role="editor"> <organization>Independent</organization> </author> <authorinitials="D." surname="Black" fullname="D. Black"> <organization/>fullname="Greg White"> <organization>CableLabs</organization> </author> <dateyear="1998" month="December"/> <abstract> <t>This document defines the IP header field, called the DS (for differentiated services) field. [STANDARDS-TRACK]</t> </abstract>day="13" month="May" year="2022"/> </front> <seriesInfoname="RFC" value="2474"/> <seriesInfo name="DOI" value="10.17487/RFC2474"/>name="Internet-Draft" value="draft-briscoe-docsis-q-protection-06"/> </reference> <!-- [I-D.briscoe-iccrg-prague-congestion-control] Expired as of 1/17/23. Missing editor role. --> <referenceanchor="RFC2309" target="https://www.rfc-editor.org/info/rfc2309" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.2309.xml">anchor="I-D.briscoe-iccrg-prague-congestion-control"> <front><title>Recommendations on Queue Management and<title>Prague CongestionAvoidance in the Internet</title>Control</title> <authorinitials="B." surname="Braden" fullname="B. Braden"> <organization/>fullname="Koen De Schepper" surname="De Schepper" initials="K."> <organization>Nokia Bell Labs</organization> </author> <authorinitials="D." surname="Clark" fullname="D. Clark"> <organization/>fullname="Olivier Tilmans"> <organization>Nokia Bell Labs</organization> </author> <authorinitials="J." surname="Crowcroft" fullname="J. Crowcroft"> <organization/>fullname="Bob Briscoe" role="editor"> <organization>Independent</organization> </author> <date month="July" day="11" year="2022"/> </front> <seriesInfo name="Internet-Draft" value="draft-briscoe-iccrg-prague-congestion-control-01"/> </reference> <!-- [I-D.cardwell-iccrg-bbr-congestion-control] IESG state Expired as of 1/17/23. xi:include incorrectly shows Hassas Yeganeh's name as "S. H. Yeganeh" when it should be "S. Hassas Yeganeh" --> <reference anchor="I-D.cardwell-iccrg-bbr-congestion-control"> <front> <title>BBR Congestion Control</title> <authorinitials="B." surname="Davie" fullname="B. Davie"> <organization/>initials="N." surname="Cardwell" fullname="Neal Cardwell"> <organization>Google</organization> </author> <authorinitials="S." surname="Deering" fullname="S. Deering"> <organization/>initials="Y." surname="Cheng" fullname="Yuchung Cheng"> <organization>Google</organization> </author> <authorinitials="D." surname="Estrin" fullname="D. Estrin"> <organization/>initials="S." surname="Hassas Yeganeh" fullname="Soheil Hassas Yeganeh"> <organization>Google</organization> </author> <authorinitials="S." surname="Floyd" fullname="S. Floyd"> <organization/>initials="I." surname="Swett" fullname="Ian Swett"> <organization>Google</organization> </author> <author initials="V." surname="Jacobson"fullname="V.fullname="Van Jacobson"><organization/><organization>Google</organization> </author> <date month="March" day="7" year="2022"/> </front> <seriesInfo name="Internet-Draft" value="draft-cardwell-iccrg-bbr-congestion-control-02"/> </reference> <!-- [I-D.mathis-iccrg-relentless-tcp] IESG state Expired as of 1/17/23 --> <xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.mathis-iccrg-relentless-tcp.xml"/> <reference anchor="BBRv2" target="https://github.com/google/bbr"> <front> <title>TCP BBR v2 Alpha/Preview Release</title> <author/> <date month="June" year="2022"/> </front> <refcontent>commit 17700ca</refcontent> </reference> <reference anchor="DCttH19" target="https://bobbriscoe.net/projects/latency/dctth_journal_draft20190726.pdf"> <front> <title>'Data Centre to the Home': Ultra-Low Latency for All</title> <authorinitials="G." surname="Minshall" fullname="G. Minshall"> <organization/>fullname="Koen De Schepper" initials="K." surname="De Schepper"> <organization>Nokia Bell Labs</organization> </author> <authorinitials="C." surname="Partridge" fullname="C. Partridge"> <organization/>fullname="Olga Bondarenko" initials="O." surname="Bondarenko"> <organization>Simula Research Lab</organization> </author> <authorinitials="L." surname="Peterson" fullname="L. Peterson"> <organization/>fullname="Olivier" initials="O." surname="Tilmans"> <organization>Nokia Bell Labs</organization> </author> <author fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization>Independent (bobbriscoe.net)</organization> </author> <date month="July" year="2019"/> </front> <refcontent>Updated RITE project Technical Report</refcontent> </reference> <reference anchor="L4Seval22" target="https://arxiv.org/abs/2209.01078"> <front> <title>Dual Queue Coupled AQM: Deployable Very Low Queuing Delay for All</title> <author fullname="Koen De Schepper" initials="K."surname="Ramakrishnan" fullname="K. Ramakrishnan"> <organization/>surname="De Schepper"> <organization>Nokia Bell Labs</organization> </author> <authorinitials="S." surname="Shenker" fullname="S. Shenker"> <organization/>fullname="Olga Albisser" initials="O." surname="Albisser"> <organization>Simula Research Lab</organization> </author> <authorinitials="J." surname="Wroclawski" fullname="J. Wroclawski"> <organization/>fullname="Olivier Tilmans" initials="O." surname="Tilmans"> <organization>Nokia Bell Labs</organization> </author> <authorinitials="L." surname="Zhang" fullname="L. Zhang"> <organization/>fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization>Independent (bobbriscoe.net)</organization> </author> <dateyear="1998" month="April"/> <abstract> <t>This memo presents two recommendations to the Internet community concerning measures to improve and preserve Internet performance. It presents a strong recommendation for testing, standardization, and widespread deployment of active queue management in routers, to improve the performance of today's Internet. It also urges a concerted effort of research, measurement, and ultimate deployment of router mechanisms to protect the Internet from flows that are not sufficiently responsive to congestion notification. This memo provides information for the Internet community. It does not specify an Internet standard of any kind.</t> </abstract>month="September" year="2022"/> </front> <seriesInfoname="RFC" value="2309"/> <seriesInfoname="DOI"value="10.17487/RFC2309"/>value="10.48550/arXiv.2209.01078"/> <refcontent>Preprint submitted to IEEE/ACM Transactions on Networking</refcontent> </reference> <referenceanchor="RFC3540" target="https://www.rfc-editor.org/info/rfc3540" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.3540.xml">anchor="PI2" target="https://dl.acm.org/citation.cfm?doid=2999572.2999578"> <front><title>Robust Explicit Congestion Notification (ECN) Signaling with Nonces</title><title>PI^2: A Linearized AQM for both Classic and Scalable TCP</title> <authorinitials="N." surname="Spring" fullname="N. Spring"> <organization/>fullname="Koen De Schepper" initials="K." surname="De Schepper"> <organization>Bell Labs</organization> </author> <authorinitials="D." surname="Wetherall" fullname="D. Wetherall"> <organization/>fullname="Olga Bondarenko" initials="O." surname="Bondarenko"> <organization>Simula Research Lab</organization> </author> <authorinitials="D." surname="Ely" fullname="D. Ely"> <organization/>fullname="Ing-jyh Tsang" initials="I." surname="Tsang"> <organization>Bell Labs</organization> </author> <author fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization>BT</organization> </author> <dateyear="2003" month="June"/> <abstract> <t>This note describes the Explicit Congestion Notification (ECN)-nonce, an optional addition to ECN that protects against accidental or malicious concealment of marked packets from the TCP sender. It improves the robustness of congestion control by preventing receivers from exploiting ECN to gain an unfair share of network bandwidth. The ECN-nonce uses the two ECN-Capable Transport (ECT)codepoints in the ECN field of the IP header, and requires a flag in the TCP header. It is computationally efficient for both routers and hosts. This memo defines an Experimental Protocol for the Internet community.</t> </abstract>month="December" year="2016"/> </front> <seriesInfoname="RFC" value="3540"/> <seriesInfoname="DOI"value="10.17487/RFC3540"/>value="10.1145/2999572.2999578"/> <refcontent>Proceedings of ACM CoNEXT 2016, pp. 105-119</refcontent> </reference> <referenceanchor="RFC3246" target="https://www.rfc-editor.org/info/rfc3246" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.3246.xml">anchor="VCP" target="https://doi.acm.org/10.1145/1080091.1080098"> <front><title>An Expedited Forwarding PHB (Per-Hop Behavior)</title> <author initials="B." surname="Davie" fullname="B. Davie"> <organization/> </author> <author initials="A." surname="Charny" fullname="A. Charny"> <organization/> </author> <author initials="J.C.R." surname="Bennet" fullname="J.C.R. Bennet"> <organization/> </author><title>One more bit is enough</title> <authorinitials="K." surname="Benson" fullname="K. Benson">fullname="Yong Xia" initials="Y." surname="Xia"> <organization/> </author> <authorinitials="J.Y." surname="Le Boudec" fullname="J.Y. Le Boudec">fullname="Lakshminarayanan Subramanian" initials="L." surname="Subramanian"> <organization/> </author> <authorinitials="W." surname="Courtney" fullname="W. Courtney">fullname="Ion Stoica" initials="I." surname="Stoica"> <organization/> </author> <author fullname="Shivkumar Kalyanaraman" initials="S."surname="Davari" fullname="S. Davari"> <organization/> </author> <author initials="V." surname="Firoiu" fullname="V. Firoiu"> <organization/> </author> <author initials="D." surname="Stiliadis" fullname="D. Stiliadis">surname="Kalyanaraman"> <organization/> </author> <dateyear="2002" month="March"/> <abstract> <t>This document defines a PHB (per-hop behavior) called Expedited Forwarding (EF). The PHB is a basic building block in the Differentiated Services architecture. EF is intended to provide a building block for low delay, low jitter and low loss services by ensuring that the EF aggregate is served at a certain configured rate. This document obsoletes RFC 2598. [STANDARDS-TRACK]</t> </abstract>month="August" year="2005"/> </front> <seriesInfoname="RFC" value="3246"/> <seriesInfoname="DOI"value="10.17487/RFC3246"/>value="10.1145/1080091.1080098"/> <refcontent>SIGCOMM '05: Proceedings of the 2005 conference on Applications, technologies, architectures, and protocols for computer communications, pp. 37–48</refcontent> </reference> <referenceanchor="RFC3649" target="https://www.rfc-editor.org/info/rfc3649" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.3649.xml">anchor="QV" target="https://riteproject.files.wordpress.com/2015/12/rite-deliverable-2-3.pdf"> <front><title>HighSpeed TCP for Large Congestion Windows</title><title>Report on Prototype Development and Evaluation of Network and Interaction Techniques</title> <authorinitials="S." surname="Floyd" fullname="S. Floyd"> <organization/>fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization>BT</organization> </author> <author fullname="Per Hurtig" initials="P." surname="Hurtig"> <organization>Karlstad University</organization> </author> <dateyear="2003" month="December"/> <abstract> <t>The proposals in this document are experimental. While they may be deployed in the current Internet, they do not represent a consensus that this is the best method for high-speed congestion control. In particular, we note that alternative experimental proposals are likely to be forthcoming, and it is not well understood how the proposals in this document will interact with such alternative proposals. This document proposes HighSpeed TCP, a modificationmonth="September" year="2015"/> </front> <refcontent>RITE Technical Report, Deliverable 2.3, Appendix C.2: "Up toTCP's congestion control mechanism for use with TCP connectionsSpeed withlarge congestion windows. The congestion control mechanismsQueue View"</refcontent> </reference> <reference anchor="Alizadeh-stability" target="https://dl.acm.org/doi/10.1145/1993744.1993753"> <front> <title>Analysis ofthe current Standard TCP constrains the congestion windows that can be achieved by TCP in realistic environments. For example, for a Standard TCP connection with 1500-byte packetsDCTCP: Stability, Convergence, anda 100 ms round-trip time, achieving a steady-state throughput of 10 Gbps would require an average congestion windowFairness</title> <author fullname="Mohamed Alizadeh" initials="M." surname="Alizadeh"/> <author fullname="Adel Javanmard" initials="A." surname="Javanmard"/> <author fullname="Balaji Prabhakar" initials="B." surname="Prabhakar"/> <date month="June" year="2011"/> </front> <seriesInfo name="DOI" value="10.1145/1993744.1993753"/> <refcontent>SIGMETRICS '11: Proceedings of83,333 segments,the ACM SIGMETRICS Joint International Conference on Measurement anda packet drop rate of at most one congestion event every 5,000,000,000 packets (or equivalently, at most one congestion event every 1 2/3 hours). This is widely acknowledged as an unrealistic constraint. To address his limitationModeling ofTCP, this document proposes HighSpeed TCP, and solicits experimentation and feedback from the wider community.</t> </abstract>Computer Systems, pp. 73-84</refcontent> <format target="https://people.csail.mit.edu/alizadeh/papers/dctcp_analysis-sigmetrics11.pdf" type="PDF"/> </reference> <reference anchor="TCPPrague" target="https://www.ietf.org/mail-archive/web/tcpprague/current/msg00001.html"> <front> <title>Notes: DCTCP evolution 'bar BoF': Tue 21 Jul 2015, 17:40, Prague</title> <author fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization>Simula</organization> </author> <date month="July" year="2015"/> </front><seriesInfo name="RFC" value="3649"/> <seriesInfo name="DOI" value="10.17487/RFC3649"/><refcontent>message to the tcpPrague mailing list</refcontent> </reference> <referenceanchor="RFC4301" target="https://www.rfc-editor.org/info/rfc4301" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.4301.xml">anchor="sub-mss-prob" target="https://arxiv.org/abs/1904.07598"> <front><title>Security Architecture<title>Scaling TCP's Congestion Window forthe Internet Protocol</title>Small Round Trip Times</title> <authorinitials="S." surname="Kent" fullname="S. Kent"> <organization/>fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization>BT</organization> </author> <author fullname="Koen De Schepper" initials="K."surname="Seo" fullname="K. Seo"> <organization/>surname="De Schepper"> <organization>Bell Labs</organization> </author> <dateyear="2005" month="December"/> <abstract> <t>This document describes an updated version of the "Security Architecture for IP", which is designed to provide security services for traffic at the IP layer. This document obsoletes RFC 2401 (November 1998). [STANDARDS-TRACK]</t> </abstract>month="May" year="2015"/> </front> <seriesInfoname="RFC" value="4301"/> <seriesInfoname="DOI"value="10.17487/RFC4301"/>value="10.48550/arXiv.1904.07598"/> <refcontent>BT Technical Report: TR-TUB8-2015-002</refcontent> </reference> <referenceanchor="RFC4302" target="https://www.rfc-editor.org/info/rfc4302" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.4302.xml">anchor="ecn-fallback" target="https://arxiv.org/abs/1911.00710"> <front><title>IP Authentication Header</title><title>TCP Prague Fall-back on Detection of a Classic ECN AQM</title> <authorinitials="S." surname="Kent" fullname="S. Kent"> <organization/>fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization>Independent</organization> </author><date year="2005" month="December"/> <abstract> <t>This document describes an updated version of the IP Authentication Header (AH), which is designed to provide authentication services in IPv4<author fullname="Asad Sajjad Ahmed" initials="A." surname="Ahmed"> <organization>Simula andIPv6. This document obsoletes RFC 2402 (November 1998). [STANDARDS-TRACK]</t> </abstract>Uni Oslo</organization> </author> <date month="February" year="2021"/> </front> <seriesInfoname="RFC" value="4302"/> <seriesInfoname="DOI"value="10.17487/RFC4302"/>value="10.48550/arXiv.1911.00710"/> <refcontent>Technical Report: TR-BB-2019-002</refcontent> </reference> <referenceanchor="RFC4303" target="https://www.rfc-editor.org/info/rfc4303" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.4303.xml">anchor="Ahmed19" target="https://www.duo.uio.no/handle/10852/70966"> <front><title>IP Encapsulating Security Payload (ESP)</title><title>Extending TCP for Low Round Trip Delay</title> <authorinitials="S." surname="Kent" fullname="S. Kent"> <organization/>fullname="Asad Sajjad Ahmed" initials="A.S." surname="Ahmed"> <organization>Simula and Uni Oslo</organization> </author> <dateyear="2005" month="December"/> <abstract> <t>This document describes an updated version of the Encapsulating Security Payload (ESP) protocol, which is designed to provide a mix of security services in IPv4 and IPv6. ESP is used to provide confidentiality, data origin authentication, connectionless integrity, an anti-replay service (a form of partial sequence integrity), and limited traffic flow confidentiality. This document obsoletes RFC 2406 (November 1998). [STANDARDS-TRACK]</t> </abstract>month="August" year="2019"/> </front><seriesInfo name="RFC" value="4303"/> <seriesInfo name="DOI" value="10.17487/RFC4303"/><refcontent>Master's Thesis, University of Oslo</refcontent> </reference> <referenceanchor="RFC4340" target="https://www.rfc-editor.org/info/rfc4340" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.4340.xml">anchor="A2DTCP" target="https://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6871352"> <front><title>Datagram Congestion Control Protocol (DCCP)</title><title>Adaptive-Acceleration Data Center TCP</title> <author fullname="Tao Zhang" initials="T." surname="Zhang"> <organization/> </author> <authorinitials="E." surname="Kohler" fullname="E. Kohler">fullname="Jianxin Wang" initials="J." surname="Wang"> <organization/> </author> <authorinitials="M." surname="Handley" fullname="M. Handley">fullname="Jiawei Huang" initials="J." surname="Huang"> <organization/> </author> <authorinitials="S." surname="Floyd" fullname="S. Floyd">fullname="Yi Huang" initials="Y." surname="Huang"> <organization/> </author> <author fullname="Jianer Chen" initials="J." surname="Chen"> <organization/> </author> <author fullname="Yi Pan" initials="Y." surname="Pan"> <organization/> </author> <dateyear="2006" month="March"/> <abstract> <t>The Datagram Congestion Control Protocol (DCCP) is a transport protocol that provides bidirectional unicast connections of congestion-controlled unreliable datagrams. DCCP is suitable for applications that transfer fairly large amounts of data and that can benefit from control over the tradeoff between timeliness and reliability. [STANDARDS-TRACK]</t> </abstract>month="June" year="2015"/> </front> <seriesInfoname="RFC" value="4340"/> <seriesInfoname="DOI"value="10.17487/RFC4340"/>value="10.1109/TC.2014.2345393"/> <refcontent>IEEE Transactions on Computers, Volume 64, Issue 6, pp. 1522-1533</refcontent> </reference> <referenceanchor="RFC4341" target="https://www.rfc-editor.org/info/rfc4341" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.4341.xml">anchor="PragueLinux" target="https://www.netdevconf.org/0x13/session.html?talk-tcp-prague-l4s"> <front><title>Profile<title>Implementing the 'TCP Prague' Requirements forDatagram Congestion Control Protocol (DCCP) Congestion Control ID 2: TCP-like Congestion Control</title>L4S</title> <authorinitials="S." surname="Floyd" fullname="S. Floyd"> <organization/>fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization>Independent</organization> </author> <authorinitials="E." surname="Kohler" fullname="E. Kohler"> <organization/>fullname="Koen De Schepper" initials="K." surname="De Schepper"> <organization>Nokia Bell Labs</organization> </author><date year="2006" month="March"/> <abstract> <t>This document contains the profile for Congestion Control Identifier 2 (CCID 2), TCP-like Congestion Control, in the Datagram Congestion Control Protocol (DCCP). CCID 2 should be used by senders who would like to take advantage of the available bandwidth in an environment with rapidly changing conditions, and who are able to adapt to the abrupt changes in the congestion window typical<author fullname="Olga Albisser" initials="O." surname="Albisser"> <organization>Simula</organization> </author> <author fullname="Joakim Misund" initials="J." surname="Misund"> <organization>University ofTCP's Additive Increase Multiplicative Decrease (AIMD) congestion control. [STANDARDS-TRACK]</t> </abstract>Oslo</organization> </author> <author fullname="Olivier Tilmans" initials="O." surname="Tilmans"> <organization>Nokia Bell Labs</organization> </author> <author fullname="Mirja Kühlewind" initials="M." surname="Kühlewind"> <organization>ETHZ</organization> </author> <author fullname="Asad Sajjad Ahmed" initials="A." surname="Ahmed"> <organization>Simula</organization> </author> <date month="March" year="2019"/> </front><seriesInfo name="RFC" value="4341"/> <seriesInfo name="DOI" value="10.17487/RFC4341"/><refcontent>Proceedings of Linux Netdev 0x13</refcontent> <format target="https://www.files.netdevconf.info/f/4d6939d5f1fb404fafd1/?dl=1" type="PDF"/> </reference> <referenceanchor="RFC4342" target="https://www.rfc-editor.org/info/rfc4342" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.4342.xml">anchor="LinuxPacedChirping" target="https://legacy.netdevconf.info/0x13/session.html?talk-chirp"> <front><title>Profile for Datagram Congestion Control Protocol (DCCP) Congestion Control ID 3: TCP-Friendly Rate Control (TFRC)</title><title>Paced Chirping - Rethinking TCP start-up</title> <authorinitials="S." surname="Floyd" fullname="S. Floyd"> <organization/>fullname="Joakim Misund" initials="J." surname="Misund"> <organization>University of Oslo</organization> </author> <author fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization>Independent</organization> </author> <date month="March" year="2019"/> </front> <refcontent>Proceedings of Linux Netdev 0x13</refcontent> </reference> <reference anchor="TCP-CA" target="https://ee.lbl.gov/papers/congavoid.pdf"> <front> <title>Congestion Avoidance and Control</title> <authorinitials="E." surname="Kohler" fullname="E. Kohler">fullname="Van Jacobson" initials="V." surname="Jacobson"> <organization/> </author> <authorinitials="J." surname="Padhye" fullname="J. Padhye">fullname="Michael J. Karels" initials="M. J." surname="Karels"> <organization/> </author> <dateyear="2006" month="March"/> <abstract> <t>This document contains the profile for Congestion Control Identifier 3, TCP-Friendly Rate Control (TFRC), in the Datagram Congestion Control Protocol (DCCP). CCID 3 should be used by senders that want a TCP-friendly sending rate, possibly with Explicit Congestion Notification (ECN), while minimizing abrupt rate changes. [STANDARDS-TRACK]</t> </abstract>month="November" year="1988"/> </front><seriesInfo name="RFC" value="4342"/> <seriesInfo name="DOI" value="10.17487/RFC4342"/><refcontent>Laurence Berkeley Labs Technical Report</refcontent> </reference> <referenceanchor="RFC5033" target="https://www.rfc-editor.org/info/rfc5033" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.5033.xml">anchor="Savage-TCP" target="https://dl.acm.org/doi/abs/10.1145/505696.505704"> <front><title>Specifying New<title>TCP Congestion ControlAlgorithms</title>with a Misbehaving Receiver</title> <author fullname="Stefan Savage" initials="S."surname="Floyd" fullname="S. Floyd"> <organization/>surname="Savage"> <organization>University of Washington</organization> </author> <authorinitials="M." surname="Allman" fullname="M. Allman"> <organization/>fullname="Neal Cardwell" initials="N." surname="Cardwell"> <organization>University of Washington</organization> </author><date year="2007" month="August"/> <abstract> <t>The IETF's standard congestion control schemes have been widely shown to be inadequate for various environments (e.g., high-speed networks). Recent research has yielded many alternate congestion control schemes that significantly differ from the IETF's congestion control principles. Using these new congestion control schemes in the global Internet has possible ramifications to both the traffic using the new congestion control and to traffic using the currently standardized congestion control. Therefore, the IETF must proceed with caution when dealing with alternate congestion control proposals. The goal<author fullname="David Wetherall" initials="D." surname="Wetherall"> <organization>University ofthis document is to provide guidance for considering alternate congestion control algorithms within the IETF. This document specifies an Internet Best Current Practices for the Internet Community, and requests discussion and suggestions for improvements.</t> </abstract>Washington</organization> </author> <author fullname="Tom Anderson" initials="T." surname="Anderson"> <organization>University of Washington</organization> </author> <date month="October" year="1999"/> </front> <seriesInfoname="BCP" value="133"/> <seriesInfo name="RFC" value="5033"/> <seriesInfoname="DOI"value="10.17487/RFC5033"/>value="10.1145/505696.505704"/> <refcontent>ACM SIGCOMM Computer Communication Review, Volume 29, Issue 5, pp. 71–78</refcontent> </reference> <referenceanchor="RFC5129" target="https://www.rfc-editor.org/info/rfc5129" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.5129.xml">anchor="Heist21" target="https://github.com/heistp/l4s-tests"> <front><title>Explicit Congestion Marking in MPLS</title> <author initials="B." surname="Davie" fullname="B. Davie"><title>L4S Tests</title> <author> <organization/> </author> <date month="August" year="2021"/> </front> <refcontent>commit e21cd91</refcontent> </reference> <reference anchor="SCReAM-L4S" target="https://github.com/EricssonResearch/scream"> <front> <title>SCReAM</title> <author/> <date month="November" year="2022"/> </front> <refcontent>commit 140e292</refcontent> </reference> <reference anchor="DualPI2Linux" target="https://www.netdevconf.org/0x13/session.html?talk-DUALPI2-AQM"> <front> <title>DUALPI2 - Low Latency, Low Loss and Scalable (L4S) AQM</title> <author fullname="Olga Albisser" initials="O." surname="Albisser"> <organization>Simula Research Lab</organization> </author> <author fullname="Koen De Schepper" initials="K." surname="De Schepper"> <organization>Nokia Bell Labs</organization> </author> <author fullname="Bob Briscoe" initials="B."surname="Briscoe" fullname="B. Briscoe"> <organization/>surname="Briscoe"> <organization>Independent</organization> </author> <authorinitials="J." surname="Tay" fullname="J. Tay"> <organization/>fullname="Olivier Tilmans" initials="O." surname="Tilmans"> <organization>Nokia Bell Labs</organization> </author> <author fullname="Henrik Steen" initials="H." surname="Steen"> <organization>Simula Research Lab</organization> </author> <dateyear="2008" month="January"/> <abstract> <t>RFC 3270 defines how to support the Diffserv architecture in MPLS networks, including how to encode Diffserv Code Points (DSCPs) in an MPLS header. DSCPs may be encoded in the EXP field, while other uses of that field are not precluded. RFC 3270 makes no statement about how Explicit Congestion Notification (ECN) marking might be encoded in the MPLS header. This document defines how an operator might define some of the EXP codepoints for explicit congestion notification, without precluding other uses. [STANDARDS-TRACK]</t> </abstract>month="March" year="2019"/> </front><seriesInfo name="RFC" value="5129"/> <seriesInfo name="DOI" value="10.17487/RFC5129"/><refcontent>Proceedings of Linux Netdev 0x13</refcontent> <format target="https://www.files.netdevconf.org/f/febbe8c6a05b4ceab641/?dl=1" type="PDF"/> </reference><!-- <?rfc include='reference.RFC.5238'?> <?rfc include='reference.RFC.5415'?> <?rfc include='reference.RFC.5764'?> <?rfc include='reference.RFC.6083'?> --><referenceanchor="RFC5348" target="https://www.rfc-editor.org/info/rfc5348" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.5348.xml">anchor="COBALT" target="https://ieeexplore.ieee.org/abstract/document/8847054"> <front><title>TCP Friendly Rate Control (TFRC): Protocol Specification</title><title>Design and Evaluation of COBALT Queue Discipline</title> <author fullname="Jendaipou Palmei" initials="J." surname="Palmei"> <organization/> </author> <author fullname="Shefali Gupta" initials="S."surname="Floyd" fullname="S. Floyd">surname="Gupta"> <organization/> </author> <authorinitials="M." surname="Handley" fullname="M. Handley">fullname="Pasquale Imputato" initials="P." surname="Imputato"> <organization/> </author> <author fullname="Jonathan Morton" initials="J."surname="Padhye" fullname="J. Padhye">surname="Morton"> <organization/> </author> <authorinitials="J." surname="Widmer" fullname="J. Widmer">fullname="Mohit P. Tahiliani" initials="M. P." surname="Tahiliani"> <organization/> </author> <author fullname="Stefan Avallone" initials="S." surname="Avallone"> <organization/> </author> <author fullname="Dave Täht" initials="D." surname="Täht"> <organization/> </author> <dateyear="2008" month="September"/> <abstract> <t>This document specifies TCP Friendly Rate Control (TFRC). TFRC is a congestion control mechanism for unicast flows operating in a best-effort Internet environment. It is reasonably fair when competing for bandwidth with TCP flows, but has a much lower variation of throughput over time compared with TCP, making it more suitable for applications such as streaming media where a relatively smooth sending rate is of importance.</t> <t>This document obsoletes RFC 3448 and updates RFC 4342. [STANDARDS-TRACK]</t> </abstract>month="July" year="2019"/> </front> <seriesInfoname="RFC" value="5348"/> <seriesInfoname="DOI"value="10.17487/RFC5348"/>value="10.1109/LANMAN.2019.8847054"/> <refcontent>IEEE International Symposium on Local and Metropolitan Area Networks (LANMAN)</refcontent> </reference> <referenceanchor="RFC5622" target="https://www.rfc-editor.org/info/rfc5622" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.5622.xml">anchor="Dukkipati06" target="https://dl.acm.org/doi/10.1145/1111322.1111336"> <front><title>Profile for Datagram Congestion Control Protocol (DCCP) Congestion ID 4: TCP-Friendly Rate Control for Small Packets (TFRC-SP)</title> <author initials="S." surname="Floyd" fullname="S. Floyd"> <organization/> </author> <author initials="E." surname="Kohler" fullname="E. Kohler"> <organization/> </author> <date year="2009" month="August"/> <abstract> <t>This document specifies a profile for Congestion Control Identifier 4, the small-packet variant of TCP-Friendly Rate Control (TFRC), in the Datagram Congestion Control Protocol (DCCP). CCID 4 is for experimental use, and uses TFRC-SP (RFC 4828), a variant of TFRC designed for applications that send small packets. CCID 4 is considered experimental because TFRC-SP is itself experimental, and is not proposed for widespread deployment in the global Internet at this time. The goal for TFRC-SP is to achieve roughly the same bandwidth in bits per second (bps) as a TCP flow using packets of up to 1500 bytes but experiencing the same level of congestion. CCID 4 is for use for senders that send small packets and would like a TCP- friendly sending rate, possibly with Explicit Congestion Notification (ECN), while minimizing abrupt rate changes. This memo defines an Experimental Protocol for the Internet community.</t> </abstract> </front> <seriesInfo name="RFC" value="5622"/> <seriesInfo name="DOI" value="10.17487/RFC5622"/> </reference> <reference anchor="RFC5865" target="https://www.rfc-editor.org/info/rfc5865" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.5865.xml"> <front> <title>A Differentiated Services Code Point (DSCP) for Capacity-Admitted Traffic</title> <author initials="F." surname="Baker" fullname="F. Baker"> <organization/> </author> <author initials="J." surname="Polk" fullname="J. Polk"> <organization/> </author> <author initials="M." surname="Dolly" fullname="M. Dolly"> <organization/> </author> <date year="2010" month="May"/> <abstract> <t>This document requests one Differentiated Services Code Point (DSCP) from the Internet Assigned Numbers Authority (IANA) for a class of real-time traffic. This traffic class conforms to the Expedited Forwarding Per-Hop Behavior. This traffic is also admitted by the network using a Call Admission Control (CAC) procedure involving authentication, authorization, and capacity admission. This differs from a real-time traffic class that conforms to the Expedited Forwarding Per-Hop Behavior but is not subject to capacity admission or subject to very coarse capacity admission. [STANDARDS-TRACK]</t> </abstract> </front> <seriesInfo name="RFC" value="5865"/> <seriesInfo name="DOI" value="10.17487/RFC5865"/> </reference> <reference anchor="RFC4960" target="https://www.rfc-editor.org/info/rfc4960" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.4960.xml"> <front> <title>Stream Control Transmission Protocol</title> <author initials="R." surname="Stewart" fullname="R. Stewart" role="editor"> <organization/> </author> <date year="2007" month="September"/> <abstract> <t>This document obsoletes RFC 2960 and RFC 3309. It describes the Stream Control Transmission Protocol (SCTP). SCTP is designed to transport Public Switched Telephone Network (PSTN) signaling messages over IP networks, but is capable of broader applications.</t> <t>SCTP is a reliable transport protocol operating on top of a connectionless packet network such as IP. It offers the following services to its users:</t> <t>-- acknowledged error-free non-duplicated transfer of user data,</t> <t>-- data fragmentation to conform to discovered path MTU size,</t> <t>-- sequenced delivery of user messages within multiple streams, with an option for order-of-arrival delivery of individual user messages,</t> <t>-- optional bundling of multiple user messages into a single SCTP packet, and</t> <t>-- network-level fault tolerance through supporting of multi-homing at either or both ends of an association.</t> <t> The design of SCTP includes appropriate congestion avoidance behavior and resistance to flooding and masquerade attacks. [STANDARDS-TRACK]</t> </abstract> </front> <seriesInfo name="RFC" value="4960"/> <seriesInfo name="DOI" value="10.17487/RFC4960"/> </reference> <reference anchor="RFC5562" target="https://www.rfc-editor.org/info/rfc5562" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.5562.xml"> <front> <title>Adding Explicit Congestion Notification (ECN) Capability to TCP's SYN/ACK Packets</title> <author initials="A." surname="Kuzmanovic" fullname="A. Kuzmanovic"> <organization/> </author> <author initials="A." surname="Mondal" fullname="A. Mondal"> <organization/> </author> <author initials="S." surname="Floyd" fullname="S. Floyd"> <organization/> </author> <author initials="K." surname="Ramakrishnan" fullname="K. Ramakrishnan"> <organization/> </author> <date year="2009" month="June"/> <abstract> <t>The proposal in this document is Experimental. While it may be deployed in the current Internet, it does not represent a consensus that this is the best possible mechanism for the use of Explicit Congestion Notification (ECN) in TCP SYN/ACK packets.</t> <t>This document describes an optional, experimental modification to RFC 3168 to allow TCP SYN/ACK packets to be ECN-Capable. For TCP, RFC 3168 specifies setting an ECN-Capable codepoint on data packets, but not on SYN and SYN/ACK packets. However, because of the high cost to the TCP transfer of having a SYN/ACK packet dropped, with the resulting retransmission timeout, this document describes the use of ECN for the SYN/ACK packet itself, when sent in response to a SYN packet with the two ECN flags set in the TCP header, indicating a willingness to use ECN. Setting the initial TCP SYN/ACK packet as ECN-Capable can be of great benefit to the TCP connection, avoiding the severe penalty of a retransmission timeout for a connection that has not yet started placing a load on the network. The TCP responder (the sender of the SYN/ACK packet) must reply to a report of an ECN-marked SYN/ACK packet by resending a SYN/ACK packet that is not ECN-Capable. If the resent SYN/ACK packet is acknowledged, then the TCP responder reduces its initial congestion window from two, three, or four segments to one segment, thereby reducing the subsequent load from that connection on the network. If instead the SYN/ACK packet is dropped, or for some other reason the TCP responder does not receive an acknowledgement in the specified time, the TCP responder follows TCP standards for a dropped SYN/ACK packet (setting the retransmission timer). This memo defines an Experimental Protocol for the Internet community.</t> </abstract> </front> <seriesInfo name="RFC" value="5562"/> <seriesInfo name="DOI" value="10.17487/RFC5562"/> </reference> <reference anchor="RFC5681" target="https://www.rfc-editor.org/info/rfc5681" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.5681.xml"> <front> <title>TCP Congestion Control</title> <author initials="M." surname="Allman" fullname="M. Allman"> <organization/> </author> <author initials="V." surname="Paxson" fullname="V. Paxson"> <organization/> </author> <author initials="E." surname="Blanton" fullname="E. Blanton"> <organization/> </author> <date year="2009" month="September"/> <abstract> <t>This document defines TCP's four intertwined congestion control algorithms: slow start, congestion avoidance, fast retransmit, and fast recovery. In addition, the document specifies how TCP should begin transmission after a relatively long idle period, as well as discussing various acknowledgment generation methods. This document obsoletes RFC 2581. [STANDARDS-TRACK]</t> </abstract> </front> <seriesInfo name="RFC" value="5681"/> <seriesInfo name="DOI" value="10.17487/RFC5681"/> </reference> <reference anchor="RFC5706" target="https://www.rfc-editor.org/info/rfc5706" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.5706.xml"> <front> <title>Guidelines for Considering Operations and Management of New Protocols and Protocol Extensions</title> <author initials="D." surname="Harrington" fullname="D. Harrington"> <organization/> </author> <date year="2009" month="November"/> <abstract> <t>New protocols or protocol extensions are best designed with due consideration of the functionality needed to operate and manage the protocols. Retrofitting operations and management is sub-optimal. The purpose of this document is to provide guidance to authors and reviewers of documents that define new protocols or protocol extensions regarding aspects of operations and management that should be considered. This memo provides information for the Internet community.</t> </abstract> </front> <seriesInfo name="RFC" value="5706"/> <seriesInfo name="DOI" value="10.17487/RFC5706"/> </reference> <reference anchor="RFC6040" target="https://www.rfc-editor.org/info/rfc6040" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.6040.xml"> <front> <title>Tunnelling of Explicit Congestion Notification</title> <author initials="B." surname="Briscoe" fullname="B. Briscoe"> <organization/> </author> <date year="2010" month="November"/> <abstract> <t>This document redefines how the explicit congestion notification (ECN) field of the IP header should be constructed on entry to and exit from any IP-in-IP tunnel. On encapsulation, it updates RFC 3168 to bring all IP-in-IP tunnels (v4 or v6) into line with RFC 4301 IPsec ECN processing. On decapsulation, it updates both RFC 3168 and RFC 4301 to add new behaviours for previously unused combinations of inner and outer headers. The new rules ensure the ECN field is correctly propagated across a tunnel whether it is used to signal one or two severity levels of congestion; whereas before, only one severity level was supported. Tunnel endpoints can be updated in any order without affecting pre-existing uses of the ECN field, thus ensuring backward compatibility. Nonetheless, operators wanting to support two severity levels (e.g., for pre-congestion notification -- PCN) can require compliance with this new specification. A thorough analysis of the reasoning for these changes and the implications is included. In the unlikely event that the new rules do not meet a specific need, RFC 4774 gives guidance on designing alternate ECN semantics, and this document extends that to include tunnelling issues. [STANDARDS-TRACK]</t> </abstract> </front> <seriesInfo name="RFC" value="6040"/> <seriesInfo name="DOI" value="10.17487/RFC6040"/> </reference> <reference anchor="RFC6077" target="https://www.rfc-editor.org/info/rfc6077" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.6077.xml"> <front> <title>Open Research Issues in Internet Congestion Control</title> <author initials="D." surname="Papadimitriou" fullname="D. Papadimitriou" role="editor"> <organization/> </author> <author initials="M." surname="Welzl" fullname="M. Welzl"> <organization/> </author> <author initials="M." surname="Scharf" fullname="M. Scharf"> <organization/> </author> <author initials="B." surname="Briscoe" fullname="B. Briscoe"> <organization/> </author> <date year="2011" month="February"/> <abstract> <t>This document describes some of the open problems in Internet congestion control that are known today. This includes several new challenges that are becoming important as the network grows, as well as some issues that have been known for many years. These challenges are generally considered to be open research topics that may require more study or application of innovative techniques before Internet-scale solutions can be confidently engineered and deployed. This document is not an Internet Standards Track specification; it is published for informational purposes.</t> </abstract> </front> <seriesInfo name="RFC" value="6077"/> <seriesInfo name="DOI" value="10.17487/RFC6077"/> </reference> <reference anchor="RFC6660" target="https://www.rfc-editor.org/info/rfc6660" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.6660.xml"> <front> <title>Encoding Three Pre-Congestion Notification (PCN) States in the IP Header Using a Single Diffserv Codepoint (DSCP)</title> <author initials="B." surname="Briscoe" fullname="B. Briscoe"> <organization/> </author> <author initials="T." surname="Moncaster" fullname="T. Moncaster"> <organization/> </author> <author initials="M." surname="Menth" fullname="M. Menth"> <organization/> </author> <date year="2012" month="July"/> <abstract> <t>The objective of Pre-Congestion Notification (PCN) is to protect the quality of service (QoS) of inelastic flows within a Diffserv domain. The overall rate of PCN-traffic is metered on every link in the PCN- domain, and PCN-packets are appropriately marked when certain configured rates are exceeded. Egress nodes pass information about these PCN-marks to Decision Points that then decide whether to admit or block new flow requests or to terminate some already admitted flows during serious pre-congestion.</t> <t>This document specifies how PCN-marks are to be encoded into the IP header by reusing the Explicit Congestion Notification (ECN) codepoints within a PCN-domain. The PCN wire protocol for non-IP protocol headers will need to be defined elsewhere. Nonetheless, this document clarifies the PCN encoding for MPLS in an informational appendix. The encoding for IP provides for up to three different PCN marking states using a single Diffserv codepoint (DSCP): not-marked (NM), threshold-marked (ThM), and excess-traffic-marked (ETM). Hence, it is called the 3-in-1 PCN encoding. This document obsoletes RFC 5696. [STANDARDS-TRACK]</t> </abstract> </front> <seriesInfo name="RFC" value="6660"/> <seriesInfo name="DOI" value="10.17487/RFC6660"/> </reference> <reference anchor="RFC6675" target="https://www.rfc-editor.org/info/rfc6675" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.6675.xml"> <front> <title>A Conservative Loss Recovery Algorithm Based on Selective Acknowledgment (SACK) for TCP</title> <author initials="E." surname="Blanton" fullname="E. Blanton"> <organization/> </author> <author initials="M." surname="Allman" fullname="M. Allman"> <organization/> </author> <author initials="L." surname="Wang" fullname="L. Wang"> <organization/> </author> <author initials="I." surname="Jarvinen" fullname="I. Jarvinen"> <organization/> </author> <author initials="M." surname="Kojo" fullname="M. Kojo"> <organization/> </author> <author initials="Y." surname="Nishida" fullname="Y. Nishida"> <organization/> </author> <date year="2012" month="August"/> <abstract> <t>This document presents a conservative loss recovery algorithm for TCP that is based on the use of the selective acknowledgment (SACK) TCP option. The algorithm presented in this document conforms to the spirit of the current congestion control specification (RFC 5681), but allows TCP senders to recover more effectively when multiple segments are lost from a single flight of data. This document obsoletes RFC 3517 and describes changes from it. [STANDARDS-TRACK]</t> </abstract> </front> <seriesInfo name="RFC" value="6675"/> <seriesInfo name="DOI" value="10.17487/RFC6675"/> </reference> <reference anchor="RFC7560" target="https://www.rfc-editor.org/info/rfc7560" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.7560.xml"> <front> <title>Problem Statement and Requirements for Increased Accuracy in Explicit Congestion Notification (ECN) Feedback</title> <author initials="M." surname="Kuehlewind" fullname="M. Kuehlewind" role="editor"> <organization/> </author> <author initials="R." surname="Scheffenegger" fullname="R. Scheffenegger"> <organization/> </author> <author initials="B." surname="Briscoe" fullname="B. Briscoe"> <organization/> </author> <date year="2015" month="August"/> <abstract> <t>Explicit Congestion Notification (ECN) is a mechanism where network nodes can mark IP packets, instead of dropping them, to indicate congestion to the endpoints. An ECN-capable receiver will feed this information back to the sender. ECN is specified for TCP in such a way that it can only feed back one congestion signal per Round-Trip Time (RTT). In contrast, ECN for other transport protocols, such as RTP/UDP and SCTP, is specified with more accurate ECN feedback. Recent new TCP mechanisms (like Congestion Exposure (ConEx) or Data Center TCP (DCTCP)) need more accurate ECN feedback in the case where more than one marking is received in one RTT. This document specifies requirements for an update to the TCP protocol to provide more accurate ECN feedback.</t> </abstract> </front> <seriesInfo name="RFC" value="7560"/> <seriesInfo name="DOI" value="10.17487/RFC7560"/> </reference> <reference anchor="RFC7567" target="https://www.rfc-editor.org/info/rfc7567" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.7567.xml"> <front> <title>IETF Recommendations Regarding Active Queue Management</title> <author initials="F." surname="Baker" fullname="F. Baker" role="editor"> <organization/> </author> <author initials="G." surname="Fairhurst" fullname="G. Fairhurst" role="editor"> <organization/> </author> <date year="2015" month="July"/> <abstract> <t>This memo presents recommendations to the Internet community concerning measures to improve and preserve Internet performance. It presents a strong recommendation for testing, standardization, and widespread deployment of active queue management (AQM) in network devices to improve the performance of today's Internet. It also urges a concerted effort of research, measurement, and ultimate deployment of AQM mechanisms to protect the Internet from flows that are not sufficiently responsive to congestion notification.</t> <t>Based on 15 years of experience and new research, this document replaces the recommendations of RFC 2309.</t> </abstract> </front> <seriesInfo name="BCP" value="197"/> <seriesInfo name="RFC" value="7567"/> <seriesInfo name="DOI" value="10.17487/RFC7567"/> </reference> <reference anchor="RFC7713" target="https://www.rfc-editor.org/info/rfc7713" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.7713.xml"> <front> <title>Congestion Exposure (ConEx) Concepts, Abstract Mechanism, and Requirements</title> <author initials="M." surname="Mathis" fullname="M. Mathis"> <organization/> </author> <author initials="B." surname="Briscoe" fullname="B. Briscoe"> <organization/> </author> <date year="2015" month="December"/> <abstract> <t>This document describes an abstract mechanism by which senders inform the network about the congestion recently encountered by packets in the same flow. Today, network elements at any layer may signal congestion to the receiver by dropping packets or by Explicit Congestion Notification (ECN) markings, and the receiver passes this information back to the sender in transport-layer feedback. The mechanism described here enables the sender to also relay this congestion information back into the network in-band at the IP layer, such that the total amount of congestion from all elements on the path is revealed to all IP elements along the path, where it could, for example, be used to provide input to traffic management. This mechanism is called Congestion Exposure, or ConEx. The companion document, "Congestion Exposure (ConEx) Concepts and Use Cases" (RFC 6789), provides the entry point to the set of ConEx documentation.</t> </abstract> </front> <seriesInfo name="RFC" value="7713"/> <seriesInfo name="DOI" value="10.17487/RFC7713"/> </reference> <reference anchor="RFC5925" target="https://www.rfc-editor.org/info/rfc5925" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.5925.xml"> <front> <title>The TCP Authentication Option</title> <author initials="J." surname="Touch" fullname="J. Touch"> <organization/> </author> <author initials="A." surname="Mankin" fullname="A. Mankin"> <organization/> </author> <author initials="R." surname="Bonica" fullname="R. Bonica"> <organization/> </author> <date year="2010" month="June"/> <abstract> <t>This document specifies the TCP Authentication Option (TCP-AO), which obsoletes the TCP MD5 Signature option of RFC 2385 (TCP MD5). TCP-AO specifies the use of stronger Message Authentication Codes (MACs), protects against replays even for long-lived TCP connections, and provides more details on the association of security with TCP connections than TCP MD5. TCP-AO is compatible with either a static Master Key Tuple (MKT) configuration or an external, out-of-band MKT management mechanism; in either case, TCP-AO also protects connections when using the same MKT across repeated instances of a connection, using traffic keys derived from the MKT, and coordinates MKT changes between endpoints. The result is intended to support current infrastructure uses of TCP MD5, such as to protect long-lived connections (as used, e.g., in BGP and LDP), and to support a larger set of MACs with minimal other system and operational changes. TCP-AO uses a different option identifier than TCP MD5, even though TCP-AO and TCP MD5 are never permitted to be used simultaneously. TCP-AO supports IPv6, and is fully compatible with the proposed requirements for the replacement of TCP MD5. [STANDARDS-TRACK]</t> </abstract> </front> <seriesInfo name="RFC" value="5925"/> <seriesInfo name="DOI" value="10.17487/RFC5925"/> </reference> <reference anchor="RFC8033" target="https://www.rfc-editor.org/info/rfc8033" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8033.xml"> <front> <title>Proportional Integral Controller Enhanced (PIE): A Lightweight Control Scheme to Address the Bufferbloat Problem</title> <author initials="R." surname="Pan" fullname="R. Pan"> <organization/> </author> <author initials="P." surname="Natarajan" fullname="P. Natarajan"> <organization/> </author> <author initials="F." surname="Baker" fullname="F. Baker"> <organization/> </author> <author initials="G." surname="White" fullname="G. White"> <organization/> </author> <date year="2017" month="February"/> <abstract> <t>Bufferbloat is a phenomenon in which excess buffers in the network cause high latency and latency variation. As more and more interactive applications (e.g., voice over IP, real-time video streaming, and financial transactions) run in the Internet, high latency and latency variation degrade application performance. There is a pressing need to design intelligent queue management schemes that can control latency and latency variation, and hence provide desirable quality of service to users.</t> <t>This document presents a lightweight active queue management design called "PIE" (Proportional Integral controller Enhanced) that can effectively control the average queuing latency to a target value. Simulation results, theoretical analysis, and Linux testbed results have shown that PIE can ensure low latency and achieve high link utilization under various congestion situations. The design does not require per-packet timestamps, so it incurs very little overhead and is simple enough to implement in both hardware and software.</t> </abstract> </front> <seriesInfo name="RFC" value="8033"/> <seriesInfo name="DOI" value="10.17487/RFC8033"/> </reference> <reference anchor="RFC8083" target="https://www.rfc-editor.org/info/rfc8083" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8083.xml"> <front> <title>Multimedia Congestion Control: Circuit Breakers for Unicast RTP Sessions</title> <author initials="C." surname="Perkins" fullname="C. Perkins"> <organization/> </author> <author initials="V." surname="Singh" fullname="V. Singh"> <organization/> </author> <date year="2017" month="March"/> <abstract> <t>The Real-time Transport Protocol (RTP) is widely used in telephony, video conferencing, and telepresence applications. Such applications are often run on best-effort UDP/IP networks. If congestion control is not implemented in these applications, then network congestion can lead to uncontrolled packet loss and a resulting deterioration of the user's multimedia experience. The congestion control algorithm acts as a safety measure by stopping RTP flows from using excessive resources and protecting the network from overload. At the time of this writing, however, while there are several proprietary solutions, there is no standard algorithm for congestion control of interactive RTP flows.</t> <t>This document does not propose a congestion control algorithm. It instead defines a minimal set of RTP circuit breakers: conditions under which an RTP sender needs to stop transmitting media data to protect the network from excessive congestion. It is expected that, in the absence of long-lived excessive congestion, RTP applications running on best-effort IP networks will be able to operate without triggering these circuit breakers. To avoid triggering the RTP circuit breaker, any Standards Track congestion control algorithms defined for RTP will need to operate within the envelope set by these RTP circuit breaker algorithms.</t> </abstract> </front> <seriesInfo name="RFC" value="8083"/> <seriesInfo name="DOI" value="10.17487/RFC8083"/> </reference> <reference anchor="RFC8085" target="https://www.rfc-editor.org/info/rfc8085" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8085.xml"> <front> <title>UDP Usage Guidelines</title> <author initials="L." surname="Eggert" fullname="L. Eggert"> <organization/> </author> <author initials="G." surname="Fairhurst" fullname="G. Fairhurst"> <organization/> </author> <author initials="G." surname="Shepherd" fullname="G. Shepherd"> <organization/> </author> <date year="2017" month="March"/> <abstract> <t>The User Datagram Protocol (UDP) provides a minimal message-passing transport that has no inherent congestion control mechanisms. This document provides guidelines on the use of UDP for the designers of applications, tunnels, and other protocols that use UDP. Congestion control guidelines are a primary focus, but the document also provides guidance on other topics, including message sizes, reliability, checksums, middlebox traversal, the use of Explicit Congestion Notification (ECN), Differentiated Services Code Points (DSCPs), and ports.</t> <t>Because congestion control is critical to the stable operation of the Internet, applications and other protocols that choose to use UDP as an Internet transport must employ mechanisms to prevent congestion collapse and to establish some degree of fairness with concurrent traffic. They may also need to implement additional mechanisms, depending on how they use UDP.</t> <t>Some guidance is also applicable to the design of other protocols (e.g., protocols layered directly on IP or via IP-based tunnels), especially when these protocols do not themselves provide congestion control.</t> <t>This document obsoletes RFC 5405 and adds guidelines for multicast UDP usage.</t> </abstract> </front> <seriesInfo name="BCP" value="145"/> <seriesInfo name="RFC" value="8085"/> <seriesInfo name="DOI" value="10.17487/RFC8085"/> </reference> <reference anchor="RFC8174" target="https://www.rfc-editor.org/info/rfc8174" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8174.xml"> <front> <title>Ambiguity of Uppercase vs Lowercase in RFC 2119 Key Words</title> <author initials="B." surname="Leiba" fullname="B. Leiba"> <organization/> </author> <date year="2017" month="May"/> <abstract> <t>RFC 2119 specifies common key words that may be used in protocol specifications. This document aims to reduce the ambiguity by clarifying that only UPPERCASE usage of the key words have the defined special meanings.</t> </abstract> </front> <seriesInfo name="BCP" value="14"/> <seriesInfo name="RFC" value="8174"/> <seriesInfo name="DOI" value="10.17487/RFC8174"/> </reference> <reference anchor="RFC8290" target="https://www.rfc-editor.org/info/rfc8290" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8290.xml"> <front> <title>The Flow Queue CoDel Packet Scheduler and Active Queue Management Algorithm</title> <author initials="T." surname="Hoeiland-Joergensen" fullname="T. Hoeiland-Joergensen"> <organization/> </author> <author initials="P." surname="McKenney" fullname="P. McKenney"> <organization/> </author> <author initials="D." surname="Taht" fullname="D. Taht"> <organization/> </author> <author initials="J." surname="Gettys" fullname="J. Gettys"> <organization/> </author> <author initials="E." surname="Dumazet" fullname="E. Dumazet"> <organization/> </author> <date year="2018" month="January"/> <abstract> <t>This memo presents the FQ-CoDel hybrid packet scheduler and Active Queue Management (AQM) algorithm, a powerful tool for fighting bufferbloat and reducing latency.</t> <t>FQ-CoDel mixes packets from multiple flows and reduces the impact of head-of-line blocking from bursty traffic. It provides isolation for low-rate traffic such as DNS, web, and videoconferencing traffic. It improves utilisation across the networking fabric, especially for bidirectional traffic, by keeping queue lengths short, and it can be implemented in a memory- and CPU-efficient fashion across a wide range of hardware.</t> </abstract> </front> <seriesInfo name="RFC" value="8290"/> <seriesInfo name="DOI" value="10.17487/RFC8290"/> </reference> <reference anchor="RFC8298" target="https://www.rfc-editor.org/info/rfc8298" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8298.xml"> <front> <title>Self-Clocked Rate Adaptation for Multimedia</title> <author initials="I." surname="Johansson" fullname="I. Johansson"> <organization/> </author> <author initials="Z." surname="Sarker" fullname="Z. Sarker"> <organization/> </author> <date year="2017" month="December"/> <abstract> <t>This memo describes a rate adaptation algorithm for conversational media services such as interactive video. The solution conforms to the packet conservation principle and uses a hybrid loss-and-delay- based congestion control algorithm. The algorithm is evaluated over both simulated Internet bottleneck scenarios as well as in a Long Term Evolution (LTE) system simulator and is shown to achieve both low latency and high video throughput in these scenarios.</t> </abstract> </front> <seriesInfo name="RFC" value="8298"/> <seriesInfo name="DOI" value="10.17487/RFC8298"/> </reference> <reference anchor="I-D.ietf-tcpm-accurate-ecn" target="https://datatracker.ietf.org/api/v1/doc/document/draft-ietf-tcpm-accurate-ecn/" xml:base="https://bib.ietf.org/public/rfc/bibxml-ids/reference.I-D.ietf-tcpm-accurate-ecn.xml"> <front> <title>More Accurate ECN Feedback in TCP</title> <author fullname="Bob Briscoe"/> <author fullname="Mirja Kühlewind"/> <author fullname="Richard Scheffenegger"/> <date day="25" month="July" year="2022"/> <abstract> <t>Explicit Congestion Notification (ECN) is a mechanism where network nodes can mark IP packets instead of dropping them to indicate incipient congestion to the end-points. Receivers with an ECN- capable transport protocol feed back this information to the sender. ECN was originally specified for TCP in such a way that only one feedback signal can be transmitted per Round-Trip Time (RTT). Recent new TCP mechanisms like Congestion Exposure (ConEx), Data Center TCP (DCTCP) or Low Latency Low Loss Scalable Throughput (L4S) need more accurate ECN feedback information whenever more than one marking is received in one RTT. This document updates the original ECN specification to specify a scheme to provide more than one feedback signal per RTT in the TCP header. Given TCP header space is scarce, it allocates a reserved header bit previously assigned to the ECN- Nonce. It also overloads the two existing ECN flags in the TCP header. The resulting extra space is exploited to feed back the IP- ECN field received during the 3-way handshake as well. Supplementary feedback information can optionally be provided in a new TCP option, which is never used on the TCP SYN. The document also specifies the treatment of this updated TCP wire protocol by middleboxes.</t> </abstract> </front> <seriesInfo name="Internet-Draft" value="draft-ietf-tcpm-accurate-ecn-20"/> </reference> <reference anchor="I-D.ietf-tsvwg-ecn-encap-guidelines" target="https://datatracker.ietf.org/api/v1/doc/document/draft-ietf-tsvwg-ecn-encap-guidelines/" xml:base="https://bib.ietf.org/public/rfc/bibxml-ids/reference.I-D.ietf-tsvwg-ecn-encap-guidelines.xml"> <front> <title>Guidelines for Adding Congestion Notification to Protocols that Encapsulate IP</title> <author fullname="Bob Briscoe"/> <author fullname="John Kaippallimalil"/> <date day="11" month="July" year="2022"/> <abstract> <t>The purpose of this document is to guide the design of congestion notification in any lower layer or tunnelling protocol that encapsulates IP. The aim is for explicit congestion signals to propagate consistently from lower layer protocols into IP. Then the IP internetwork layer can act as a portability layer to carry congestion notification from non-IP-aware congested nodes up to the transport layer (L4). Following these guidelines should assure interworking among IP layer and lower layer congestion notification mechanisms, whether specified by the IETF or other standards bodies. This document updates the advice to subnetwork designers about ECN in RFC 3819.</t> </abstract> </front> <seriesInfo name="Internet-Draft" value="draft-ietf-tsvwg-ecn-encap-guidelines-17"/> </reference> <reference anchor="I-D.ietf-tsvwg-rfc6040update-shim" target="https://datatracker.ietf.org/api/v1/doc/document/draft-ietf-tsvwg-rfc6040update-shim/" xml:base="https://bib.ietf.org/public/rfc/bibxml-ids/reference.I-D.ietf-tsvwg-rfc6040update-shim.xml"> <front> <title>Propagating Explicit Congestion Notification Across IP Tunnel Headers Separated by a Shim</title> <author fullname="Bob Briscoe"/> <date day="11" month="July" year="2022"/> <abstract> <t>RFC 6040 on "Tunnelling of Explicit Congestion Notification" made the rules for propagation of ECN consistent for all forms of IP in IP tunnel. This specification updates RFC 6040 to clarify that its scope includes tunnels where two IP headers are separated by at least one shim header that is not sufficient on its own for wide area packet forwarding. It surveys widely deployed IP tunnelling protocols that use such shim header(s) and updates the specifications of those that do not mention ECN propagation (L2TPv2, L2TPv3, GRE, Teredo and AMT). This specification also updates RFC 6040 with configuration requirements needed to make any legacy tunnel ingress safe.</t> </abstract> </front> <seriesInfo name="Internet-Draft" value="draft-ietf-tsvwg-rfc6040update-shim-15"/> </reference> <reference anchor="I-D.ietf-trill-ecn-support" target="https://datatracker.ietf.org/api/v1/doc/document/draft-ietf-trill-ecn-support/" xml:base="https://bib.ietf.org/public/rfc/bibxml-ids/reference.I-D.ietf-trill-ecn-support.xml"> <front> <title>TRILL (TRansparent Interconnection of Lots of Links): ECN (Explicit Congestion Notification) Support</title> <author fullname="Donald E. Eastlake"/> <author fullname="Bob Briscoe"/> <date day="25" month="February" year="2018"/> <abstract> <t>Explicit congestion notification (ECN) allows a forwarding element to notify downstream devices, including the destination, of the onset of congestion without having to drop packets. This can improve network efficiency through better congestion control without packet drops. This document extends ECN to TRILL (TRansparent Interconnection of Lots of Links) switches, including integration with IP ECN, and provides for ECN marking in the TRILL Header Extension Flags Word (see RFC 7179).</t> </abstract> </front> <seriesInfo name="Internet-Draft" value="draft-ietf-trill-ecn-support-07"/> </reference> <reference anchor="I-D.ietf-tsvwg-aqm-dualq-coupled" target="https://datatracker.ietf.org/api/v1/doc/document/draft-ietf-tsvwg-aqm-dualq-coupled/" xml:base="https://bib.ietf.org/public/rfc/bibxml-ids/reference.I-D.ietf-tsvwg-aqm-dualq-coupled.xml"> <front> <title>DualQ Coupled AQMs for Low Latency, Low Loss and Scalable Throughput (L4S)</title> <author fullname="Koen De Schepper"/> <author fullname="Bob Briscoe"/> <author fullname="Greg White"/> <date day="7" month="July" year="2022"/> <abstract> <t>This specification defines a framework for coupling the Active Queue Management (AQM) algorithms in two queues intended for flows with different responses to congestion. This provides a way for the Internet to transition from the scaling problems of standard TCP Reno-friendly ('Classic') congestion controls to the family of 'Scalable' congestion controls. These are designed for consistently very Low queuing Latency, very Low congestion Loss and Scaling of per-flow throughput (L4S) by using Explicit Congestion Notification (ECN) in a modified way. Until the Coupled DualQ, these L4S senders could only be deployed where a clean-slate environment could be arranged, such as in private data centres. The coupling acts like a semi-permeable membrane: isolating the sub-millisecond average queuing delay and zero congestion loss of L4S from Classic latency and loss; but pooling the capacity between any combination of Scalable and Classic flows with roughly equivalent throughput per flow. The DualQ achieves this indirectly, without having to inspect transport layer flow identifiers and without compromising the performance of the Classic traffic, relative to a single queue. The DualQ design has low complexity and requires no configuration for the public Internet.</t> </abstract> </front> <seriesInfo name="Internet-Draft" value="draft-ietf-tsvwg-aqm-dualq-coupled-24"/> </reference> <reference anchor="I-D.stewart-tsvwg-sctpecn" target="https://www.ietf.org/archive/id/draft-stewart-tsvwg-sctpecn-05.txt" xml:base="https://bib.ietf.org/public/rfc/bibxml-ids/reference.I-D.stewart-tsvwg-sctpecn.xml"> <front> <title>ECN for Stream Control Transmission Protocol (SCTP)</title> <author fullname="Randall R. Stewart"/> <author fullname="Michael Tuexen"/> <author fullname="Xuesong Dong"/> <date day="15" month="January" year="2014"/> <abstract> <t>This document describes the addition of the ECN to the Stream Control Transmission Protocol (SCTP).</t> </abstract> </front> <seriesInfo name="Internet-Draft" value="draft-stewart-tsvwg-sctpecn-05"/> </reference> <reference anchor="I-D.ietf-tsvwg-l4s-arch" target="https://datatracker.ietf.org/api/v1/doc/document/draft-ietf-tsvwg-l4s-arch/" xml:base="https://bib.ietf.org/public/rfc/bibxml-ids/reference.I-D.ietf-tsvwg-l4s-arch.xml"> <front> <title>Low Latency, Low Loss, Scalable Throughput (L4S) Internet Service: Architecture</title> <author fullname="Bob Briscoe"/> <author fullname="Koen De Schepper"/> <author fullname="Marcelo Bagnulo"/> <author fullname="Greg White"/> <date day="27" month="July" year="2022"/> <abstract> <t>This document describes the L4S architecture, which enables Internet applications to achieve Low queuing Latency, Low Loss, and Scalable throughput (L4S). The insight on which L4S is based is that the root cause of queuing delay is in the congestion controllers of senders, not in the queue itself. With the L4S architecture all Internet applications could (but do not have to) transition away from congestion control algorithms that cause substantial queuing delay, to a new class of congestion controls that induce very little queuing, aided by explicit congestion signalling from the network. This new class of congestion controls can provide low latency for capacity-seeking flows, so applications can achieve both high bandwidth and low latency.</t> <t>The architecture primarily concerns incremental deployment. It defines mechanisms that allow the new class of L4S congestion controls to coexist with 'Classic' congestion controls in a shared network. These mechanisms aim to ensure that the latency and throughput performance using an L4S-compliant congestion controller is usually much better (and rarely worse) than performance would have been using a 'Classic' congestion controller, and that competing flows continuing to use 'Classic' controllers are typically not impacted by the presence of L4S. These characteristics are important to encourage adoption of L4S congestion control algorithms and L4S compliant network elements.</t> <t>The L4S architecture consists of three components: network support to isolate L4S traffic from classic traffic; protocol features that allow network elements to identify L4S traffic; and host support for L4S congestion controls. The protocol is defined separately as an experimental change to Explicit Congestion Notification (ECN).</t> </abstract> </front> <seriesInfo name="Internet-Draft" value="draft-ietf-tsvwg-l4s-arch-19"/> </reference> <reference anchor="I-D.briscoe-tsvwg-l4s-diffserv" target="https://datatracker.ietf.org/api/v1/doc/document/draft-briscoe-tsvwg-l4s-diffserv/" xml:base="https://bib.ietf.org/public/rfc/bibxml-ids/reference.I-D.briscoe-tsvwg-l4s-diffserv.xml"> <front> <title>Interactions between Low Latency, Low Loss, Scalable Throughput (L4S) and Differentiated Services</title> <author fullname="Bob Briscoe"/> <date day="2" month="July" year="2018"/> <abstract> <t>L4S and Diffserv offer somewhat overlapping services (low latency and low loss), but bandwidth allocation is out of scope for L4S. Therefore there is scope for the two approaches to complement each other, but also to conflict. This informational document explains how the two approaches interact, how they can be arranged to complement each other and in which cases one can stand alone without needing the other.</t> </abstract> </front> <seriesInfo name="Internet-Draft" value="draft-briscoe-tsvwg-l4s-diffserv-02"/> </reference> <reference anchor="I-D.ietf-tsvwg-nqb" target="https://datatracker.ietf.org/api/v1/doc/document/draft-ietf-tsvwg-nqb/" xml:base="https://bib.ietf.org/public/rfc/bibxml-ids/reference.I-D.ietf-tsvwg-nqb.xml"> <front> <title>A Non-Queue-Building Per-Hop Behavior (NQB PHB) for Differentiated Services</title> <author fullname="Greg White"/> <author fullname="Thomas Fossati"/> <date day="4" month="March" year="2022"/> <abstract> <t>This document specifies properties and characteristics of a Non- Queue-Building Per-Hop Behavior (NQB PHB). The purpose of this NQB PHB is to provide a separate queue that enables smooth, low-data- rate, application-limited traffic flows, which would ordinarily share a queue with bursty and capacity-seeking traffic, to avoid the latency, latency variation and loss caused by such traffic. This PHB is implemented without prioritization and without rate policing, making it suitable for environments where the use of either these features may be restricted. The NQB PHB has been developed primarily for use by access network segments, where queuing delays and queuing loss caused by Queue-Building protocols are manifested, but its use is not limited to such segments. In particular, applications to cable broadband links, Wi-Fi links, and mobile network radio and core segments are discussed. This document recommends a specific Differentiated Services Code Point (DSCP) to identify Non-Queue- Building flows.</t> </abstract> </front> <seriesInfo name="Internet-Draft" value="draft-ietf-tsvwg-nqb-10"/> </reference> <reference anchor="I-D.ietf-tsvwg-l4sops" target="https://datatracker.ietf.org/api/v1/doc/document/draft-ietf-tsvwg-l4sops/" xml:base="https://bib.ietf.org/public/rfc/bibxml-ids/reference.I-D.ietf-tsvwg-l4sops.xml"> <front> <title>Operational Guidance for Deployment of L4S in the Internet</title> <author fullname="Greg White"/> <date day="28" month="April" year="2022"/> <abstract> <t>This document is intended to provide guidance in order to ensure successful deployment of Low Latency Low Loss Scalable throughput (L4S) in the Internet. Other L4S documents provide guidance for running an L4S experiment, but this document is focused solely on potential interactions between L4S flows and flows using the original ('Classic') ECN over a Classic ECN bottleneck link. The document discusses the potential outcomes of these interactions, describes mechanisms to detect the presence of Classic ECN bottlenecks, and identifies opportunities to prevent and/or detect and resolve fairness problems in such networks. This guidance is aimed at operators of end-systems, operators of networks, and researchers.</t> </abstract> </front> <seriesInfo name="Internet-Draft" value="draft-ietf-tsvwg-l4sops-03"/> </reference> <reference anchor="RFC8311" target="https://www.rfc-editor.org/info/rfc8311" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8311.xml"> <front> <title>Relaxing Restrictions on Explicit Congestion Notification (ECN) Experimentation</title> <author initials="D." surname="Black" fullname="D. Black"> <organization/> </author> <date year="2018" month="January"/> <abstract> <t>This memo updates RFC 3168, which specifies Explicit Congestion Notification (ECN) as an alternative to packet drops for indicating network congestion to endpoints. It relaxes restrictions in RFC 3168 that hinder experimentation towards benefits beyond just removal of loss. This memo summarizes the anticipated areas of experimentation and updates RFC 3168 to enable experimentation in these areas. An Experimental RFC in the IETF document stream is required to take advantage of any of these enabling updates. In addition, this memo makes related updates to the ECN specifications for RTP in RFC 6679 and for the Datagram Congestion Control Protocol (DCCP) in RFCs 4341, 4342, and 5622. This memo also records the conclusion of the ECN nonce experiment in RFC 3540 and provides the rationale for reclassification of RFC 3540 from Experimental to Historic; this reclassification enables new experimental use of the ECT(1) codepoint.</t> </abstract> </front> <seriesInfo name="RFC" value="8311"/> <seriesInfo name="DOI" value="10.17487/RFC8311"/> </reference> <reference anchor="I-D.ietf-tcpm-generalized-ecn" target="https://datatracker.ietf.org/api/v1/doc/document/draft-ietf-tcpm-generalized-ecn/" xml:base="https://bib.ietf.org/public/rfc/bibxml-ids/reference.I-D.ietf-tcpm-generalized-ecn.xml"> <front> <title>ECN++: Adding Explicit Congestion Notification (ECN) to TCP Control Packets</title> <author fullname="Marcelo Bagnulo"/> <author fullname="Bob Briscoe"/> <date day="27" month="July" year="2022"/> <abstract> <t>This document describes an experimental modification to ECN when used with TCP. It allows the use of ECN on the following TCP packets: SYNs, pure ACKs, Window probes, FINs, RSTs and retransmissions.</t> </abstract> </front> <seriesInfo name="Internet-Draft" value="draft-ietf-tcpm-generalized-ecn-10"/> </reference> <reference anchor="ARED01" target="https://www.icir.org/floyd/red.html"> <front> <title>Adaptive RED: An Algorithm for Increasing the Robustness of RED's Active Queue Management</title> <author fullname="Sally Floyd" initials="S." surname="Floyd"> <organization>ACIRI</organization> </author> <author fullname="Ramakrishna Gummadi" initials="R." surname="Gummadi"> <organization>ACIRI</organization> </author> <author fullname="S. Shenker" initials="S." surname="Shenker"> <organization>ACIRI</organization> </author> <date month="August" year="2001"/> </front> <seriesInfo name="ACIRI Technical Report" value=""/> </reference> <reference anchor="RFC8257" target="https://www.rfc-editor.org/info/rfc8257" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8257.xml"> <front> <title>Data Center TCP (DCTCP): TCP Congestion Control for Data Centers</title> <author initials="S." surname="Bensley" fullname="S. Bensley"> <organization/> </author> <author initials="D." surname="Thaler" fullname="D. Thaler"> <organization/> </author> <author initials="P." surname="Balasubramanian" fullname="P. Balasubramanian"> <organization/> </author> <author initials="L." surname="Eggert" fullname="L. Eggert"> <organization/> </author> <author initials="G." surname="Judd" fullname="G. Judd"> <organization/> </author> <date year="2017" month="October"/> <abstract> <t>This Informational RFC describes Data Center TCP (DCTCP): a TCP congestion control scheme for data-center traffic. DCTCP extends the Explicit Congestion Notification (ECN) processing to estimate the fraction of bytes that encounter congestion rather than simply detecting that some congestion has occurred. DCTCP then scales the TCP congestion window based on this estimate. This method achieves high-burst tolerance, low latency, and high throughput with shallow- buffered switches. This memo also discusses deployment issues related to the coexistence of DCTCP and conventional TCP, discusses the lack of a negotiating mechanism between sender and receiver, and presents some possible mitigations. This memo documents DCTCP as currently implemented by several major operating systems. DCTCP, as described in this specification, is applicable to deployments in controlled environments like data centers, but it must not be deployed over the public Internet without additional measures.</t> </abstract> </front> <seriesInfo name="RFC" value="8257"/> <seriesInfo name="DOI" value="10.17487/RFC8257"/> </reference> <reference anchor="RFC8312" target="https://www.rfc-editor.org/info/rfc8312" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8312.xml"> <front> <title>CUBIC for Fast Long-Distance Networks</title> <author initials="I." surname="Rhee" fullname="I. Rhee"> <organization/> </author> <author initials="L." surname="Xu" fullname="L. Xu"> <organization/> </author> <author initials="S." surname="Ha" fullname="S. Ha"> <organization/> </author> <author initials="A." surname="Zimmermann" fullname="A. Zimmermann"> <organization/> </author> <author initials="L." surname="Eggert" fullname="L. Eggert"> <organization/> </author> <author initials="R." surname="Scheffenegger" fullname="R. Scheffenegger"> <organization/> </author> <date year="2018" month="February"/> <abstract> <t>CUBIC is an extension to the current TCP standards. It differs from the current TCP standards only in the congestion control algorithm on the sender side. In particular, it uses a cubic function instead of a linear window increase function of the current TCP standards to improve scalability and stability under fast and long-distance networks. CUBIC and its predecessor algorithm have been adopted as defaults by Linux and have been used for many years. This document provides a specification of CUBIC to enable third-party implementations and to solicit community feedback through experimentation on the performance of CUBIC.</t> </abstract> </front> <seriesInfo name="RFC" value="8312"/> <seriesInfo name="DOI" value="10.17487/RFC8312"/> </reference> <reference anchor="RFC8511" target="https://www.rfc-editor.org/info/rfc8511" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8511.xml"> <front> <title>TCP Alternative Backoff with ECN (ABE)</title> <author initials="N." surname="Khademi" fullname="N. Khademi"> <organization/> </author> <author initials="M." surname="Welzl" fullname="M. Welzl"> <organization/> </author> <author initials="G." surname="Armitage" fullname="G. Armitage"> <organization/> </author> <author initials="G." surname="Fairhurst" fullname="G. Fairhurst"> <organization/> </author> <date year="2018" month="December"/> <abstract> <t>Active Queue Management (AQM) mechanisms allow for burst tolerance while enforcing short queues to minimise the time that packets spend enqueued at a bottleneck. This can cause noticeable performance degradation for TCP connections traversing such a bottleneck, especially if there are only a few flows or their bandwidth-delay product (BDP) is large. The reception of a Congestion Experienced (CE) Explicit Congestion Notification (ECN) mark indicates that an AQM mechanism is used at the bottleneck, and the bottleneck network queue is therefore likely to be short. Feedback of this signal allows the TCP sender-side ECN reaction in congestion avoidance to reduce the Congestion Window (cwnd) by a smaller amount than the congestion control algorithm's reaction to inferred packet loss. Therefore, this specification defines an experimental change to the TCP reaction specified in RFC 3168, as permitted by RFC 8311.</t> </abstract> </front> <seriesInfo name="RFC" value="8511"/> <seriesInfo name="DOI" value="10.17487/RFC8511"/> </reference> <reference anchor="RFC8985" target="https://www.rfc-editor.org/info/rfc8985" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8985.xml"> <front> <title>The RACK-TLP Loss Detection Algorithm for TCP</title> <author initials="Y." surname="Cheng" fullname="Y. Cheng"> <organization/> </author> <author initials="N." surname="Cardwell" fullname="N. Cardwell"> <organization/> </author> <author initials="N." surname="Dukkipati" fullname="N. Dukkipati"> <organization/> </author> <author initials="P." surname="Jha" fullname="P. Jha"> <organization/> </author> <date year="2021" month="February"/> <abstract> <t>This document presents the RACK-TLP loss detection algorithm for TCP. RACK-TLP uses per-segment transmit timestamps and selective acknowledgments (SACKs) and has two parts. Recent Acknowledgment (RACK) starts fast recovery quickly using time-based inferences derived from acknowledgment (ACK) feedback, and Tail Loss Probe (TLP) leverages RACK and sends a probe packet to trigger ACK feedback to avoid retransmission timeout (RTO) events. Compared to the widely used duplicate acknowledgment (DupAck) threshold approach, RACK-TLP detects losses more efficiently when there are application-limited flights of data, lost retransmissions, or data packet reordering events. It is intended to be an alternative to the DupAck threshold approach.</t> </abstract> </front> <seriesInfo name="RFC" value="8985"/> <seriesInfo name="DOI" value="10.17487/RFC8985"/> </reference> <reference anchor="RFC8888" target="https://www.rfc-editor.org/info/rfc8888" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8888.xml"> <front> <title>RTP Control Protocol (RTCP) Feedback for Congestion Control</title> <author initials="Z." surname="Sarker" fullname="Z. Sarker"> <organization/> </author> <author initials="C." surname="Perkins" fullname="C. Perkins"> <organization/> </author> <author initials="V." surname="Singh" fullname="V. Singh"> <organization/> </author> <author initials="M." surname="Ramalho" fullname="M. Ramalho"> <organization/> </author> <date year="2021" month="January"/> <abstract> <t>An effective RTP congestion control algorithm requires more fine-grained feedback on packet loss, timing, and Explicit Congestion Notification (ECN) marks than is provided by the standard RTP Control Protocol (RTCP) Sender Report (SR) and Receiver Report (RR) packets. This document describes an RTCP feedback message intended to enable congestion control for interactive real-time traffic using RTP. The feedback message is designed for use with a sender-based congestion control algorithm, in which the receiver of an RTP flow sends back to the sender RTCP feedback packets containing the information the sender needs to perform congestion control.</t> </abstract> </front> <seriesInfo name="RFC" value="8888"/> <seriesInfo name="DOI" value="10.17487/RFC8888"/> </reference> <reference anchor="RFC9000" target="https://www.rfc-editor.org/info/rfc9000" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.9000.xml"> <front> <title>QUIC: A UDP-Based Multiplexed and Secure Transport</title> <author initials="J." surname="Iyengar" fullname="J. Iyengar" role="editor"> <organization/> </author> <author initials="M." surname="Thomson" fullname="M. Thomson" role="editor"> <organization/> </author> <date year="2021" month="May"/> <abstract> <t>This document defines the core of the QUIC transport protocol. QUIC provides applications with flow-controlled streams for structured communication, low-latency connection establishment, and network path migration. QUIC includes security measures that ensure confidentiality, integrity, and availability in a range of deployment circumstances. Accompanying documents describe the integration of TLS for key negotiation, loss detection, and an exemplary congestion control algorithm.</t> </abstract> </front> <seriesInfo name="RFC" value="9000"/> <seriesInfo name="DOI" value="10.17487/RFC9000"/> </reference> <reference anchor="RFC9001" target="https://www.rfc-editor.org/info/rfc9001" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.9001.xml"> <front> <title>Using TLS to Secure QUIC</title> <author initials="M." surname="Thomson" fullname="M. Thomson" role="editor"> <organization/> </author> <author initials="S." surname="Turner" fullname="S. Turner" role="editor"> <organization/> </author> <date year="2021" month="May"/> <abstract> <t>This document describes how Transport Layer Security (TLS) is used to secure QUIC.</t> </abstract> </front> <seriesInfo name="RFC" value="9001"/> <seriesInfo name="DOI" value="10.17487/RFC9001"/> </reference> <reference anchor="RFC9147" target="https://www.rfc-editor.org/info/rfc9147" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.9147.xml"> <front> <title>The Datagram Transport Layer Security (DTLS) Protocol Version 1.3</title> <author initials="E." surname="Rescorla" fullname="E. Rescorla"> <organization/> </author> <author initials="H." surname="Tschofenig" fullname="H. Tschofenig"> <organization/> </author> <author initials="N." surname="Modadugu" fullname="N. Modadugu"> <organization/> </author> <date year="2022" month="April"/> <abstract> <t>This document specifies version 1.3 of the Datagram Transport Layer Security (DTLS) protocol. DTLS 1.3 allows client/server applications to communicate over the Internet in a way that is designed to prevent eavesdropping, tampering, and message forgery.</t> <t>The DTLS 1.3 protocol is based on the Transport Layer Security (TLS) 1.3 protocol and provides equivalent security guarantees with the exception of order protection / non-replayability. Datagram semantics of the underlying transport are preserved by the DTLS protocol.</t> <t>This document obsoletes RFC 6347.</t> </abstract> </front> <seriesInfo name="RFC" value="9147"/> <seriesInfo name="DOI" value="10.17487/RFC9147"/> </reference> <reference anchor="I-D.sridharan-tcpm-ctcp" target="https://datatracker.ietf.org/api/v1/doc/document/draft-sridharan-tcpm-ctcp/" xml:base="https://bib.ietf.org/public/rfc/bibxml-ids/reference.I-D.sridharan-tcpm-ctcp.xml"> <front> <title>Compound TCP: A New TCP Congestion Control for High-Speed and Long Distance Networks</title> <author fullname="Murali Sridharan"/> <author fullname="Kun Tan"/> <author fullname="Deepak Bansal"/> <author fullname="Dave Thaler"/> <date day="29" month="October" year="2007"/> <abstract> <t>Compound TCP (CTCP) is a modification to TCP's congestion control mechanism for use with TCP connections with large congestion windows. This document describes the Compound TCP algorithm in detail, and solicits experimentation and feedback from the wider community. The key idea behind CTCP is to add a scalable delay-based component to the standard TCP's loss-based congestion control. The sending rate of CTCP is controlled by both loss and delay components. The delay-based component has a scalable window increasing rule that not only efficiently uses the link capacity, but on sensing queue build up, proactively reduces the sending rate.</t> </abstract> </front> <seriesInfo name="Internet-Draft" value="draft-sridharan-tcpm-ctcp-02"/> </reference> <reference anchor="I-D.briscoe-docsis-q-protection" target="https://datatracker.ietf.org/api/v1/doc/document/draft-briscoe-docsis-q-protection/" xml:base="https://bib.ietf.org/public/rfc/bibxml-ids/reference.I-D.briscoe-docsis-q-protection.xml"> <front> <title>The DOCSIS(r) Queue Protection Algorithm to Preserve Low Latency</title> <author fullname="Bob Briscoe"/> <author fullname="Greg White"/> <date day="13" month="May" year="2022"/> <abstract> <t>This informational document explains the specification of the queue protection algorithm used in DOCSIS technology since version 3.1. A shared low latency queue relies on the non-queue-building behaviour of every traffic flow using it. However, some flows might not take such care, either accidentally or maliciously. If a queue is about to exceed a threshold level of delay, the queue protection algorithm can rapidly detect the flows most likely to be responsible. It can then prevent harm to other traffic in the low latency queue by ejecting selected packets (or all packets) of these flows. The document is designed for four types of audience: a) congestion control designers who need to understand how to keep on the 'good' side of the algorithm; b) implementers of the algorithm who want to understand it in more depth; c) designers of algorithms with similar goals, perhaps for non-DOCSIS scenarios; and d) researchers interested in evaluating the algorithm.</t> </abstract> </front> <seriesInfo name="Internet-Draft" value="draft-briscoe-docsis-q-protection-06"/> </reference> <reference anchor="I-D.briscoe-iccrg-prague-congestion-control" target="https://datatracker.ietf.org/api/v1/doc/document/draft-briscoe-iccrg-prague-congestion-control/" xml:base="https://bib.ietf.org/public/rfc/bibxml-ids/reference.I-D.briscoe-iccrg-prague-congestion-control.xml"> <front> <title>Prague Congestion Control</title> <author fullname="Koen De Schepper"/> <author fullname="Olivier Tilmans"/> <author fullname="Bob Briscoe"/> <date day="11" month="July" year="2022"/> <abstract> <t>This specification defines the Prague congestion control scheme, which is derived from DCTCP and adapted for Internet traffic by implementing the Prague L4S requirements. Over paths with L4S support at the bottleneck, it adapts the DCTCP mechanisms to achieve consistently low latency and full throughput. It is defined independently of any particular transport protocol or operating system, but notes are added that highlight issues specific to certain transports and OSs. It is mainly based on the current default options of the reference Linux implementation of TCP Prague, but it includes experience from other implementations where available. It separately describes non-default and optional parts, as well as future plans.</t> <t>The implementation does not satisfy all the Prague requirements (yet) and the IETF might decide that certain requirements need to be relaxed as an outcome of the process of trying to satisfy them all. In two cases, research code is replaced by placeholders until full evaluation is complete.</t> </abstract> </front> <seriesInfo name="Internet-Draft" value="draft-briscoe-iccrg-prague-congestion-control-01"/> </reference> <reference anchor="I-D.cardwell-iccrg-bbr-congestion-control" target="https://datatracker.ietf.org/api/v1/doc/document/draft-cardwell-iccrg-bbr-congestion-control/" xml:base="https://bib.ietf.org/public/rfc/bibxml-ids/reference.I-D.cardwell-iccrg-bbr-congestion-control.xml"> <front> <title>BBR Congestion Control</title> <author fullname="Neal Cardwell"/> <author fullname="Yuchung Cheng"/> <author fullname="Soheil Hassas Yeganeh"/> <author fullname="Ian Swett"/> <author fullname="Van Jacobson"/> <date day="7" month="March" year="2022"/> <abstract> <t>This document specifies the BBR congestion control algorithm. BBR ("Bottleneck Bandwidth and Round-trip propagation time") uses recent measurements of a transport connection's delivery rate, round-trip time, and packet loss rate to build an explicit model of the network path. BBR then uses this model to control both how fast it sends data and the maximum volume of data it allows in flight in the network at any time. Relative to loss-based congestion control algorithms such as Reno [RFC5681] or CUBIC [RFC8312], BBR offers substantially higher throughput for bottlenecks with shallow buffers or random losses, and substantially lower queueing delays for bottlenecks with deep buffers (avoiding "bufferbloat"). BBR can be implemented in any transport protocol that supports packet-delivery acknowledgment. Thus far, open source implementations are available for TCP [RFC793] and QUIC [RFC9000]. This document specifies version 2 of the BBR algorithm, also sometimes referred to as BBRv2 or bbr2.</t> </abstract> </front> <seriesInfo name="Internet-Draft" value="draft-cardwell-iccrg-bbr-congestion-control-02"/> </reference> <reference anchor="I-D.mathis-iccrg-relentless-tcp" target="https://www.ietf.org/archive/id/draft-mathis-iccrg-relentless-tcp-00.txt" xml:base="https://bib.ietf.org/public/rfc/bibxml-ids/reference.I-D.mathis-iccrg-relentless-tcp.xml"> <front> <title>Relentless Congestion Control</title> <author fullname="Matt Mathis"/> <date day="4" month="March" year="2009"/> <abstract> <t>Relentless congestion control is a simple modification that can be applied to almost any AIMD style congestion control: instead of applying a multiplicative reduction to cwnd after a loss, cwnd is reduced by the number of lost segments. It can be modeled as a strict implementation of van Jacobson's Packet Conservation Principle. During recovery, new segments are injected into the network in exact accordance with the segments that are reported to have been delivered to the receiver by the returning ACKs. This algorithm offers a valuable new congestion control property: the TCP portion of the control loop has exactly unity gain, which should make it easier to implement simple controllers in network devices to accurately control queue sizes across a huge range of scales. Relentless Congestion Control conforms to neither the details nor the philosophy of current congestion control standards. These standards are based on the idea that the Internet can attain sufficient fairness by having relatively simple network devices send uniform congestion signals to all flows, and mandating that all protocols have equivalent responses to these congestion signals. To function appropriately in a shared environment, Relentless Congestion Control requires that the network allocates capacity through some technique such as Fair Queuing, Approximate Fair Dropping, etc. The salient features of these algorithms are that they segregate the traffic into distinct flows, and send different congestion signals to each flow. This alternative congestion control paradigm is described in a separate document, also under consideration by the ICCRG. The goal of the document is to illustrate some new protocol features and properties might be possible if we relax the "TCP-friendly" mandate. A secondary goal of Relentless TCP is to make a distinction between the bottlenecks that belong to protocol itself, vs standard congestion control and the "TCP-friendly" paradigm.</t> </abstract> </front> <seriesInfo name="Internet-Draft" value="draft-mathis-iccrg-relentless-tcp-00"/> </reference> <reference anchor="BBRv2" target="https://github.com/google/bbr/blob/v2alpha/README.md"> <front> <title>BRTCP BBR v2 Alpha/Preview Release</title> <author fullname="Neal Cardwell" initials="N" surname="Cardwell"> <organization/> </author> <date/> </front> <seriesInfo name="GitHub repository;" value="Linux congestion control module"/> </reference> <!--{ToDo: DCttH ref will need to be updated, once stable}--> <reference anchor="DCttH19" target="https://bobbriscoe.net/pubs.html#DCttH_TR"> <front> <title>`Data Centre to the Home': Ultra-Low Latency for All</title> <author fullname="Koen De Schepper" initials="K." surname="De Schepper"> <organization>Nokia Bell Labs</organization> </author> <author fullname="Olga Bondarenko" initials="O." surname="Bondarenko"> <organization>Simula Research Lab</organization> </author> <author fullname="Olivier" initials="O." surname="Tilmans"> <organization>Nokia Bell Labs</organization> </author> <author fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization>Independent (bobbriscoe.net)</organization> </author> <date month="July" year="2019"/> </front> <seriesInfo name="Updated RITE project Technical Report" value=""/> <format target="https://bobbriscoe.net/projects/latency/dctth_journal_draft20190726.pdf" type="PDF"/> </reference> <reference anchor="PI2" target="https://dl.acm.org/citation.cfm?doid=2999572.2999578"> <front> <title>PI^2 : A Linearized AQM for both Classic and Scalable TCP</title> <author fullname="Koen De Schepper" initials="K." surname="De Schepper"> <organization>Bell Labs</organization> </author> <author fullname="Olga Bondarenko" initials="O." surname="Bondarenko"> <organization>Simula Research Lab</organization> </author> <author fullname="Ing-jyh Tsang" initials="I." surname="Tsang"> <organization>Bell Labs</organization> </author> <author fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization>BT</organization> </author> <date month="December" year="2016"/> </front> <seriesInfo name="Proc. ACM CoNEXT 2016" value="pp.105-119"/> </reference> <reference anchor="VCP" target="https://doi.acm.org/10.1145/1080091.1080098"> <front> <title>One more bit is enough</title> <author fullname="Yong Xia" initials="Y." surname="Xia"> <organization/> </author> <author fullname="Lakshminarayanan Subramanian" initials="L." surname="Subramanian"> <organization/> </author> <author fullname="Ion Stoica" initials="I." surname="Stoica"> <organization/> </author> <author fullname="Shivkumar Kalyanaraman" initials="S." surname="Kalyanaraman"> <organization/> </author> <date month="" year="2005"/> </front> <seriesInfo name="Proc. SIGCOMM'05, ACM CCR" value="35(4)37--48"/> <format target="https://conferences.sigcomm.org/sigcomm/2005/paper-XiaSub.pdf" type="PDF"/> </reference> <reference anchor="QV" target="https://riteproject.files.wordpress.com/2015/12/rite-deliverable-2-3.pdf"> <front> <title>Up to Speed with Queue View</title> <author fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization>BT</organization> </author> <author fullname="Per Hurtig" initials="P." surname="Hurtig"> <organization>Uni Karlstad</organization> </author> <date month="August" year="2015"/> </front> <seriesInfo name="RITE Technical Report" value="D2.3; Appendix C.2"/> </reference> <reference anchor="Alizadeh-stability" target="https://people.csail.mit.edu/alizadeh/papers/dctcp_analysis-sigmetrics11.pdf"> <front> <title>Analysis of DCTCP: Stability, Convergence, and Fairness</title> <author fullname="Mohamed Alizadeh" initials="M." surname="Alizadeh"/> <author fullname="Adel Javanmard" initials="A." surname="Javanmard"/> <author fullname="Balaji Prabhakar" initials="B." surname="Prabhakar"/> <date month="June" year="2011"/> </front> <seriesInfo name="ACM SIGMETRICS 2011" value=""/> </reference> <reference anchor="TCPPrague" target="https://www.ietf.org/mail-archive/web/tcpprague/current/msg00001.html"> <front> <title>Notes: DCTCP evolution 'bar BoF': Tue 21 Jul 2015, 17:40, Prague</title> <author fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization>Simula</organization> </author> <date month="July" year="2015"/> </front> <seriesInfo name="tcpprague mailing list archive" value=""/> </reference> <reference anchor="sub-mss-prob" target="https://arxiv.org/abs/1904.07598"> <front> <title>Scaling TCP's Congestion Window for Small Round Trip Times</title> <author fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization>BT</organization> </author> <author fullname="Koen De Schepper" initials="K." surname="De Schepper"> <organization>Bell Labs</organization> </author> <date month="May" year="2015"/> </front> <seriesInfo name="BT Technical Report" value="TR-TUB8-2015-002"/> <format target="https://arxiv.org/pdf/1904.07598" type="PDF"/> </reference> <reference anchor="ecn-fallback" target="https://arxiv.org/abs/1911.00710"> <front> <title>TCP Prague Fall-back on Detection of a Classic ECN AQM</title> <author fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization>Independent</organization> </author> <author fullname="Asad Sajjad Ahmed" initials="A.S." surname="Ahmed"> <organization>Simula and Uni Oslo</organization> </author> <date month="April" year="2020"/> </front> <seriesInfo name="bobbriscoe.net Technical Report" value="TR-BB-2019-002"/> </reference> <reference anchor="Ahmed19" target="https://www.duo.uio.no/handle/10852/70966"> <front> <title>Extending TCP for Low Round Trip Delay</title> <author fullname="Asad Sajjad Ahmed" initials="A.S." surname="Ahmed"> <organization>Simula and Uni Oslo</organization> </author> <date month="August" year="2019"/> </front> <seriesInfo name="Master's Thesis, Uni Oslo" value=""/> <format target="https://www.duo.uio.no/bitstream/handle/10852/70966/main.pdf?sequence=5&isAllowed=y" type="PDF"/> </reference> <reference anchor="Paced-Chirping" target="https://riteproject.files.wordpress.com/2018/07/misundjoakimmastersthesissubmitted180515.pdf"> <front> <title>Rapid Acceleration in TCP Prague</title> <author fullname="Joakim Misund" initials="J." surname="Misund"> <organization>University of Oslo and Simula</organization> </author> <date month="May" year="2018"/> </front> <seriesInfo name="Master's Thesis" value=""/> </reference> <reference anchor="A2DTCP" target="https://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6871352"> <front> <title>Adaptive-Acceleration Data Center TCP</title> <author fullname="Tao Zhang" initials="T." surname="Zhang"> <organization/> </author> <author fullname="Jianxin Wang" initials="J." surname="Wang"> <organization/> </author> <author fullname="Jiawei Huang" initials="J." surname="Huang"> <organization/> </author> <author fullname="Yi Huang" initials="Y." surname="Huang"> <organization/> </author> <author fullname="Jianer Chen" initials="J." surname="Chen"> <organization/> </author> <author fullname="Yi Pan" initials="Y." surname="Pan"> <organization/> </author> <date month="June" year="2015"/> </front> <seriesInfo name="IEEE Transactions on Computers" value="64(6):1522-1533"/> </reference> <reference anchor="PragueLinux" target="https://www.netdevconf.org/0x13/session.html?talk-tcp-prague-l4s"> <front> <title>Implementing the `TCP Prague' Requirements for Low Latency Low Loss Scalable Throughput (L4S)</title> <author fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization>Independent</organization> </author> <author fullname="Koen De Schepper" initials="K." surname="De Schepper"> <organization>Nokia Bell Labs</organization> </author> <author fullname="Olga Albisser" initials="O." surname="Albisser"> <organization>Simula</organization> </author> <author fullname="Joakim Misund" initials="J." surname="Misund"> <organization>University of Oslo</organization> </author> <author fullname="Olivier Tilmans" initials="O." surname="Tilmans"> <organization>Nokia Bell Labs</organization> </author> <author fullname="Mirja Kühlewind" initials="M." surname="Kühlewind"> <organization>ETHZ</organization> </author> <author fullname="Asad Sajjad Ahmed" initials="A.S." surname="Ahmed"> <organization>Simula</organization> </author> <date month="March" year="2019"/> </front> <seriesInfo name="Proc. Linux Netdev 0x13" value=""/> <format target="https://www.files.netdevconf.info/f/4d6939d5f1fb404fafd1/?dl=1" type="PDF"/> </reference> <reference anchor="LinuxPacedChirping" target="https://www.netdevconf.org/0x13/session.html?talk-chirp"> <front> <title>Paced Chirping - Rethinking TCP start-up</title> <author fullname="Joakim Misund" initials="J." surname="Misund"> <organization>University of Oslo</organization> </author> <author fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization>Independent</organization> </author> <date month="March" year="2019"/> </front> <seriesInfo name="Proc. Linux Netdev 0x13" value=""/> <format target="https://www.files.netdevconf.info/f/da8cc04a608a448f890e/?dl=1" type="PDF"/> </reference> <reference anchor="TCP-CA" target="https://ee.lbl.gov/papers/congavoid.pdf"> <front> <title>Congestion Avoidance and Control</title> <author fullname="Van Jacobson" initials="V." surname="Jacobson"> <organization/> </author> <author fullname="Michael J. Karels" initials="M.J." surname="Karels"> <organization/> <address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address> </author> <date month="November" year="1988"/> </front> <seriesInfo name="Laurence Berkeley Labs Technical Report" value=""/> </reference> <reference anchor="Savage-TCP"> <front> <title>TCP Congestion Control with a Misbehaving Receiver</title> <author fullname="Stefan Savage" initials="S." surname="Savage"> <organization>Uni Washington</organization> </author> <author fullname="Neal Cardwell" initials="N." surname="Cardwell"> <organization>Uni Washington</organization> </author> <author fullname="David Wetherall" initials="D." surname="Wetherall"> <organization>Uni Washington</organization> </author> <author fullname="Tom Anderson" initials="T." surname="Anderson"> <organization>Uni Washington</organization> </author> <date month="October" year="1999"/> </front> <seriesInfo name="ACM SIGCOMM Computer Communication Review" value="29(5):71--78"/> </reference> <reference anchor="Heist21" target="https://github.com/heistp/l4s-tests/"> <front> <title>L4S Tests</title> <author fullname="Pete Heist" initials="P." surname="Heist"> <organization/> </author> <author fullname="Jonathan Morton" initials="J." surname="Morton"> <organization/> </author> <date month="May" year="2021"/> </front> <seriesInfo name="GitHub" value="README"/> </reference> <reference anchor="SCReAM-L4S" target="https://github.com/EricssonResearch/scream/blob/master/README.md"> <front> <title>SCReAM</title> <author fullname="Ingemar Johansson" initials="I" surname="Johansson"> <organization/> </author> <date/> </front> <seriesInfo name="GitHub repository;" value=""/> <format target="https://github.com/google/bbr/blob/v2alpha/README.md" type="Source code"/> </reference> <reference anchor="DualPI2Linux" target="https://www.netdevconf.org/0x13/session.html?talk-DUALPI2-AQM"> <front> <title>DUALPI2 - Low Latency, Low Loss and Scalable (L4S) AQM</title> <author fullname="Olga Albisser" initials="O." surname="Albisser"> <organization>Simula Research Lab</organization> </author> <author fullname="Koen De Schepper" initials="K." surname="De Schepper"> <organization>Nokia Bell Labs</organization> </author> <author fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization>Independent</organization> </author> <author fullname="Olivier Tilmans" initials="O." surname="Tilmans"> <organization>Nokia Bell Labs</organization> </author> <author fullname="Henrik Steen" initials="H." surname="Steen"> <organization>Simula Research Lab</organization> </author> <date month="March" year="2019"/> </front> <seriesInfo name="Proc. Linux Netdev 0x13" value=""/> <format target="https://www.files.netdevconf.org/f/febbe8c6a05b4ceab641/?dl=1" type="PDF"/> </reference> <reference anchor="COBALT" target="https://doi.org/10.1109/LANMAN.2019.8847054"> <front> <title>Design and Evaluation of COBALT Queue Discipline</title> <author initials="J." surname="Palmei"/> <author initials="S." surname="Gupta"/> <author initials="P." surname="Imputato"/> <author initials="J." surname="Morton"/> <author initials="M." surname="Tahiliani"/> <author initials="S." surname="Avallone"/> <author initials="D." surname="Taht"/> <date year="2019"/> </front> <seriesInfo name="In Proc. IEEE Int'l Symp. on Local and Metropolitan Area Networks" value="2019, pp1--6"/> </reference> <reference anchor="Dukkipati06" target="https://dl.acm.org/doi/10.1145/1111322.1111336"> <front> <title>Why Flow-Completion Time is the Right Metric<title>Why Flow-Completion Time is the Right Metric for Congestion Control</title> <author fullname="Nandita Dukkipati" initials="N." surname="Dukkipati"> <organization>StanfordUni</organization>University</organization> </author> <author fullname="Nick McKeown" initials="N." surname="McKeown"> <organization>StanfordUni</organization>University</organization> </author> <date month="January" year="2006"/> </front> <seriesInfoname="ACM CCR" value="36(1):59--62"/> <format target="http://yuba.stanford.edu/rcp/flowCompTime-dukkipati.pdf" type="PDF"/>name="DOI" value="10.1145/1111322.1111336"/> <refcontent>ACM SIGCOMM Computer Communication Review, Volume 36, Issue 1, pp. 59-62</refcontent> </reference> <reference anchor="Bufferbloat" target="https://bufferbloat.net/"> <front> <title>Bufferbloat</title><author fullname=""> <organization/><author> <organization>The Bufferbloat community</organization> </author> <date/> </front><annotation>(last accessed 27 Aug 2022)</annotation></reference> </references> </references> <section anchor="l4sps_tcp-prague-reqs" numbered="true" toc="default"> <name>Rationale for the 'Prague L4S Requirements'</name> <t>This appendix is informative, not normative. It gives a list of modifications to currentscalableScalable congestion controls so that they can be deployed over the public Internet and coexist safely with existing traffic. The list complements the normative requirements in <xref target="l4sid_transport_req" format="default"/> that a sender has to comply with before it can set the L4S identifier in packets it sends into the Internet. As well as rationale for safety improvements (the requirements in <xref target="l4sid_transport_req"format="default"/>)format="default"/>), this appendix also includes preferable performance improvements (optimizations).</t> <t>The requirements and recommendations in <xref target="l4sid_transport_req"format="default"/>)format="default"/> have become known as thePrague'Prague L4SRequirements,Requirements', because they were originally identified at an ad hoc meeting duringIETF-94IETF 94 inPrague <xrefPrague <xref target="TCPPrague" format="default"/>. They were originally called the 'TCP Prague Requirements', but they are not solely applicable to TCP, so the name and wording has been generalized for all transport protocols, and the name 'TCP Prague' is now used for a specific implementation of the requirements.</t> <t>At the time of writing,DCTCP <xrefDCTCP <xref target="RFC8257" format="default"/> is the most widely usedscalableScalable transport protocol. In its current form, DCTCP is specified to be deployable only in controlled environments. Deploying it in the public Internet would lead to a number of issues,bothfrom both the safety and the performance perspective. The modifications and additional mechanisms listed in this section will be necessary for its deployment over the global Internet. Where an example is needed, DCTCP is used as a base, but the requirements in <xref target="l4sid_transport_req" format="default"/> apply equally to otherscalableScalable congestion controls, covering adaptive real-time media, etc., not just capacity-seeking behaviours.</t> <section numbered="true" toc="default"> <name>Rationale for the Requirements for Scalable Transport Protocols</name><t/><section numbered="true" toc="default"> <name>Use of L4S Packet Identifier</name> <t>Description: AscalableScalable congestion control needs to distinguish the packets it sends from those sent by Classic congestion controls (see the precise normative requirement wording in <xref target="l4sid_codepoint" format="default"/>).</t> <t>Motivation: It needs to be possible for a network node to classify L4S packets without flow state into a queue that applies an L4SECN markingECN-marking behaviour and isolates L4S packets from the queuing delay of Classic packets.</t> </section> <section numbered="true" toc="default"> <name>Accurate ECN Feedback</name> <t>Description: The transport protocol for ascalableScalable congestion control needs to provide timely, accurate feedback about the extent of ECN marking experienced by all packets (see the precise normative requirement wording in <xref target="l4sid_feedback" format="default"/>).</t> <t>Motivation: Classic congestion controls only need feedback about the existence of a congestion episode within a round trip, not precisely how many packets weremarked with ECNECN-marked or dropped. Therefore, in 2001, when ECN feedback was added toTCP <xrefTCP <xref target="RFC3168" format="default"/>, it could not inform the sender of more than one ECN mark per RTT. Since then, requirements for more accurate ECN feedback in TCP have been defined in <xref target="RFC7560"format="default"/>format="default"/>, and <xref target="I-D.ietf-tcpm-accurate-ecn" format="default"/> specifies a change to the TCP protocol to satisfy these requirements. Most other transport protocols already satisfy this requirement (see <xref target="l4sid_feedback" format="default"/>).</t> </section> <section anchor="l4sid_sec_replaceable" numbered="true" toc="default"> <name>Capable of Replacement by Classic Congestion Control</name> <t>Description: It needs to be possible to replace the implementation of ascalableScalable congestion control with a Classic control (see the precise normative requirement wording in <xreftarget="l4sid_congestion_response"target="l4sid_Prague_req-replaceable" format="default"/>).</t> <t>Motivation: L4S is an experimentalprotocol, thereforeprotocol; therefore, it seems prudent to be able to disable it at source in case of insurmountable problems, perhaps due to some unexpected interaction on a particular sender; over a particular path or network; or with a particularreceiverreceiver, or even ultimately an insurmountable problem with the experiment as a whole.</t> </section> <section anchor="l4sid_sec_fallback_on_loss" numbered="true" toc="default"> <name>FallbackBack to Classic Congestion Control on Packet Loss</name> <t>Description: As well as responding to ECN markings in a scalable way, ascalableScalable congestion control needs to react to packet loss in a way that will coexist safely with a Reno congestioncontrol <xrefcontrol <xref target="RFC5681" format="default"/> (see the precise normative requirement wording in <xreftarget="l4sid_congestion_response"target="l4sid_Prague_req-loss-response" format="default"/>).</t> <t>Motivation: Part of the safety conditions for deploying ascalableScalable congestion control on the public Internet is to make sure that it behaves properly when it builds a queue at a network bottleneck that has not been upgraded to support L4S. Packet loss can have many causes, but it usually has to be conservatively assumed that it is a sign of congestion. Therefore, on detecting packet loss, ascalableScalable congestion control will need to fall back to Classic congestion control behaviour. If it does not comply, it could starve Classic traffic.</t> <t>AscalableScalable congestion control can be used for different types of transport,e.g. fore.g., for real-time media or for reliable transport like TCP. Therefore, the particular Classic congestion control behaviour to fall back on will need to be dependent on the specific congestion control implementation. In the particular case of DCTCP, the DCTCPspecification <xrefspecification <xref target="RFC8257" format="default"/> states that"It is RECOMMENDED that an implementation deal with"A DCTCP sender <bcp14>MUST</bcp14> react to loss episodes in the same way as conventionalTCP." For safe deployment, <xref target="l4sid_congestion_response" format="default"/> requiresTCP,...". To ensure anyspecification of a scalableScalable congestion controlforis safe to deploy over the publicInternet to defineInternet, Item <xref target="l4sid_Prague_req-loss-response" format="counter"/> of <xref target="l4sid_congestion_response" format="default"/> in theabove requirementpresent spec does not require precisely the same response as Reno TCP, but it does require a"MUST".</t>response that will coexist safely with Classic congestion controls like Reno.</t> <t>Even though a bottleneck is L4S capable, it might still become overloaded and have to drop packets. In this case, the sender may receive a high proportion of packets marked with the CEbit setcodepoint and also experience loss. Current DCTCP implementations each react differently to this situation.At least one implementation reactsOne approach is to react only to the drop signal(e.g. by(e.g., by halving theCWND) and at leastcwnd); anotherDCTCP implementation reactsapproach is toboth signals (e.g. by halving the CWND duereact tothe drop and also further reducing the CWND based on the proportion of marked packet).both signals, which reduces cwnd by more than half. Athird approach for the public Internetcompromise between these two has been proposedthat adjustswhere the loss response is adjusted to result in a halving when combined withtheany ECNresponse.response earlier in the same round. We believe that further experimentation is needed to understand what is the best behaviour for the public Internet, which may or may not be one of these existing approaches.</t> </section> <section anchor="l4sid_sec_fallback_on_classic_CE" numbered="true" toc="default"> <name>Coexistence with Classic Congestion Control at Classic ECNbottlenecks</name>Bottlenecks</name> <t>Description: Monitoring has to be in place so that a non-L4S but ECN-capable AQM can be detected at path bottlenecks. This is in case such an AQM has been implemented in a shared queue, in which case any long-runningscalableScalable flow would predominate over any simultaneous long-running Classic flow sharing the queue. The precise requirement wording in <xreftarget="l4sid_congestion_response"target="l4sid_Prague_req-classic-ecn-response" format="default"/> is written so that such a problem couldeitherbe resolved either inreal-time,real time or via administrative intervention.</t> <t>Motivation: Similarly to the discussion in <xref target="l4sid_sec_fallback_on_loss" format="default"/>, this requirement in <xref target="l4sid_congestion_response" format="default"/> is a safety condition to ensure an L4S congestion control coexists well with Classic flows when it builds a queue at a shared network bottleneck that has not been upgraded to support L4S. Nonetheless, if necessary, it is considered reasonable to resolve such problems over management timescales (possibly involving human intervention) because:</t> <ul spacing="normal"> <li>although a Classic flow can considerably reduce its throughput in the face of a competingscalableScalable flow, it still makes progress and does not starve;</li> <li>implementations of a Classic ECN AQM in a queue that is intended to be shared are believed to berare;</li>rare; and</li> <li>detection of such AQMs is not always clear-cut; so focused out-of-band testing (or even contacting the relevant network operator) would improve certainty.</li> </ul><t>Therefore, the<t>The relevant normative requirement (<xref target="l4sid_congestion_response" format="default"/>) is therefore divided into three stages: monitoring,detectiondetection, and action:</t> <dl newline="false" spacing="normal"> <dt>Monitoring:</dt> <dd>Monitoring involves collection of the measurement data to be analysed. Monitoring is expressed as a'MUST'"<bcp14>MUST</bcp14>" for uncontrolled environments, although the placement of the monitoring function is left open. Whether monitoring has to be applied inreal-timereal time is expressed as a'SHOULD'."<bcp14>SHOULD</bcp14>". This allows for the possibility that the operator of an L4S sender(e.g. a CDN)(e.g., a Content Distribution Network (CDN)) might prefer to test out-of-band for signs of Classic ECN AQMs, perhaps to avoid continually consuming resources to monitor live traffic.</dd> <dt>Detection:</dt> <dd>Detection involves analysis of the monitored data to detect the likelihood of a Classic ECN AQM. Detection can either directly detect actual coexistence problems betweenflows,flows orit canaim to identify AQM technologies that are likely to present coexistence problems, based on knowledge of AQMs deployed at the time. The requirements recommend that detection occurs live inreal-time.real time. However, detection is allowed to be deferred(e.g. it(e.g., it might involve further testing targeted at candidateAQMs);</dd>AQMs).</dd> <dt>Action:</dt> <dd> <t>This involves the act of switching the sender to a Classic congestion control. This might occur inreal-timereal time within the congestion control for the subsequent duration of a flow, or it might involve administrative action to switch to Classic congestion control for a specific interface or for a certain set of destination addresses.</t> <t>Instead of the sender taking action itself, the operator of the sender(e.g. a(e.g., a CDN) might prefer to ask the network operator to modify the Classic AQM's treatment of L4S packets;or toensure L4S packets bypass the AQM; ortoupgrade the AQM to support L4S (see the L4S operationalguidance <xrefguidance <xref target="I-D.ietf-tsvwg-l4sops" format="default"/>).OnceIf L4S flows then no longer shared the Classic ECNAQMAQM, they would obviously no longer detect it, and the requirement to act on it would no longer apply.</t> </dd> </dl> <t>The whole set of normative requirements concerning Classic ECN AQMs in <xref target="l4sid_congestion_response" format="default"/> is worded so that it does not apply in controlled environments, such as private networks ordata centredata-centre networks. CDN servers placed within an access ISP's network can be considered as a single controlled environment, but any onward networks served by the access network, including all the attached customer networks, would be unlikely to fall under the same degree of coordinated control. Monitoring is expressed as a'MUST'"<bcp14>MUST</bcp14>" for these uncontrolled segments of paths(e.g. beyond(e.g., beyond the access ISP in a home network), because there is a possibility that there might be a shared queue Classic ECN AQM in that segment. Nonetheless, the intent of the wording is to only require occasional monitoring of these uncontrolledregions,regions and not to burden CDN operators if monitoring never uncovers any potential problems.</t> <t>More detailed discussion of all the above options and alternatives can be found in the L4S operationalguidance <xrefguidance <xref target="I-D.ietf-tsvwg-l4sops" format="default"/>.</t> <t>Having said all the above, the approach recommended in <xref target="l4sid_congestion_response" format="default"/> is to monitor,detectdetect, and act inreal-timereal time on live traffic. A passive monitoring algorithm to detect a Classic ECN AQM at the bottleneck and fall back to Classic congestion control is described in an extensive technicalreport <xrefreport <xref target="ecn-fallback" format="default"/>, which also provides a link to Linux sourcecode,code and a large online visualization of its evaluation results. Very briefly, the algorithm primarily monitors RTT variation using the same algorithm that maintains the mean deviation of TCP's smoothed RTT, but it smooths over a duration of the order of a Classic sawtooth. The outcome is also conditioned on other metrics such as the presence of CE marking and congestion avoidance phase having stabilized. The report also identifies further work to improve the approach, forinstanceinstance, improvements withlow capacitylow-capacity links and combining the measurements with a cache of what had been learned about a path in previous connections. The report also suggests alternative approaches.</t> <t>Although using passive measurements within live traffic (as above) can detect a Classic ECN AQM, it is much harder (perhaps impossible) to determine whether or not the AQM is in a shared queue. Nonetheless, this is much easier using active test trafficout-of-band,out-of-band because two flows can be used. Section 4 of the samereport <xrefreport <xref target="ecn-fallback" format="default"/> describes a simple technique to detect a Classic ECN AQM and determine whether it is in a shared queue, which is summarized here.</t> <t>An L4S-enabled test server could be set up so that, when a test client accesses it, it serves a script that gets the client to open two parallel long-running flows. It could serve one with a Classic congestion control (C, that sets ECT(0)) and one with ascalableScalable CC (L, that sets ECT(1)). If neither flow induces any ECN marks, it can be presumed that the path does not contain a Classic ECN AQM. If either flow induces some ECN marks, the server could measure the relative flow rates andround tripround-trip times of the two flows. <xref target="l4sid_tab_active_AQN_test" format="default"/> shows the AQM that can be inferred for various cases (presumingtheno more types of AQMbehavioursbehaviour than those known at the time of writing).</t> <table anchor="l4sid_tab_active_AQN_test" align="center"><name>Out-of-band testing<name>Out-of-Band Testing withtwo parallel flows. L:=L4S, C:=Classic.</name>Two Parallel Flows</name> <thead> <tr> <th align="left">Rate</th> <th align="left">RTT</th> <th align="left">Inferred AQM</th> </tr> </thead> <tbody> <tr> <td align="left">L > C</td> <td align="left">L = C</td> <td align="left">Classic ECN AQM (FIFO)</td> </tr> <tr> <td align="left">L = C</td> <td align="left">L = C</td> <td align="left">Classic ECN AQM (FQ)</td> </tr> <tr> <td align="left">L = C</td> <td align="left">L < C</td> <td align="left">FQ-L4S AQM</td> </tr> <tr> <td align="left">L ~= C</td> <td align="left">L < C</td> <tdalign="left">Coupled DualQalign="left">DualQ Coupled AQM</td> </tr> </tbody> <tfoot> <tr> <th colspan="3" align="center">L = L4S; C = Classic</th> </tr> </tfoot> </table> <t>Finally, we motivate the recommendation in <xref target="l4sid_congestion_response" format="default"/> that ascalableScalable congestion control is not expected to change to setting ECT(0) while it adapts its behaviour to coexist with Classic flows. This is because the sender needs to continue to check whether it made the right decision-and switch back if it was wrong, or if a different link becomes the bottleneck:</t> <ul spacing="normal"> <li>If, as recommended, the sender changes only its behaviour but not its codepoint to Classic, its codepoint will still be compatible with either an L4S or a Classic AQM. If the bottleneck does actually support both, it will still classify ECT(1) into the same L4S queue, where the sender can measure that switching to Classic behaviour waswrong,wrong so that it can switch back.</li> <li>In contrast, if the sender changes both its behaviour and its codepoint to Classic, even if the bottleneck supports both, it will classify ECT(0) into the Classic queue, reinforcing the sender's incorrect decision so that it never switches back.</li> <li>Also, not changing its codepoint avoids the risk of being flipped to a different path by a load balancer or multipath routing that hashes on the whole of theex-ToSformer Type-of-Service (ToS) byte(unfortunately(which is unfortunately still a common pathology).</li> </ul><t>Note<aside><t>Note that if a flow is configured to <em>only</em> use a Classic congestion control, it is then entirely appropriate not to useECT(1).</t>ECT(1).</t></aside> </section> <section anchor="l4sid_sec_RTT_dependence" numbered="true" toc="default"> <name>Reduce RTTdependence</name>Dependence</name> <t>Description: AscalableScalable congestion control needs to reduce RTT bias as much as possible at least over thelow to typicallow-to-typical range of RTTs that will interact in the intended deployment scenario (see the precise normative requirement wording in <xreftarget="l4sid_congestion_response"target="l4sid_Prague_req-rtt-independence" format="default"/>).</t> <t>Motivation: The throughput of Classic congestion controls is known to be inversely proportional to RTT, so one would expect flows over very low RTT paths to nearly starve flows over larger RTTs. However, Classic congestion controls have never allowed a very low RTT path to exist because they induce a large queue. For instance, consider two paths with base RTT1 ms1 ms and100 ms.100 ms. If a Classic congestion control induces a100 ms100 ms queue, it turns these RTTs into101 ms101 ms and200 ms200 ms, leading to a throughput ratio of about 2:1. Whereas if ascalableScalable congestion control induces only a1 ms1 ms queue, the ratio is 2:101, leading to a throughput ratio of about 50:1.</t> <t>Therefore, with very small queues, long RTT flows will essentially starve, unlessscalableScalable congestion controls comply withthisthe requirement in <xref target="l4sid_congestion_response" format="default"/>.</t> <t>Over higher than typical RTTs, L4S flows can use the same RTT bias as in current Classic congestion controls and still work satisfactorily.So,So there is no additional requirement in <xref target="l4sid_congestion_response" format="default"/> for high RTT L4S flows to remove RTT bias--- theycancan, but they don't have to.</t> <t>One way for a Scalable congestion control to satisfy these requirements is to make its additive increase behave as if it were a standard Reno flow but over a larger RTT by using a virtual RTT (rtt_virt) that is a function of the actual RTT (rtt). Example functions might be:</t> <ulempty="true" spacing="normal">spacing="normal" empty="true"> <li> <tt>rtt_virt = max(rtt,25ms)</tt></li>25 ms)</tt></li> <li> <tt>rtt_virt = rtt +10ms</tt></li>10 ms</tt></li> </ul> <t>These example functions are chosen so that, as the actual RTT reduces from high to low, the virtual RTT reduces less (see <xref target="I-D.briscoe-iccrg-prague-congestion-control" format="default"/> for details).</t> <t>However, short RTT flows can more rapidly respond to changes in available capacity, whether due to other flows arriving and departing or radio capacity varying. So it would be wrong to require short RTT flows to be as sluggish aslong-RTTlong RTT flows, which would unnecessarilyunder-utilizeunderutilize capacity and result in unnecessary overshoots and undershoots (instability). Therefore, rather than requiring strictRTT-independence,RTT independence, the wordingsaysin Item <xref target="l4sid_Prague_req-rtt-independence" format="counter"/> of <xref target="l4sid_congestion_response" format="default"/> is "as independent of RTT as possible without compromising stability or utilization". This allows shorter RTT flows to exploit their agility advantage.</t> </section> <section anchor="l4sid_sec_min_cwnd" numbered="true" toc="default"> <name>ScalingdownDown tofractional congestion windows</name>Fractional Congestion Windows</name> <t>Description: AscalableScalable congestion control needs to remain responsive to congestion when typical RTTs over the public Internet are significantly smaller because they are no longer inflated by queuing delay (see the precise normative requirement wording in <xreftarget="l4sid_congestion_response"target="l4sid_Prague_req-fractional-window" format="default"/>).</t> <t>Motivation: As currently specified, the minimum congestion window of ECN-capable TCP (and its derivatives) is expected to be 2 sender maximum segment sizes (SMSS), or 1 SMSS after a retransmission timeout. Once the congestion window reaches this minimum, if there is furtherECN-marking,ECN marking, TCP is meant to wait for a retransmission timeout before sending another segment (seesection 6.1.2 of the ECN spec <xref<xref target="RFC3168"format="default"/>).sectionFormat="of" section="6.1.2">the ECN spec</xref>). In practice, most known window-based congestion control algorithms become unresponsive to ECN congestion signals at this point. No matter how much ECN marking, the congestion window no longer reduces. Instead, the sender's lack of any further congestion response forces the queue to grow, overriding any AQM and increasing queuing delay (making the window large enough to become responsive again). This can result in a stable but deeper queue, or it might drive the queue to loss,thenin which case the retransmission timeout mechanism acts as a backstop.</t> <t>Most window-based congestion controls for other transport protocols have a similar minimum window, albeit when measured in bytes for those that use smaller packets.</t> <t>L4S mechanisms significantly reducequeueingqueuing delay so, over the same path, the RTT becomes lower.ThenThen, this problem becomes surprisinglycommon <xrefcommon <xref target="sub-mss-prob" format="default"/>. This is because, for the same link capacity, smaller RTT implies a smaller window. For instance, consider a residential setting with an upstream broadband Internet access of8 Mb/s,8 Mb/s, assuming a max segment size of1500 B.1500 B. Two upstream flows will each have the minimum window of2 SMSS2 SMSS if the RTT is6 ms6 ms or less, which is quite common when accessing a nearby data centre.So,So any more than two such parallel TCP flows will become unresponsive to ECN and increase queuing delay.</t> <t>UnlessscalableScalable congestion controls address the requirement in <xref target="l4sid_congestion_response" format="default"/> from the start, they will frequently become unresponsive to ECN, negating thelow latencylow-latency benefit of L4S, for themselves and for others.</t> <t>That would seem to imply thatscalableScalable congestion controllers ought to be required to be able work with a congestion window less than1 SMSS.1 SMSS. For instance, if an ECN-capable TCP gets anECN-markECN mark when it is already sitting at a window of1 SMSS, RFC 31681 SMSS, <xref target="RFC3168" format="default"/> requires it to defer sending for a retransmission timeout. A less drastic but more complex mechanism can maintain a congestion window less than1 SMSS1 SMSS (significantly less if necessary), as described in <xref target="Ahmed19" format="default"/>. Other approaches are likely to be feasible.</t> <t>However, the requirement in <xref target="l4sid_congestion_response" format="default"/> is worded as a"SHOULD""<bcp14>SHOULD</bcp14>" because it is believed that the existence of a minimum window is not all bad. When competing with an unresponsive flow, a minimum window naturally protects the flow from starvation by at least keeping some data flowing.</t> <t>By stating the requirement to go lower than1 SMSS1 SMSS as a"SHOULD","<bcp14>SHOULD</bcp14>", while the requirement inRFC 3168<xref target="RFC3168" format="default"/> still stands as well, we shall be able to watch the choices of minimum window evolve in differentscalableScalable congestion controllers.</t><!-- Requirement #3.5: Smoothing ECN feedback Description: The ratio of marked packets CE marks received by the scalable transport sender are averaged --></section> <section anchor="l4sid_sec_reordering_tolerance" numbered="true" toc="default"> <name>Measuring Reordering Tolerance in Time Units</name> <t>Description: When detecting loss, ascalableScalable congestion control needs to be tolerant to reordering over an adaptive time interval, which scales with throughput, rather than counting only in fixed units of packets, which does not scale (see the precise normative requirement wording in <xreftarget="l4sid_congestion_response"target="l4sid_Prague_req-reordering" format="default"/>).</t> <t>Motivation: A primary purpose of L4S is scalable throughput (it's in the name). Scalability in all dimensions is, of course, also a goal of all IETF technology. The inverse linear congestion response in <xref target="l4sid_congestion_response" format="default"/> is necessary, but not sufficient, to solve the congestion control scalability problem identified in <xref target="RFC3649" format="default"/>. As well as maintaining frequent ECN signals as rate scales, it is also important to ensure that a potentially false perception of loss does not limit throughput scaling.</t><t>End-systems<t>End systems cannot know whether a missing packet is due to loss or reordering, except in hindsight--- if it appears later. So they can only deem that there has been a loss if a gap in the sequence space has not been filled, either after a certain number of subsequent packets has arrived(e.g. the(e.g., the 3 DupACK rule of standard TCP congestioncontrol <xrefcontrol <xref target="RFC5681" format="default"/>) or after a certain amount of time(e.g. the(e.g., the RACKapproach <xrefapproach <xref target="RFC8985" format="default"/>).</t> <t>As we attempt to scale packet rate over the years:</t> <ul spacing="normal"> <li>Even if only <em>some</em> sending hosts still deem that loss has occurred by counting reordered packets, <em>all</em> networks will have to keep reducing the time over which they keep packets in order. If some link technologies keep the time within which reordering occurs roughly unchanged, then loss over these links, as perceived by these hosts, will appear to continually rise over the years.</li> <li>In contrast, if all senders detect loss in units of time, the time over which the network has to keep packets in order stays roughly invariant.</li> </ul> <t>Therefore, hosts have an incentive to detect loss in time units (so as not to fool themselves too often into detecting losses when there are none). And for hosts that are changing their congestion control implementation to L4S, there is no downside to including time-based loss detection code in the change (loss recovery implemented in hardware is an exception, which is covered later). Therefore, requiring L4S hosts to detect loss in time-based units would not be a burden.</t> <t>If the requirement in <xref target="l4sid_congestion_response" format="default"/> were not placed on L4S hosts, even though it would be no burden on hosts to comply, all networks would face unnecessary uncertainty over whether some L4S hosts might be detecting loss by counting packets.ThenThen, <em>all</em> link technologieswillwould have to unnecessarily keep reducing the time within which reordering occurs. That is not a problem for some link technologies, but it becomes increasingly challenging for other link technologies to continue to scale, particularly those relying on channel bonding for scaling, such as LTE,5G5G, andDOCSIS.</t>Data Over Cable Service Interface Specification (DOCSIS).</t> <t>Given Internet paths traverse many link technologies, any scaling limit for these more challenging access link technologies would become a scaling limit for the Internet as a whole.</t> <t>It might be asked how it helps to place this loss detection requirement only on L4S hosts, because networks will still face uncertainty over whether non-L4S flows are detecting loss by counting DupACKs. The answer is that those link technologies for which it is challenging to keep squeezing the reordering time will only need to do so for non-L4S traffic (which they can do because the L4S identifier is visible at the IP layer). Therefore, they can focus their processing and memory resources into scaling non-L4S (Classic) traffic. Then, the higher the proportion of L4S traffic, the less of a scaling challenge they will have.</t> <t>To summarize, there is no reason for L4S hosts not to be part of the solution instead of part of the problem.</t> <t>Requirement("MUST")("<bcp14>MUST</bcp14>") or recommendation("SHOULD")?("<bcp14>SHOULD</bcp14>")? As explained above, this is a subtle interoperability issue between hosts and networks, which seems to need a"MUST"."<bcp14>MUST</bcp14>". Unless networks can be certain that all L4S hosts follow the time-based approach, they still have to cater for the worst case--- continually squeeze reordering into a smaller and smaller duration--- just for hosts that might be using the counting approach. However, it was decided to express this as a recommendation, using"SHOULD"."<bcp14>SHOULD</bcp14>". The main justification was that networks can still be fairly certain that L4S hosts will follow this recommendation, because following it offers only gain and no pain.</t> <t>Details:</t> <t>Thespeed oftime spent recovering a lossrecoveryis much more significant for short flows thanlong, thereforelong; therefore, a good compromise is to adapt the reorderingwindow;window from a small fraction of the RTT at the start of aflow,flow to a larger fraction of the RTT for flows that continue for many round trips.</t> <t>This is broadly the approach adopted byTCPRACK(Recent ACKnowledgements) <xref<xref target="RFC8985" format="default"/>. However, RACK starts with the 3 DupACK approach, because the RTT estimate is not necessarily stable. As long as the initial window is paced, such initial use of 3 DupACK counting would amount to time-based loss detection and therefore would satisfy the time-based loss detection recommendation of <xref target="l4sid_congestion_response" format="default"/>. This is because pacing of the initial window would ensure that 3 DupACKs early in the connection would be spread over a small fraction of the round trip.</t> <t>As mentioned above, hardware implementations of loss recovery using DupACK counting exist(e.g. some(e.g., some implementations ofRoCEv2 for RDMA).Remote Direct Memory Access over Converged Ethernet version 2 (RoCEv2)). For low latency, these implementations can change their congestion control to implement L4S, because the congestion control (as distinct from loss recovery) is implemented in software. But they cannot easily satisfy this loss recovery requirement. However, it is believed they do not need to, because such implementations are believed to solely exist in controlled environments, where the network technology keeps reordering extremely low anyway. This is why controlled environments with hardly any reordering are excluded from the scope of the normative recommendation in <xref target="l4sid_congestion_response" format="default"/>.</t> <t>Detecting loss in time units also prevents the ACK-splitting attacks described in <xref target="Savage-TCP" format="default"/>.</t> </section> </section> <section numbered="true" toc="default"> <name>Scalable Transport Protocol Optimizations</name><t/><section numbered="true" toc="default"> <name>Setting ECT in Control Packets and Retransmissions</name> <t>Description: This item concerns TCP and its derivatives(e.g. SCTP)(e.g., SCTP) as well asRTP/RTCP <xrefRTP/RTCP <xref target="RFC6679" format="default"/>. The original specification of ECN for TCP precluded the use of ECN on control packets and retransmissions. Similarly,RFC 6679<xref target="RFC6679" format="default"/> precludes the use of ECT on RTCP datagrams, in case the path changes after it has been checked for ECN traversal. To improve performance,scalableScalable transport protocols ought to enable ECN at the IP layer in TCP control packets (SYN, SYN-ACK, pure ACKs, etc.) and in retransmitted packets. The same is true for other transports,e.g. SCTP,e.g., SCTP and RTCP.</t> <t>Motivation (TCP):RFC 3168<xref target="RFC3168" format="default"/> prohibits the use of ECN on these types of TCPpacket,packets, based on a number of arguments. This means these packets are not protected from congestion loss by ECN, which considerably harms performance, particularly for short flows.ECN++ <xrefECN++ <xref target="I-D.ietf-tcpm-generalized-ecn" format="default"/> proposes experimental use of ECN on all types of TCPpacketpackets as long as AccECNfeedback <xreffeedback <xref target="I-D.ietf-tcpm-accurate-ecn" format="default"/> is available (which itself satisfies the accurate feedback requirement in <xref target="l4sid_feedback" format="default"/> for using ascalableScalable congestion control).</t> <t>Motivation (RTCP): L4S experiments in general will need to observe the rule in the RTP ECNspec <xrefspec <xref target="RFC6679" format="default"/> that precludes ECT on RTCP datagrams. Nonetheless, as ECN usage becomes more widespread, it would be useful to conduct specific experiments with ECN-capable RTCP to gather data on whether such caution is necessary.</t> </section> <section numbered="true" toc="default"> <name>Faster than Additive Increase</name> <t>Description: It would improve performance ifscalableScalable congestion controls did not limit their congestion window increase to the standard additive increase of1 SMSS1 SMSS per roundtrip <xreftrip <xref target="RFC5681" format="default"/> during congestion avoidance. The same is true for derivatives of TCP congestion control, including similar approaches used for real-time media.</t> <t>Motivation: As currentlydefined <xrefdefined <xref target="RFC8257" format="default"/>, DCTCP uses the conventional Reno additive increase in the congestion avoidance phase. When the available capacity suddenly increases(e.g. when(e.g., when another flowfinishes,finishes or if radio capacity increases) it can take very many round trips to take advantage of the new capacity. TCPCubic <xrefCUBIC <xref target="RFC8312" format="default"/> was designed to solve this problem, but as flow rates have continued to increase, the delay accelerating into available capacity has become prohibitive. See, for instance, the examples in Section5.1<xref target="RFC9330" section="5.1" sectionFormat="bare"/> of the L4Sarchitecture <xref target="I-D.ietf-tsvwg-l4s-arch"architecture <xref target="RFC9330" format="default"/>. Even when out of itsReno-compatibilityReno-friendly mode, every8x8 times scaling ofCubic'sCUBIC's flow rate leads to2x2 times more acceleration delay.</t> <t>In the steady state, DCTCP induces about 2 ECN marks per round trip, so it is possible to quickly detect when these signals have disappeared and seek available capacity more rapidly, while minimizing the impact on other flows (Classic andscalable) <xrefScalable) <xref target="LinuxPacedChirping" format="default"/>. Alternatively, approaches such asAdaptive Acceleration (A2DTCP <xrefAdaptive-Acceleration Data Center TCP (A2DTCP) <xref target="A2DTCP" format="default"/>) have been proposed to address this problem in data centres, which might be deployable over the public Internet.</t> </section> <section numbered="true" toc="default"> <name>Faster Convergence at Flow Start</name> <t>Description: It would improve performance ifscalableScalable congestion controls converged (reached their steady-state share of the capacity) faster than Classic congestion controls or at least no slower. This affects the flow start behaviour of any L4S congestion control derived from a Classic transport that uses TCP slow start, including those for real-time media.</t> <t>Motivation: As an example, a new DCTCP flow takes longer than a Classic congestion control to obtain its share of the capacity of the bottleneck when there are already ongoing flows using the bottleneck capacity. In adata centre environmentdata-centre environment, DCTCP takes abouta factor of1.5 to 2 times longer to converge due to the much higher typical level of ECN marking that DCTCP background traffic induces, which causes new flows to exit slow startearly <xrefearly <xref target="Alizadeh-stability" format="default"/>. In testing for use over the publicInternetInternet, the convergence time of DCTCP relative to a regular loss-based TCP slow start is even lessfavourable <xref target="Paced-Chirping"favourable <xref target="LinuxPacedChirping" format="default"/> due to the shallowECN markingECN-marking threshold needed for L4S. It is exacerbated by the typically greater mismatch between the link rate of the sending host and typical Internet access bottlenecks. This problem is detrimental ingeneral,general but would particularly harm the performance of short flows relative to Classic congestion controls.</t> </section> </section> </section> <section anchor="l4sid_ECT1_CE" numbered="true" toc="default"> <name>Compromises in the Choice of L4S Identifier</name> <t>This appendix is informative, not normative. As explained in <xref target="l4sid_reqs" format="default"/>, there is insufficient space in the IP header (v4 or v6) to fully accommodate every requirement. So the choice of L4S identifier involvestradeoffs.trade-offs. This appendix records the pros and cons of the choice that was made.</t> <t>Non-normative recap of the chosen codepoint scheme:</t> <ulempty="true" spacing="normal">spacing="normal" empty="true"> <li> <t>Packets with ECT(1) and conditionally packets with CE signify L4S semantics as an alternative to the semantics of ClassicECN <xrefECN <xref target="RFC3168" format="default"/>, specifically:</t> <ul spacing="normal"> <li>The ECT(1) codepoint signifies that the packet was sent by an L4S-capable sender.</li> <li>Given the shortage of codepoints, both the L4S and Classic ECN sides of an AQM have to use the same CE codepoint to indicate that a packet has experienced congestion. If a packet that had already been marked CE in an upstream buffer arrived at a subsequent AQM, this AQM would then have to guess whether to classify CE packets as L4S or Classic ECN. Choosing the L4S treatment is a safer choice, because then a few Classic packets might arriveearly,early rather than a few L4S packets arriving late.</li> <li>Additional information might be available if the classifier were transport-aware.ThenThen, it could classify a CE packet for Classic ECN treatment if the most recent ECT packet in the same flow had beenmarkedset to ECT(0). However, the L4S service ought nottoneed transport-layer awareness.</li> </ul> </li> </ul> <t>Cons:</t> <dl newline="false" spacing="normal"> <dt>Consumes the last ECN codepoint:</dt> <dd>The L4S service could potentially supersede the service provided by ClassicECN, thereforeECN; therefore, using ECT(1) to identify L4S packets could ultimately mean that the ECT(0) codepoint was 'wasted' purely to distinguish one form of ECN from its successor.</dd> <dt>ECN hard in some lower layers:</dt> <dd>It is not always possible to support the equivalent of an IP-ECN field in an AQM acting in a buffer below the IPlayer <xreflayer <xref target="I-D.ietf-tsvwg-ecn-encap-guidelines" format="default"/>. Then, depending on thelower layerlower-layer scheme, the L4S service might have to drop rather than mark frames even though they might encapsulate an ECN-capable packet.</dd> <dt>Risk of reordering Classic CE packets within a flow:</dt> <dd anchor="l4sid_CE_reordering"> <t>Classifying all CE packets into the L4S queue risks any CE packets that were originally ECT(0) being incorrectly classified as L4S. If there were delay in the Classic queue, these incorrectly classified CE packets would arrive early, which is a form of reordering. Reordering within a microflow can cause TCP senders (and senders of similar transports) to retransmit spuriously. However, the risk of spurious retransmissions would be extremely low for the following reasons:</t> <ol spacing="normal" type="1"><li>It is quite unusual to experience queuing at more than one bottleneck on the same path (the available capacities have to be identical).</li> <li>In only a subset of these unusual cases would the first bottleneck support Classic ECN markingwhileand the secondsupportedL4S ECNmarking, whichmarking. This would be the only scenario where some ECT(0) packets could be CE marked by an AQM supporting Classic ECNthenwhile subsequently theremainder experiencedremaining ECT(0) packets would experience further delay through the Classic side of a subsequent L4S DualQ AQM.</li> <li> <t>Even then, when a few packets are delivered early, it takes very unusual conditions to cause a spurious retransmission, in contrast to when some packets are delivered late. The first bottleneck has to applyCE-marksCE marks to at least N contiguouspacketspackets, and the second bottleneck has to inject an uninterrupted sequence of at least N of these packets between two packets earlier in the stream (where N is the reordering window that the transport protocol allows before it considers a packet is lost).</t> <ulempty="true" spacing="normal">spacing="normal" empty="true"> <li>Forexampleexample, consider N=3, and consider the sequence of packets 100, 101, 102, 103,...and imagineImagine that packets150,151,152150, 151, 152 from later in the flow are injected as follows: 100, 150, 151, 101, 152, 102,103...103,... If this were late reordering, even one packet arriving out of sequence would trigger a spurious retransmission, but there is no spurious retransmission here with early reordering, because packet 101 moves the cumulative ACK counter forward before 3 packets have arrived out of order. Later, when packets 148, 149,153...153,... arrive, even though there is a 3-packet hole, there will be no problem, because the packets to fill the hole are already in the receive buffer.</li> </ul> </li> <li>Even with the current TCP recommendation ofN=3 <xrefN=3 <xref target="RFC5681"format="default"/>format="default"/>, spurious retransmissions will be unlikely for all the above reasons. AsRACK <xrefRACK <xref target="RFC8985" format="default"/> is becoming widely deployed, it tends to adapt its reordering window to a larger value of N, which will make the chance of a contiguous sequence of N early arrivals vanishingly small.</li> <li>Even a run of 2 CE marks within a Classic ECN flow is unlikely, given FQ-CoDel is the only known widely deployed AQM that supports Classic ECNmarkingmarking, and it takes great care to separate out flows and to space any markings evenly along each flow.</li> </ol> <t>It is extremely unlikely that the above set of 5 eventualities that are each unusual in themselves would all happen simultaneously. But, even if they did, the consequences would hardly be dire: the odd spurious fast retransmission. Whenever the traffic source (a Classic congestion control) mistakes the reordering of a string of CE marks for a loss, one might think that it will reduce its congestion window as well as emitting a spurious retransmission. However, it would have already reduced its congestion window when the CE markings arrived early. If it is usingABE <xrefABE <xref target="RFC8511" format="default"/>, it might reduce cwnd a little more for a loss than for a CE mark. But it will revert that reduction once it detects that the retransmission was spurious.</t> <t>In conclusion, the impact of early reordering on spurious retransmissions due to CE being ambiguous will generally be vanishingly small.</t> </dd> <dt>Insufficient anti-replay window in some pre-existing VPNs:</dt> <dd>If delay is reduced for a subset of the flows within a VPN, the anti-replay feature of some VPNs is known to potentially mistake the difference in delay for a replay attack. <xref target="l4sid_VPN_anti-replay" format="default"/> recommends that the anti-replay window at the VPN egress is sufficiently sized, as required by the relevant specifications. However, in some VPNimplementationsimplementations, the maximum anti-replay window is insufficient to cater for a large delay difference at prevailing packet rates. <xref target="l4sid_VPN_anti-replay" format="default"/> suggests alternative work-rounds for such cases, butend-usersend users using L4S over a VPN will need to be able to recognize the symptoms of this problem, in order to seek out these work-rounds.</dd> <dt>Hard to distinguish Classic ECN AQM:</dt> <dd> <t>With this scheme, when a source receives ECN feedback, it is not explicitly clear which type of AQM generated the CE markings. This is not a problem for Classic ECN sources that send ECT(0) packets, because an L4S AQM will recognize the ECT(0) packets as Classic and apply the appropriate ClassicECN markingECN-marking behaviour.</t> <t>However, in the absence of explicit disambiguation of the CE markings, an L4S source needs to use heuristic techniques to work out which type of congestion response to apply (see <xref target="l4sid_sec_fallback_on_classic_CE" format="default"/>). Otherwise, if long-running Classicflow(s)flows are sharing a Classic ECN AQM bottleneck with long-running L4Sflow(s), which thenflows, and the L4S flows apply an L4S response to the Classic CE signals,the L4S flowsthey would then outcompete the Classicflow(s).flows. Experiments have shown that L4S flows can take about 20 times more capacity share than equivalent Classic flows. Nonetheless, as link capacity reduces(e.g. to(e.g., to 4 Mb/s), the inequality reduces. So Classic flows always make progress and are not starved.</t> <t>When L4S was first proposed (in 2015, 14 years after the Classic ECNspec <xrefspec <xref target="RFC3168" format="default"/> was published), it was believed that Classic ECN AQMs had failed to bedeployed,deployed because research measurements had found little or no evidence of CE marking. In subsequentyearsyears, Classic ECN was included inper-flow-queuing (FQ) deployments, howeverFQ deployments; however, an FQ scheduler stops an L4S flow outcompeting Classic, because it enforces equality between flow rates. It is not known whether there have been any non-FQ deployments of Classic ECN AQMs in the subsequentyears,years or whether there will be any in future.</t> <t>An algorithm for detecting a Classic ECN AQM as soon as a flow stabilizes after start-up has beenproposed <xrefproposed <xref target="ecn-fallback" format="default"/> (see <xref target="l4sid_sec_fallback_on_classic_CE" format="default"/> for a brief summary). Testbed evaluations of v2 of the algorithm have shown detection is reasonably good for Classic ECN AQMs, in a wide range of circumstances. However, although it can correctly detect an L4S ECN AQM in many circumstances,itsit is often incorrect at low link capacities and/or high RTTs. Although this is the safe way round, there is a danger that it will discourage use of the algorithm.</t> </dd> <dt>Non-L4S service for control packets:</dt> <dd>Solely for the case of TCP, the Classic ECNRFCs <xrefRFCs <xref target="RFC3168" format="default"/> and <xref target="RFC5562" format="default"/> require a sender to clear theECNIP-ECN field to Not-ECT on retransmissions and on certain controlpacketspackets, specifically pure ACKs, windowprobesprobes, and SYNs. When L4S packets are classified by theECNIP-ECN field, these TCP control packets would not be classified into an L4Squeue,queue and could therefore be delayed relative to the other packets in the flow. This would not cause reordering (because retransmissions are already out of order, and these control packets typically carry no data). However, it would make critical TCP control packets more vulnerable to loss and delay. To address this problem,ECN++ <xrefECN++ <xref target="I-D.ietf-tcpm-generalized-ecn" format="default"/> proposes an experiment in which all TCP control packets and retransmissions are ECN-capable as long as appropriate ECN feedback is available in each case.</dd> </dl> <t>Pros:</t> <dl newline="false" spacing="normal"> <dt>Should worke2e:</dt>end to end:</dt> <dd>TheECNIP-ECN field generally propagatesend-to-endend to end across the Internet without being wiped or mangled, at least over fixed networks. Unlike the DSCP, the setting of the ECN field is at least meant to be forwarded unchanged by networks that do not support ECN.</dd> <dt>Should work in tunnels:</dt> <dd>The L4S identifiers work across and within any tunnel that propagates theECNIP-ECN field in any of the variant ways it has been defined since ECN-tunneling was first specified in the year2001 <xref2001 <xref target="RFC3168" format="default"/>. However, it is likely that some tunnels still do not implement ECN propagation at all.</dd> <dt>Should work for many link technologies:</dt> <dd>At most, but not all, path bottlenecks there isIP-awareness,IP awareness, so that L4S AQMs can be located where the IP-ECN field can be manipulated. Bottlenecks atlower layerlower-layer nodes withoutIP-awareness eitherIP awareness have to either use drop to signal congestion or have a specific congestion notification facilityhas to bedefined for that link technology, including propagation to and from IP-ECN. The programme to define these isprogressingprogressing, and in each case sofarfar, the scheme already defined for ECN inherently supports L4S as well (see <xref target="l4sid_encaps_no_change" format="default"/>).</dd> <dt>Could migrate to one codepoint:</dt> <dd>If all Classic ECN senders eventually evolve to use the L4S service, the ECT(0) codepoint could be reused for some futurepurpose,purpose but only once use of ECT(0) packetshadhas reduced to zero, ornear-zero,near zero, which might never happen.</dd> <dt>L4 not required:</dt> <dd>Being based on theECNIP-ECN field, this scheme does not need the network to accesstransport layertransport-layer flowidentifiers.IDs. Nonetheless, it does not preclude solutions that do.</dd> </dl> </section> <section numbered="true" toc="default"> <name>Potential Competing Uses for the ECT(1) Codepoint</name> <t>The ECT(1) codepoint of theECNIP-ECN field has already been assigned once for the ECNnonce <xrefnonce spec <xref target="RFC3540" format="default"/>, which has now been categorized ashistoric <xrefHistoric <xref target="RFC8311" format="default"/>. ECN is probably the only remaining field in the Internet Protocol that is common to IPv4 and IPv6 and still has potential to workend-to-end,end to end, with tunnels and with lower layers. Therefore, ECT(1) should not be reassigned to a different experimental use (L4S) without carefully assessing competing potential uses. These fall into thefollowing categories:</t>categories described below.</t> <section anchor="l4sid_competing_integrity" numbered="true" toc="default"> <name>Integrity of Congestion Feedback</name> <t>Receiving hosts can fool a sender into downloading faster by suppressing feedback of ECN marks (or of losses if retransmissions are not necessary or available otherwise).</t> <t>ThehistoricHistoric ECN nonceprotocol <xrefspec <xref target="RFC3540" format="default"/> proposed that a TCP sender could set eitherofECT(0) or ECT(1) in each packet of a flow and remember the sequence it had set. If any packet was lost or congestion marked, the receiver would miss that bit of the sequence. An ECNNoncenonce receiver had to feed back theleast significantleast-significant bit of the sum, so it could not suppress feedback of a loss or mark without a 50-50 chance of guessing the sum incorrectly.</t> <t>It is highly unlikely that ECT(1) will be needed as a nonce for integrity protection of congestion notifications in future. The ECNNonce RFC <xrefnonce spec <xref target="RFC3540" format="default"/> has been reclassified ashistoric,Historic, partly because other ways (that do not consume a codepoint in the IP header) have been developed to protect feedback integrity of TCP and othertransports <xreftransports <xref target="RFC8311" format="default"/>. For instance:</t> <ul spacing="normal"><li>the<li>The sender can test the integrity of a small random sample of the receiver's feedback by occasionally setting the IP-ECN field to a value normally only set by the network.ThenThen, it can test whether the receiver's feedback faithfully reports what it expects (seeparaParagraph 2 ofSection 20.2 of the ECN spec <xref<xref target="RFC3168"format="default"/>.sectionFormat="of" section="20.2">the ECN spec</xref>. This works forlossloss, and it will work for the accurate ECNfeedback <xreffeedback <xref target="RFC7560" format="default"/> intended for L4S. Like the(historic)(Historic) ECNnonce,nonce spec, this technique does not protect against a misbehaving sender. But it allows a well-behaved sender to check that each receiver is correctly feeding back congestion notifications.</li> <li>A network can check that its ECN markings (or packet losses) have been passed correctlyroundaround the full feedback loop by auditingcongestion exposure (ConEx) <xrefCongestion Exposure (ConEx) <xref target="RFC7713" format="default"/>. This assures that the integrity of congestion notifications and feedback messages must have both been preserved. ConEx information is also available anywhere along the network path, so it can be used to enforce a congestion response. Whether the receiver or a downstream network is suppressing congestion feedback or the sender is unresponsive to the feedback, or both, ConEx is intended toneutraliseneutralize any advantage that any of these three parties would otherwise gain.</li> <li>Congestion feedback fields intransport layertransport-layer headers are immutableend-to-end,end to end and therefore amenable to end-to-end integrity protection. This preserves the integrity of a receiver's feedback messages to the sender, but it does not protect against misbehaving receivers or misbehaving senders. The TCPauthentication option (TCP-AO <xrefAuthentication Option (TCP-AO) <xref target="RFC5925"format="default"/>),format="default"/>, QUIC's end-to-end protection <xref target="RFC9001"format="default"/>format="default"/>, or end-to-end IPsec integrity protection <xref target="RFC4303" format="default"/> can be used to detect any tampering with congestion feedback (whether malicious oraccidental). respectivelyaccidental), respectively, in TCP,QUICQUIC, or any transport. TCP-AO covers the main TCP header and TCP options by default, but it is often too brittle to use on many end-to-end paths, where middleboxes can make verification fail in their attempts to improve performance or security,e.g. bye.g., by resegmentation or shifting the sequence space.</li> </ul> <t>At the time of writing,Itit is becoming common to protect the integrity of transport feedback using QUIC. However, it is still not common to protect the integrity of the wider congestionfeedback,feedback loop, whether based on loss or Classic ECN. If this position changes during the L4S experiment, one or more of the above techniques might need to be developed and deployed.</t> </section> <section numbered="true" toc="default"> <name>Notification of Less Severe Congestion than CE</name> <t>Various researchers have proposed to use ECT(1) as a less severe congestion notification than CE, particularly to enable flows to fill available capacity more quickly after an idle period, when another flow departs or when a flow starts,e.g. VCP <xrefe.g., the Variable-structure congestion Control Protocol (VCP) <xref target="VCP"format="default"/>,format="default"/> and Queue View(QV) <xref(QV) <xref target="QV" format="default"/>.</t> <t>Before assigning ECT(1) as an identifier for L4S, we must carefully consider whether it might be better to hold ECT(1) in reserve for futurestandardisationstandardization of rapid flow acceleration, which is an important and enduringproblem <xrefproblem <xref target="RFC6077" format="default"/>.</t> <t>Pre-Congestion Notification (PCN) is another scheme that assigns alternative semantics to theECNIP-ECN field. It uses ECT(1) to signify a less severe level of pre-congestion notification thanCE <xrefCE <xref target="RFC6660" format="default"/>. However, theECNIP-ECN field only takes on the PCN semantics if packets carry a Diffserv codepoint defined to indicate PCN marking within a controlled environment. PCN is required to be applied solely to the outer header of a tunnel across the controlled region in order not to interfere with any end-to-end use of the ECN field. Therefore, a PCN region on the path would not interfere with the L4S service identifier defined in <xref target="l4sid_identification" format="default"/>.</t> </section> </section><!-- <section title="Change Log (to be Deleted before Publication)"> <t>A detailed version history can be accessed at <http://datatracker.ietf.org/doc/draft-briscoe-aqm-ecn-roadmap/history/></t> <t><list style="hanging"> <t hangText="From briscoe-...-00 to briscoe-...-01:">Technical changes:<list style="symbols"> <t/> </list>Editorial changes:<list style="symbols"> <t/> </list></t> </list></t> </section> --><section numbered="false" toc="default"> <name>Acknowledgements</name> <t>Thanks toRichard Scheffenegger, John Leslie, David Taeht, Jonathan Morton, Gorry Fairhurst, Michael Welzl, Mikael Abrahamsson and Andrew McGregor<contact fullname="Richard Scheffenegger"/>, <contact fullname="John Leslie"/>, <contact fullname="David Täht"/>, <contact fullname="Jonathan Morton"/>, <contact fullname="Gorry Fairhurst"/>, <contact fullname="Michael Welzl"/>, <contact fullname="Mikael Abrahamsson"/>, and <contact fullname="Andrew McGregor"/> for the discussions that led to this specification.Ing-jyh<contact fullname="Ing-jyh (Inton)TsangTsang"/> was a contributor to the earlydraftsdraft versions of this document.And thanksThanks toMikael Abrahamsson, Lloyd Wood, Nicolas Kuhn, Greg White, Tom Henderson, David Black, Gorry Fairhurst, Brian Carpenter, Jake Holland, Rod Grimes, Richard Scheffenegger, Sebastian Moeller, Neal Cardwell, Praveen Balasubramanian, Reza<contact fullname="Mikael Abrahamsson"/>, <contact fullname="Lloyd Wood"/>, <contact fullname="Nicolas Kuhn"/>, <contact fullname="Greg White"/>, <contact fullname="Tom Henderson"/>, <contact fullname="David Black"/>, <contact fullname="Gorry Fairhurst"/>, <contact fullname="Brian Carpenter"/>, <contact fullname="Jake Holland"/>, <contact fullname="Rod Grimes"/>, <contact fullname="Richard Scheffenegger"/>, <contact fullname="Sebastian Moeller"/>, <contact fullname="Neal Cardwell"/>, <contact fullname="Praveen Balasubramanian"/>, <contact fullname="Reza MarandianHagh, Pete Heist, Stuart Cheshire, Vidhi Goel, Mirja Kuehlewind, Ermin Sakic and Martin DukeHagh"/>, <contact fullname="Pete Heist"/>, <contact fullname="Stuart Cheshire"/>, <contact fullname="Vidhi Goel"/>, <contact fullname="Mirja Kühlewind"/>, <contact fullname="Ermin Sakic"/>, and <contact fullname="Martin Duke"/> for providing help and reviewing thisdraft, anddocument. And thanks toIngemar Johansson<contact fullname="Ingemar Johansson"/> for reviewing and providing substantial text. Thanks also to the area reviewers:Valery Smyslov, Maria<contact fullname="Valery Smyslov"/>, <contact fullname="Maria InesRobles, Bernard Aboba, Lars Eggert, Roman Danyliw and Eric Vyncke.Robles"/>, <contact fullname="Bernard Aboba"/>, <contact fullname="Lars Eggert"/>, <contact fullname="Roman Danyliw"/>, and <contact fullname="Éric Vyncke"/>. Thanks toSebastian Moeller<contact fullname="Sebastian Moeller"/> for identifying the interaction with VPN anti-replay and toJonathan Morton<contact fullname="Jonathan Morton"/> for identifying the attack based on this. Particular thanks to tsvwg chairsGorry Fairhurst, David Black<contact fullname="Gorry Fairhurst"/>, <contact fullname="David Black"/>, andWes Eddy<contact fullname="Wes Eddy"/> for patiently helping this and the other L4Sdraftsdocuments through the IETF process. <xref target="l4sps_tcp-prague-reqs"format="default"/> listingformat="default"/>, which lists the Prague L4SRequirementsRequirements, is based on text authored byMarcelo<contact fullname="Marcelo BagnuloBraunBraun"/> that was originally an appendix to <xreftarget="I-D.ietf-tsvwg-l4s-arch"target="RFC9330" format="default"/>. That text was in turn based on the collective output of the attendees listed in the minutes of a 'bar BoF' on DCTCP Evolution duringIETF-94 <xrefIETF 94 <xref target="TCPPrague" format="default"/>.</t> <t>The authors' contributions werepart-fundedpartly funded by the European Community under its Seventh Framework Programme through the Reducing Internet Transport Latency (RITE) project (ICT-317700). The contribution ofKoen<contact fullname="Koen DeSchepperSchepper"/> was alsopart-fundedpartly funded by the 5Growth and DAEMON EU H2020 projects.Bob Briscoe<contact fullname="Bob Briscoe"/> wasalso fundedpartly funded by the Research Council of Norway through the TimeIn project,partly by CableLabsCableLabs, andpartly bythe Comcast Innovation Fund. The views expressed here are solely those of the authors.</t> </section> </back> </rfc>