<?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="info" consensus="true" docName="draft-ietf-tsvwg-l4s-arch-20" number="9330" 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 --><title abbrev="L4S Architecture">Low Latency, Low Loss, and Scalable Throughput (L4S) Internet Service: Architecture</title> <seriesInfoname="Internet-Draft" value="draft-ietf-tsvwg-l4s-arch-20"/>name="RFC" value="9330"/> <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="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="Marcelo Bagnulo" initials="M."surname="Bagnulo Braun">surname="Bagnulo"> <organization>Universidad Carlos III de Madrid</organization> <address> <postal> <street>Av. Universidad 30</street><city>Leganes, Madrid 28911</city><city>Madrid</city> <code>28911</code> <country>Spain</country> </postal> <phone>34 91 6249500</phone> <email>marcelo@it.uc3m.es</email> <uri>https://www.it.uc3m.es</uri> </address> </author> <author fullname="Greg White" initials="G." surname="White"> <organization>CableLabs</organization> <address> <postal> <street/><country>US</country><country>United States of America</country> </postal> <email>G.White@CableLabs.com</email> </address> </author> <datemonth="" year=""/> <area>Transport</area> <workgroup>Transport Area Working Group</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>Pacing</keyword> <keyword>Burstiness</keyword> <abstract> <t>This document describes the L4S architecture, which enables Internet applications to achieveLowlow queuingLatency, Low Loss,latency, low congestion loss, andScalablescalable throughput(L4S).control. L4S is based on the insight that the root cause of queuing delay is in the capacity-seeking congestion controllers of senders, not in the queue itself. With the L4Sarchitecturearchitecture, all Internet applications could (but do not have to) transition away from congestion control algorithms that cause substantial queuingdelay, todelay and instead adopt a new class of congestion controls that can seek capacity with very little queuing. These are aided by a modified form ofexplicit congestion notificationExplicit Congestion Notification (ECN) from the network. With this new architecture, applications can have both low latency and high throughput.</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. The aim is for L4S latency and throughput to be usually much better (and rarelyworse),worse) while typically not impacting Classic performance.</t> </abstract> </front> <middle> <section anchor="l4sps_intro" numbered="true" toc="default"> <name>Introduction</name> <t>At any one time, it is increasingly common for all of the traffic in a bottleneck link(e.g. a(e.g., a household's Internetaccess)access or Wi-Fi) to come from applications that prefer low delay: interactiveWeb, Webweb, web services, voice, conversational video, interactive video, interactive remote presence, instant messaging, online and cloud-rendered gaming, remote desktop, cloud-basedapplicationsapplications, cloud-rendered virtual reality or augmented reality, and video-assisted remote control of machinery and industrial processes. In the last decade or so, much has been done to reduce propagation delay by placing caches or servers closer to users. However, queuing remains a major, albeit intermittent, component of latency. Forinstanceinstance, spikes of hundreds of milliseconds are not uncommon, even with state-of-the-artactive queue management (AQM) <xrefActive Queue Management (AQM) <xref target="COBALT"format="default"/>,format="default"/> <xref target="DOCSIS3AQM" format="default"/>.QueuingA Classic AQM in an access networkbottlenecksbottleneck is typically configured to buffer the sawteeth of lone flows, which can cause peak overall network delay to roughly double during a long-running flow, relative to expected base (unloaded) pathdelay <xrefdelay <xref target="BufferSize" format="default"/>. Low loss is also important because, for interactive applications, losses translate into even longer retransmission delays.</t> <t>It has been demonstrated that, once access network bit rates reach levels now common in the developed world, increasing link capacity offers diminishing returns if latency (delay) is not addressed <xref target="Dukkipati06"format="default"/>,format="default"/> <xref target="Rajiullah15" format="default"/>. Therefore, the goal is an Internet service with veryLow queueing Latency,low queuing latency, veryLow Losslow loss, andScalable throughput (L4S).scalable throughput. Very low queuing latency means less than1 millisecond1 millisecond (ms) on average and less than about2 ms2 ms at the 99th percentile. End-to-end delay above50 ms <xref50 ms <xref target="Raaen14"format="default"/>format="default"/>, or even above20 ms <xref20 ms <xref target="NASA04"format="default"/>format="default"/>, starts to feel unnatural for more demanding interactive applications.SoTherefore, removing unnecessary delay variability increases the reach of these applications (the distance over which they are comfortable touse).use) and/or provides additional latency budget that can be used for enhanced processing. This document describes the L4S architecture for achieving these goals.</t> <t>Differentiated services (Diffserv) offers Expedited Forwarding(EF <xref(EF) <xref target="RFC3246"format="default"/>)format="default"/> for some packets at the expense of others, but this makes no difference when all (or most) of the traffic at a bottleneck at any one time requires low latency. In contrast, L4S still works well when all traffic is L4S--- a service that gives without taking needs none of the configuration or management baggage (trafficpolicing,policing or traffic contracts) associated with favouring some traffic flows over others.</t> <t>Queuing delay degrades performanceintermittently <xrefintermittently <xref target="Hohlfeld14" format="default"/>. It occurs i) when a large enough capacity-seeking(e.g. TCP)(e.g., TCP) flow is running alongside the user's traffic in the bottleneck link, which is typically in the accessnetwork. Ornetwork, or ii) when the low latency application is itself a large capacity-seeking or adaptive rate(e.g. interactive video) flow.flow (e.g., interactive video). At these times, the performance improvement from L4S must be sufficientthatfor network operatorswillto be motivated to deploy it.</t> <t>Active Queue Management (AQM) is part of the solution to queuing under load. AQM improves performance for all traffic, but there is a limit to how much queuing delay can be reduced by solely changing thenetwork;network without addressing the root of the problem.</t> <t>The root of the problem is the presence of standard congestion control(Reno <xref(Reno <xref target="RFC5681" format="default"/>) or compatible variants(e.g. CUBIC <xref(e.g., CUBIC <xref target="RFC8312" format="default"/>) that are used in TCP and in othertransportstransports, such as QUIC <xref target="RFC9000" format="default"/>. We shall use the term 'Classic' for these Reno-friendly congestion controls. Classic congestion controls induce relatively largesaw-tooth-shapedsawtooth-shaped excursionsup theof queueand down again, which have been growing as flow rate scales <xref target="RFC3649" format="default"/>.occupancy. So if a network operator naively attempts to reduce queuing delay by configuring an AQM to operate at a shallower queue, a Classic congestion control will significantly underutilize the link at the bottom of everysaw-tooth.</t>sawtooth. These sawteeth have also been growing in duration as flow rate scales (see <xref target="l4sps_why_primary_components" format="default"/> and <xref target="RFC3649" format="default"/>).</t> <t>It has been demonstratedthatthat, if the sending host replaces a Classic congestion control with a 'Scalable' alternative,when a suitable AQM is deployed in the networkthe performance under load of all the above interactive applications can be significantlyimproved. For instance, queuing delay under heavy load withimproved once a suitable AQM is deployed in the network. Taking the exampleDCTCP/DualQsolution cited below that uses Data Center TCP (DCTCP) <xref target="RFC8257" format="default"/> and a Dual-Queue Coupled AQM <xref target="RFC9332" format="default"/> on a DSL or Ethernetlinklink, queuing delay under heavy load is roughly1 to 2 milliseconds1-2 ms at the 99th percentile without losing link utilization <xreftarget="DualPI2Linux" format="default"/>,target="L4Seval22" format="default"/> <xreftarget="DCttH19"target="DualPI2Linux" format="default"/> (for other link types, see <xref target="l4sarch_link-specifics" format="default"/>). This compares with5-20 ms5-20 ms on <em>average</em> with a Classic congestion control and current state-of-the-artAQMsAQMs, such asFQ-CoDel <xrefFlow Queue CoDel <xref target="RFC8290" format="default"/>,PIE <xrefProportional Integral controller Enhanced (PIE) <xref target="RFC8033"format="default"/>format="default"/>, or DOCSISPIE <xrefPIE <xref target="RFC8034" format="default"/> and about20-30 ms20-30 ms at the 99thpercentile <xrefpercentile <xref target="DualPI2Linux" format="default"/>.</t> <t>L4S is designed for incremental deployment. It is possible to deploy the L4S service at a bottleneck link alongside the existing best effortsservice <xrefservice <xref target="DualPI2Linux" format="default"/> so that unmodified applications can start using it as soon as the sender's stack is updated. Access networks are typically designed with one link as the bottleneck for each site (which might be a home, smallenterpriseenterprise, or mobile device), so deployment at either or both ends of this link should give nearly all the benefit in the respective direction. With some transport protocols, namely TCPand SCTP,<xref target="I-D.ietf-tcpm-accurate-ecn" format="default"/>, the sender has to check that the receiver has been suitably updated to give more accurate feedback, whereas with more recent transportprotocolsprotocols, such as QUIC <xref target="RFC9000" format="default"/> andDCCP,Datagram Congestion Control Protocol (DCCP) <xref target="RFC4340" format="default"/>, all receivers have always been suitable.</t> <t>This document presents the L4S architecture. It consists of three components: network support to isolate L4S traffic fromclassicClassic traffic; protocol features that allow network elements to identify L4S traffic; and host support for L4S congestion controls. The protocol is defined separately in <xreftarget="I-D.ietf-tsvwg-ecn-l4s-id"target="RFC9331" format="default"/> as an experimental change to Explicit Congestion Notification (ECN). This document describes and justifies the component parts and how they interact to provide thescalable,low latency, lowlossloss, and scalable Internet service. It also details the approach to incremental deployment, as briefly summarized above.</t> <section numbered="true" toc="default"> <name>Document Roadmap</name> <t>This document describes the L4S architecture in three passes.First thisFirst, the brief overview in <xref target="l4s-arch_arch_overview" format="default"/> gives the veryhigh levelhigh-level idea and states the main components with minimal rationale. This is only intended to give some context for the terminology definitions that follow in <xref target="l4sps_Terminology"format="default"/>,format="default"/> and to explain the structure of the rest of the document.ThenThen, <xref target="l4sps_components" format="default"/> goes into more detail on each component with somerationale,rationale but still mostly stating what the architecture is, rather than why. Finally, <xref target="l4sps_rationale" format="default"/> justifies why each element of the solution was chosen (<xref target="l4sps_why_primary_components" format="default"/>) and why these choices were different from other solutions (<xref target="l4sps_why-not" format="default"/>).</t><t>Having described<t>After thearchitecture,architecture has been described, <xref target="l4sarch_applicability" format="default"/> clarifies itsapplicability; that is,applicability by describing the applications anduse-casesuse cases that motivated the design, the challenges applying the architecture to various link technologies, and various incremental deploymentmodels: includingmodels (including the two main deployment topologies, different sequences for incrementaldeploymentdeployment, and various interactions withpre-existing approaches.preexisting approaches). The document ends with the usual tailpieces, including extensive discussion of traffic policing and other security considerations in <xref target="l4sps_Security_Considerations" format="default"/>.</t> </section> </section> <section anchor="l4s-arch_arch_overview" numbered="true" toc="default"> <name>L4S Architecture Overview</name><t>Below<t>Below, we outline the three main components to the L4Sarchitecture;architecture: 1) thescalableScalable congestion control on the sending host; 2) the AQM at the network bottleneck; and 3) the protocol between them.</t> <t>But first, the main point to grasp is that low latency is not provided by thenetwork -network; low latency results from the careful behaviour of thescalableScalable congestion controllers used by L4S senders. The network does have arole -role, primarily to isolate the low latency of the carefully behaving L4S traffic from the higher queuing delay needed by traffic withpre-existingpreexisting Classic behaviour. The network also alters the way it signals queue growth to thetransport -transport. It uses the Explicit Congestion Notification (ECN) protocol, but it signals the very start of queue growth- immediatelyimmediately, without the smoothing delay typical of Classic AQMs. Because ECN support is essential for L4S, senders use the ECN field as the protocol that allows the network to identify which packets are L4S and which are Classic.</t><dl newline="false" spacing="normal"> <dt>1) Host:</dt> <dd><ol spacing="normal" type="%d)"> <li><t>Host:</t> <t>Scalable congestion controls already exist. They solve the scaling problem with Classic congestion controls, such as Reno orCubic.CUBIC. Because flow rate has scaled since TCP congestion control was first designed in 1988, assuming the flow lasts long enough, it now takes hundreds of round trips (and growing) to recover after a congestion signal (whether a loss or an ECNmark)mark), as shown in the examples in <xref target="l4sps_why_primary_components" format="default"/> and <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.</t> <t>With ascalableScalable congestion control, 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 andutilizationutilization, whatever the flow rate, as well as ensuring that high throughput is more robust to disturbances. ThescalableScalable control used most widely (in controlled environments) isData Center TCP (DCTCP <xrefDCTCP <xref target="RFC8257"format="default"/>),format="default"/>, which has been implemented and deployed in Windows Server Editions (since 2012), inLinuxLinux, and in FreeBSD. Although DCTCP as-is functions well over wide-arearound trip times,round-trip times (RTTs), most implementations lack certain safety features that would be necessary for use outside controlledenvironmentsenvironments, like data centres (see <xref target="l4sarch_sec_non-l4s-neck" format="default"/>).So scalableTherefore, Scalable congestion control needs to be implemented in TCP and other transport protocols (QUIC,SCTP,Stream Control Transmission Protocol (SCTP), RTP/RTCP,RMCAT,RTP Media Congestion Avoidance Techniques (RMCAT), etc.). Indeed, between the present document being drafted and published, the followingscalableScalable congestion controls were implemented: Prague over TCPPrague <xrefand QUIC <xref target="I-D.briscoe-iccrg-prague-congestion-control" format="default"/> <xref target="PragueLinux" format="default"/>,QUIC Prague,an L4S variant of the RMCAT SCReAMcontroller <xref target="SCReAM" format="default"/>controller <xref target="SCReAM-L4S" format="default"/>, and the L4S ECN part ofBBRv2 <xrefBottleneck Bandwidth and Round-trip propagation time (BBRv2) <xref target="BBRv2" format="default"/> intended for TCP and QUIC transports.</t></dd> <dt>2) Network:</dt> <dd></li> <li><t>Network:</t> <t>L4S traffic needs to be isolated from the queuing latency of Classic traffic. One queue per application flow (FQ) is one way to achieve this,e.g. FQ-CoDel <xrefe.g., FQ-CoDel <xref target="RFC8290" format="default"/>. However, using just two queues is sufficient and does not require inspection of transport layer headers in the network, which is not always possible (see <xref target="l4sps_why-not" format="default"/>). With just two queues, it might seem impossible to know how much capacity to schedule for each queue without inspecting how many flows at any one time are using each. And it would be undesirable to arbitrarily divide access network capacity into two partitions. TheDual QueueDual-Queue Coupled AQM was developed as a minimal complexity solution to this problem. It acts like a 'semi-permeable' membrane that partitions latency but not bandwidth. As such, the two queues are fortransitiontransitioning from Classic to L4S behaviour, not bandwidth prioritization.</t> <t><xref target="l4sps_components" format="default"/> gives ahigh levelhigh-level explanation of how theper-flow-queueper-flow queue (FQ) and DualQ variants of L4S work, and <xreftarget="I-D.ietf-tsvwg-aqm-dualq-coupled"target="RFC9332" format="default"/> gives a full explanation of the DualQ Coupled AQM framework. A specific marking algorithm is not mandated for L4S AQMs. Appendices of <xreftarget="I-D.ietf-tsvwg-aqm-dualq-coupled"target="RFC9332" format="default"/> give non-normative examples that have been implemented andevaluated,evaluated and give recommended default parameter settings. It is expected that L4S experiments will improve knowledge of parameter settings and whether the set of marking algorithms needs to belimited.<!--{ToDo: Add ref to Mohit's draft re L4S FQ, once available.}-->limited. </t></dd> <dt>3) Protocol:</dt> <dd>A</li> <li><t>Protocol:</t> <t>A sending host needs to distinguish L4S and Classic packets with an identifier so that the network can classify them into their separate treatments. The L4S identifierspec. <xref target="I-D.ietf-tsvwg-ecn-l4s-id"spec <xref target="RFC9331" format="default"/> concludes that all alternatives involve compromises, but the ECT(1) andCECongestion Experienced (CE) codepoints of the ECN field represent a workable solution. As already explained, the network also uses ECN to immediately signal the very start of queue growth to thetransport.</dd> </dl>transport.</t> </li> </ol> </section> <section anchor="l4sps_Terminology" numbered="true" toc="default"> <name>Terminology</name><t>[Note to the RFC Editor (to be removed before publication as an RFC): The following definitions are copied from the L4S ECN spec <xref target="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/> for the reader's convenience. Except, here, Classic CC and Scalable CC are condensed because they refer to <xref target="l4sps_why_primary_components" format="default"/> later. Also the definition of Traffic Policing is not needed in <xref target="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/>.]</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"/>. The scaling problem with Classic congestion control is explained, with examples, in <xref target="l4sps_why_primary_components" format="default"/> and in <xref target="RFC3649" format="default"/>.</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. 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 <xref target="Mathis09"e.g., Relentless TCP <xref target="I-D.mathis-iccrg-relentless-tcp" format="default"/>, Prague for TCPPrague <xrefand QUIC <xref target="I-D.briscoe-iccrg-prague-congestion-control"format="default"/>, <xrefformat="default"/> <xref target="PragueLinux" format="default"/>, 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 of SCReAM for real-timemedia <xref target="SCReAM" format="default"/>, <xrefmedia <xref target="SCReAM-L4S" format="default"/> <xref target="RFC8298"format="default"/>).format="default"/>. SeeSection 4.3 of<xreftarget="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/>target="RFC9331" format="default" sectionFormat="of" section="4.3"/> 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 congestioncontrol <xrefcontrol <xref target="I-D.briscoe-iccrg-prague-congestion-control" format="default"/>, which was derived from DCTCP<xref<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).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 enough to 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 Reno is used in non-TCPtransportstransports, such asQUIC <xrefQUIC <xref target="RFC9000" format="default"/>.</dd> <dt>Classic ECN:</dt> <dd> <t>The original Explicit Congestion Notification (ECN)protocol <xrefprotocol <xref target="RFC3168"format="default"/>, whichformat="default"/> that requires ECN signals to be treated as equivalent to drops, both when generated in the network and when responded to by the sender.</t><t>L4S uses the ECN field as an identifier <xref target="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/> with<t>For L4S, the names used for the four codepoints of the 2-bit IP-ECN field are unchanged from those defined in the ECNspec <xrefspec <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 ECN 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> <dt>Trafficpolicing:</dt>Policing:</dt> <dd>Limiting traffic by dropping packets or shifting them to a lower service class (as opposed to introducing delay, which is termedtraffic shaping).'traffic shaping'). Policing can involve limiting the average rate and/or burst size. Policing focused on limiting queuing but not the average flow rate is termedcongestion policing, latency policing, burst policing'congestion policing', 'latency policing', 'burst policing', orqueue protection'queue protection' in this document. Otherwise, the term rate policing is used.</dd> </dl> </section> <section anchor="l4sps_components" numbered="true" toc="default"> <name>L4S Architecture Components</name> <t>The L4S architecture is composed of the elements in the following three subsections.</t> <section anchor="l4sps_protocol_components" numbered="true" toc="default"> <name>Protocol Mechanisms</name> <t>The L4S architecture involves: a) unassignment of the previous use of the identifier; b) reassignment of the same identifier; and c) optional further identifiers:</t> <ol spacing="normal" type="a"><li> <t>An essential aspect of ascalableScalable congestion control is the use of explicit congestion signals.'Classic' ECN <xrefClassic ECN <xref target="RFC3168" format="default"/> requires an ECN signal to be treated as equivalent to drop, both when it is generated in the network and when it is responded to by hosts. L4S needs networks and hosts to support a more fine-grained meaning for each ECN signal that is less severe than a drop, so that the L4S signals:</t> <ul spacing="normal"> <li>can be much morefrequent;</li>frequent and</li> <li>can be signalled immediately, without the significant delay required to smooth out fluctuations in the queue.</li> </ul> <t>To enable L4S, thestandards trackStandards Track Classic ECNspec. <xrefspec <xref target="RFC3168" format="default"/> has had to be updated to allow L4S packets to depart from the'equivalent to drop''equivalent-to-drop' constraint. <xref target="RFC8311" format="default"/> is astandards trackStandards Track update to relax specific requirements inRFC 3168<xref target="RFC3168" format="default"/> (and certain otherstandards trackStandards Track RFCs), which clears the way for the experimental changes proposed for L4S. Also, the ECT(1) codepoint was previously assigned as the experimental ECNnonce <xrefnonce <xref target="RFC3540" format="default"/>, whichRFC 8311<xref target="RFC8311" format="default"/> recategorizes as historic to make the codepoint available again.</t> </li> <li> <t><xreftarget="I-D.ietf-tsvwg-ecn-l4s-id"target="RFC9331" format="default"/> specifies that ECT(1) is used as the identifier to classify L4S packets into a separate treatment from Classic packets. This satisfies the requirement for identifying an alternative ECN treatment in <xref target="RFC4774" format="default"/>.</t> <t>The CE codepoint is used to indicate Congestion Experienced by both L4S and Classic treatments. This raises the concern that a Classic AQM earlier on the path might have marked some ECT(0) packets as CE.ThenThen, these packets will be erroneously classified into the L4S queue.Appendix B of the L4S ECN spec <xref target="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/><xref target="RFC9331" format="default" section="B" sectionFormat="of"/> explains why five unlikely eventualities all have to coincide for this to have any detrimental effect, which even then would only involve a vanishingly small likelihood of a spurious retransmission.</t> </li> <li>A network operator might wish to include certain unresponsive, non-L4S traffic in the L4S queue if it is deemed to be paced smoothly enoughpacedand at a low enough rate not to build aqueue. Forqueue, for instance, VoIP, low rate datagrams to sync online games, relatively low rate application-limited traffic, DNS,LDAP,Lightweight Directory Access Protocol (LDAP), etc. This traffic would need to be tagged with specific identifiers,e.g. a low latencye.g., a low-latency DiffservCodepointcodepoint such as Expedited Forwarding(EF <xref(EF) <xref target="RFC3246"format="default"/>),format="default"/>, Non-Queue-Building(NQB <xref(NQB) <xref target="I-D.ietf-tsvwg-nqb"format="default"/>),format="default"/>, or operator-specific identifiers.</li> </ol> </section> <section anchor="l4sps_network_components" numbered="true" toc="default"> <name>Network Components</name> <t>The L4S architecture aims to provide low latency without the <em>need</em> for per-flow operations in network components. Nonetheless, the architecture does not preclude per-flow solutions. The following bullets describe the known arrangements: a) the DualQ Coupled AQM with an L4S AQM in one queue coupled from a Classic AQM in the other; b)Per-Flow Queuesper-flow queues with an instance of a Classic and an L4S AQM in each queue; and c) Dual queues with per-flowAQMs,AQMs but no per-flow queues:</t> <ol spacing="normal" type="a"><li> <t>TheDual QueueDual-Queue Coupled AQM (illustrated in <xref target="l4sps_fig_components" format="default"/>) achieves the 'semi-permeable' membrane property mentioned earlier as follows:</t> <ul spacing="normal"> <li>Latency isolation: Two separate queues are used to isolate L4S queuing delay from the larger queue that Classic traffic needs to maintain fullutilization. <!--Each has its own AQM with the L4S AQM configured for a very shallow (sub-millisecond) target delay, and the Classic AQM for 5-15 ms, which is needed to absorb the larger saw-toothing pattern of Classic congestion controls (otherwise they under-utilize the link). --> </li>utilization.</li> <li>Bandwidth pooling: The two queues act as if they are a single pool of bandwidth in which flows of either type get roughly equal throughput without the scheduler needing to identify any flows. This is achieved by having an AQM in each queue, but the Classic AQM provides a congestion signal to both queues in a manner that ensures a consistent response from the two classes of congestion control. Specifically, the Classic AQM generates a drop/mark probability based on congestion in its own queue, which it uses both to drop/mark packets in its own queue and to affect the marking probability in the L4S queue. The strength of the coupling of the congestion signalling between the two queues is enough to make the L4S flows slow down to leave the right amount of capacity for the Classic flows (as they would if they were the same type of traffic sharing the same queue).</li> </ul><t>Then<t>Then, the scheduler can serve the L4S queue with priority (denoted by the '1' on the higher priority input), because the L4S traffic isn't offering up enough traffic to use all the priority that it is given. Therefore:</t> <ul spacing="normal"> <li>for latency isolation on shorttime-scales (sub-round-trip)timescales (sub-round-trip), the prioritization of the L4S queue protects its low latency by allowing bursts to dissipate quickly;</li> <li>but for bandwidth pooling on longertime-scalestimescales (round-trip andlonger)longer), the Classic queue creates an equal and opposite pressure against the L4S traffic to ensure that neither has priority when it comes to bandwidth--- the tension between prioritizing L4S and coupling the marking from the Classic AQM results in approximate per-flow fairness.</li> </ul> <t>To protect againstunresponsive traffic taking advantage ofthe prioritization ofthepersistent L4Squeue and starvingtraffic deadlocking the Classicqueue,queue for a while in some implementations, it is advisable for the priority to be conditional, not strict (seeAppendix A of the<xref target="RFC9332" format="default" section="A" sectionFormat="of">the DualQspec <xref target="I-D.ietf-tsvwg-aqm-dualq-coupled" format="default"/>).spec</xref>). </t> <t>When there is no Classic traffic, the L4S queue's own AQM comes into play. It starts congestion marking with a very shallow queue, so L4S traffic maintains very low queuing delay.</t> <t>If either queue becomes persistently overloaded, drop of some ECN-capable packets is introduced, as recommended inSection 7 of the ECN spec <xref<xref target="RFC3168"format="default"/>sectionFormat="of" section="7">the ECN spec</xref> andSection 4.2.1 of the AQM recommendations <xref<xref target="RFC7567"format="default"/>. Then both queues introduce the same level of dropsectionFormat="of" section="4.2.1">the AQM recommendations</xref>. The trade-offs with different approaches are discussed in <xref target="RFC9332" sectionFormat="of" section="4.2.3">the DualQ spec</xref> (not shown in thefigure).</t>figure here).</t> <t>TheDual QueueDual-Queue Coupled AQM has been specified as generically aspossible <xref target="I-D.ietf-tsvwg-aqm-dualq-coupled"possible <xref target="RFC9332" format="default"/> without specifying the particular AQMs to use in the two queues so that designers are free to implement diverse ideas. Informational appendices in thatdraftdocument give pseudocode examples of two different specific AQM approaches: one called DualPI2 (pronounced Dual PISquared) <xrefSquared) <xref target="DualPI2Linux" format="default"/> that uses the PI2 variant ofPIE,PIE and a zero-config variant ofREDRandom Early Detection (RED) called Curvy RED. A DualQ Coupled AQM based on PIE has also been specified and implemented for Low LatencyDOCSIS <xrefDOCSIS <xref target="DOCSIS3.1" format="default"/>.</t> <figure anchor="l4sps_fig_components"> <name>Components of an L4S DualQ Coupled AQMSolution: 1) Scalable Sending Host; 2) Isolation in separate network queues; and 3) Packet Identification Protocol</name>Solution</name> <artwork align="center" name="" type="" alt=""><![CDATA[ (3) (2) .-------^------..------------^------------------. ,-(1)-----. _____ ; ________ : L4S -------. | | :|Scalable| : _\ ||__\_|mark | :| sender | : __________ / / || / |_____|\ _________ :|________|\; | |/ -------' ^ \1|condit'nl| `---------'\_| IP-ECN | Coupling : \|priority |_\ ________ / |Classifier| : /|scheduler| / |Classic |/ |__________|\ -------. __:__ / |_________| | sender | \_\ || | ||__\_|mark/|/ |________| / || | || / |drop | Classic -------' |_____| (1) Scalable sending host (2) Isolation in separate network queues (3) Packet identification protocol ]]></artwork> </figure> </li> <li>Per-Flow Queues and AQMs: A scheduler with per-flowqueuesqueues, such as FQ-CoDel orFQ-PIEFQ-PIE, can be used for L4S. Forinstanceinstance, within each queue of an FQ-CoDel system, as well as a CoDel AQM, there is typically also the option of ECN marking at an immediate (unsmoothed) shallow threshold to support use in data centres (seeSec.5.2.7 of the FQ-CoDel spec <xref<xref target="RFC8290"format="default"/>).sectionFormat="of" section="5.2.7">the FQ-CoDel spec</xref>). In Linux, this has been modified so that the shallow threshold can be solely applied to ECT(1)packets <xrefpackets <xref target="FQ_CoDel_Thresh" format="default"/>. Then, if there is a flow ofnon-ECNNot-ECT or ECT(0) packets in theper-flow-queue,per-flow queue, the Classic AQM(e.g. CoDel)(e.g., CoDel) is applied;whilewhereas, if there is a flow of ECT(1) packets in the queue, the shallower (typically sub-millisecond) threshold is applied. In addition, ECT(0) andnot-ECTNot-ECT packets could potentially be classified into a separateflow-queueflow queue from ECT(1) and CE packets to avoid them mixing if they share a commonflow-identifier (e.g. inflow identifier (e.g., in a VPN).</li> <li><t>Dual-queues,<t>Dual queues but per-flow AQMs: It should also be possible to use dual queues forisolation,isolation but with per-flow marking to controlflow-ratesflow rates (instead of the coupled per-queue marking of theDual QueueDual-Queue Coupled AQM). One of the two queues would be for isolating L4S packets, which would be classified by the ECN codepoint. Flow rates could be controlled by flow-specific marking. The policy goal of the marking could be to differentiate flow rates(e.g. <xref(e.g., <xref target="Nadas20" format="default"/>, which requires additional signalling of a per-flow'value'),'value') or to equalizeflow-ratesflow rates (perhaps in a similar way to Approx FairCoDel <xrefCoDel <xref target="AFCD"format="default"/>,format="default"/> <xref target="I-D.morton-tsvwg-codel-approx-fair"format="default"/>,format="default"/> but with two queues not one).</t> <t>Notethatthat, whenever the term 'DualQ' is used loosely without saying whether marking isper-queueper queue orper-flow,per flow, it means adual queuedual-queue AQM with per-queue marking.</t> </li> </ol> </section> <section anchor="l4sps_host_components" numbered="true" toc="default"> <name>Host Mechanisms</name> <t>The L4S architecture includes two main mechanisms in the end host that we enumerate next:</t> <ol spacing="normal" type="a"><li> <t>ScalableCongestion Controlcongestion control at the sender: <xref target="l4s-arch_arch_overview" format="default"/> defines ascalableScalable congestion control as one where the average time from one congestion signal to the next (the recovery time) remains invariant astheflow rate scales, all other factors being equal.Data Center TCPDCTCP is the most widely used example. It has been documented as an informational record of the protocol currently in use in controlledenvironments <xrefenvironments <xref target="RFC8257" format="default"/>. Adraftlist of safety and performance improvements for ascalableScalable congestion control to be usable on the public Internet has been drawn up(the(see the so-called 'Prague L4S requirements' inAppendix A of<xreftarget="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/>).target="RFC9331" format="default" sectionFormat="of" section="A"/>). The subset that involve risk of harm to others have been captured as normative requirements inSection 4 of<xreftarget="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/>.target="RFC9331" format="default" sectionFormat="of" section="4"/>. TCPPrague <xrefPrague <xref target="I-D.briscoe-iccrg-prague-congestion-control" format="default"/> has been implemented in Linux as a reference implementation to address theserequirements <xrefrequirements <xref target="PragueLinux" format="default"/>.</t> <t>Transport protocols other than TCP use various congestion controls that are designed to be friendly with Reno. Before they can use the L4S service, they will need to be updated to implement ascalableScalable congestion response, which they will have to indicate by using the ECT(1) codepoint. Scalable variants are under consideration for more recent transportprotocols, e.g. QUIC,protocols (e.g., QUIC), and the L4S ECN part ofBBRv2 <xrefBBRv2 <xref target="BBRv2"format="default"/>,format="default"/> <xref target="I-D.cardwell-iccrg-bbr-congestion-control" format="default"/> is ascalableScalable congestion control intended for the TCP and QUIC transports, amongst others. Also, an L4S variant of the RMCAT SCReAMcontroller <xrefcontroller <xref target="RFC8298" format="default"/> has beenimplemented <xref target="SCReAM"implemented <xref target="SCReAM-L4S" format="default"/> for media transported over RTP.</t><t>Section 4.3 of the<t> <xref target="RFC9331" format="default" sectionFormat="of" section="4.3">the L4S ECNspec <xref target="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/>spec</xref> definesscalableScalable congestion control in moredetail,detail and specifies the requirements that an L4SscalableScalable congestion control has to comply with.</t> </li> <li> <t>The ECN feedback in some transport protocols is already sufficiently fine-grained for L4S (specificallyDCCP <xrefDCCP <xref target="RFC4340" format="default"/> andQUIC <xrefQUIC <xref target="RFC9000" format="default"/>). But others either requireupdateupdates or are in the process of being updated:</t> <ul spacing="normal"> <li>For the case of TCP, the feedback protocol for ECN embeds the assumption from ClassicECN <xrefECN <xref target="RFC3168" format="default"/> that an ECN mark is equivalent to a drop, making it unusable for ascalableScalable TCP. Therefore, the implementation of TCP receivers will have to beupgraded <xrefupgraded <xref target="RFC7560" format="default"/>. Work to standardize and implement more accurate ECN feedback for TCP (AccECN) is inprogress <xrefprogress <xref target="I-D.ietf-tcpm-accurate-ecn"format="default"/>,format="default"/> <xref target="PragueLinux" format="default"/>.</li> <li>ECN feedback was only roughly sketched inanthe appendix of the now obsoleted second specification of SCTP <xref target="RFC4960" format="default"/>, while a fuller specification was proposed in a long-expireddraft <xrefdocument <xref target="I-D.stewart-tsvwg-sctpecn" format="default"/>. A new design would need to be implemented and deployed before SCTP could support L4S.</li> <li>For RTP, sufficient ECN feedback was definedin <xrefin <xref target="RFC6679" format="default"/>, but <xref target="RFC8888" format="default"/> defines the lateststandards trackStandards Track improvements.</li> </ul> </li> </ol> </section> </section> <section anchor="l4sps_rationale" numbered="true" toc="default"> <name>Rationale</name><t/><section anchor="l4sps_why_primary_components" numbered="true" toc="default"> <name>Why These Primary Components?</name> <dl newline="false" spacing="normal"> <dt>Explicit congestion signalling (protocol):</dt> <dd> <t>Explicit congestion signalling is a key part of the L4S approach. In contrast, use of drop as a congestion signal createsatension because drop is both an impairment (less would be better) and a useful signal (more would be better):</t> <ul spacing="normal"> <li>Explicit congestion signals can be used many times per roundtrip,trip to keep tightcontrol,control without any impairment. Under heavy load, even more explicit signals can beapplied,applied so that the queue can be kept short whatever the load. In contrast, Classic AQMs have to introduce very high packet drop at high load to keep the queue short. By using ECN, an L4S congestion control's sawtooth reduction can be smaller and therefore return to the operating point more often, without worrying that more sawteeth will cause more signals. The consequent smaller amplitude sawteeth fit between an empty queue and a very shallow marking threshold(~1 ms(~1 ms in the public Internet), so queue delay variation can be very low, without risk ofunder-utilization.</li>underutilization.</li> <li>Explicit congestion signals can be emitted immediately to track fluctuations of the queue. L4S shifts smoothing from the network to the host. The network doesn't know theround tripround-trip times (RTTs) of any of the flows. So if the network is responsible for smoothing (as in the Classic approach), it has to assume a worst case RTT, otherwise long RTT flows would become unstable. This delays Classic congestion signals by 100-200 ms. In contrast, each host knows its ownround trip time.RTT. So, in the L4S approach, the host can smooth each flow over its own RTT, introducing no more smoothing delay than strictly necessary (usually only a few milliseconds). A host can also choose not to introduce any smoothing delay if appropriate,e.g. duringe.g., during flow start-up.</li> </ul> <t>Neither of the above are feasible if explicit congestion signalling has to be considered 'equivalent to drop' (as was required with ClassicECN <xrefECN <xref target="RFC3168" format="default"/>), because drop is an impairment as well as a signal. So drop cannot be excessively frequent, and drop cannot beimmediate, otherwiseimmediate; otherwise, too many drops would turn out to have been due to only a transient fluctuation in the queue that would not have warranted dropping a packet in hindsight. Therefore, in an L4S AQM, the L4S queue uses a new L4S variant of ECN that is not equivalent todrop (see section 5.2 of thedrop (see <xref target="RFC9331" format="default" sectionFormat="of" section="5.2">the L4S ECNspec <xref target="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/>),spec</xref>), while the Classic queue uses either ClassicECN <xrefECN <xref target="RFC3168" format="default"/> or drop, which are still equivalent to each other.</t> <t>Before Classic ECN was standardized, there were various proposals to give an ECN mark a different meaning from drop. However, there was no particular reason to agree on any one of the alternative meanings, so 'equivalent to drop' was the only compromise that could be reached.RFC 3168<xref target="RFC3168" format="default"/> contains a statement that:</t> <ulempty="true" spacing="normal"> <li>"Anempty="true"> <li><t indent="1">An environment where all end nodes were ECN-Capable could allow new criteria to be developed for setting the CE codepoint, and new congestion control mechanisms for end-node reaction to CE packets. However, this is a research issue, and as such is not addressed in thisdocument."</li> </ul>document.</t></li></ul> </dd> <dt>Latency isolation (network):</dt> <dd>L4S congestion controls keep queue delaylowlow, whereas Classic congestion controls need a queue of the order of the RTT to avoidunder-utilization.underutilization. One queue cannot have twolengths, thereforelengths; therefore, L4S traffic needs to be isolated in a separate queue(e.g. DualQ)(e.g., DualQ) or queues(e.g. FQ).</dd>(e.g., FQ).</dd> <dt>Coupled congestion notification:</dt> <dd>Coupling the congestion notification between two queues as in the DualQ Coupled AQM is not necessarily essential, but it is a simple way to allow senders to determine theirrate,rate packet by packet, rather than be overridden by a network scheduler. An alternative is for a network scheduler to control the rate of each application flow (see the discussion in <xref target="l4sps_why-not" format="default"/>).</dd> <dt>L4S packet identifier (protocol):</dt> <dd>Once there are at least two treatments in the network, hosts need an identifier at the IP layer to distinguish which treatment they intend to use.</dd> <dt>Scalable congestion notification:</dt> <dd>AscalableScalable congestion control in the host keeps the signalling frequency from the networkhighhigh, whatever the flow rate, so that queue delay variations can be small when conditions are stable, and rate can track variations in available capacity as rapidly as possible otherwise.</dd> <dt>Low loss:</dt> <dd>Latency is not the only concern of L4S. The 'Low Loss' part of the name denotes that L4S generally achieves zero congestion loss due to its use of ECN. Otherwise, loss would itself cause delay, particularly for short flows, due to retransmissiondelay <xrefdelay <xref target="RFC2884" format="default"/>.</dd> <dt>Scalable throughput:</dt> <dd> <t>The"Scalable throughput"'Scalable throughput' part of the name denotes that the per-flow throughput ofscalableScalable congestion controls should scale indefinitely, avoiding the imminent scaling problems with Reno-friendly congestion controlalgorithms <xrefalgorithms <xref target="RFC3649" format="default"/>. It was known when TCP congestion avoidance was first developed in 1988 that it would not scale to high bandwidth-delay products (see footnote 6 in <xref target="TCP-CA" format="default"/>). Today, regular broadband flow rates over WAN distances are already beyond the scaling range of Classic Reno congestion control. So`less'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. </t> <t>For instance, we will consider a scenario with a maximum RTT of30 ms30 ms at the peak of each sawtooth. As Reno packet rate scales8x8 times from 1,250 to 10,000 packet/s (from 15 to120 Mb/s120 Mb/s with1500 B1500 B packets), the time to recover from a congestion event rises proportionately by8x8 times as well, from422 ms422 ms to3.38 s.3.38 s. It is clearly problematic for a congestion control to take multiple seconds to recover from each congestion event.Cubic <xrefCUBIC <xref target="RFC8312" format="default"/> was developed to be less unscalable, but it is approaching its scaling limit; with the same max RTT of30 ms,30 ms, at120 Mb/s Cubic120 Mb/s, CUBIC is still fully in its Reno-friendly mode, so it takes about4.3 s4.3 s to recover. However, oncetheflow rate scales by8x8 times again to960 Mb/s960 Mb/s it enters trueCubicCUBIC mode, with a recovery time of12.2 s.12.2 s. From then on, each further scaling by8x8 times doublesCubic'sCUBIC's recovery time (because the cube root of 8 is 2),e.g. at 7.68 Gb/se.g., at 7.68 Gb/s, the recovery time is24.3 s.24.3 s. In contrast, ascalableScalable congestion control like DCTCP orTCPPrague induces 2 congestion signals per round trip on average, which remains invariant for any flow rate, keeping dynamic control very tight.</t> <t>For a feel of where the global average lone-flow download sits on this scale at the time of writing (2021), according to <xref target="BDPdata"format="default"/> globally averagedformat="default"/>, the global average fixed access capacity was 103 Mb/s in 2020 andaveragedthe average base RTT to a CDN was25-34ms25 to 34 ms in 2019. Averaging of per-country data was weighted by Internet user population (data collected globally is necessarily of variable quality, but the paper does double-check that the outcome compares well against a second source). So a lone CUBIC flow would at best take about 200 round trips (5 s) to recover from each of its sawtooth reductions, if the flow even lasted that long. This is described as 'at best' because it assumes everyone uses an AQM, whereas inrealityreality, most users still have a (probably bloated) tail-drop buffer. In the tail-drop case, the likely average recovery time would be at least4x4 times 5 s, if not more, because RTT under load would be at least double that of an AQM, and the recovery time of Reno-friendly flows depends on the square of RTT.</t> <t>Although work on scaling congestion controls tends to start with TCP as the transport, the above is not intended to exclude other transports(e.g. SCTP,(e.g., SCTP and QUIC) or less elastic algorithms(e.g. RMCAT),(e.g., RMCAT), which all tend to adopt the same or similar developments.</t> </dd> </dl> </section> <section anchor="l4sps_why-not" numbered="true" toc="default"> <name>What L4SaddsAdds to Existing Approaches</name> <t>All the following approaches address some part of the same problem space as L4S. In each case, it is shown that L4S complements them or improves on them, rather than being a mutually exclusive alternative:</t> <dl newline="false" spacing="normal"> <dt>Diffserv:</dt> <dd> <t>Diffserv addresses the problem of bandwidth apportionment for important traffic as well as queuing latency for delay-sensitive traffic. Of these, L4S solely addresses the problem of queuing latency. Diffserv will still be necessary where important traffic requires priority(e.g. for(e.g., for commercialreasons,reasons or for protection of critical infrastructure traffic)--- see <xref target="I-D.briscoe-tsvwg-l4s-diffserv" format="default"/>. Nonetheless, the L4S approach can provide low latency for all traffic within each Diffserv class (including the case where there is only the one default Diffserv class).</t> <t>Also, Diffserv can only provide a latency benefit if a small subset of the traffic on a bottleneck link requests low latency. As already explained, it has no effect when all the applications in use at one time at a single site(home,(e.g., a home, smallbusinessbusiness, or mobile device) require low latency. In contrast, because L4S works for all traffic, it needs none of the management baggage (trafficpolicing,policing or traffic contracts) associated with favouring some packets over others. This lack of management baggage ought to give L4S a better chance of end-to-end deployment.</t> <t>In particular,becauseif networkstenddo nottotrust end systems to identify which packets should befavoured over others, where networksfavoured, they assign packets to Diffserv classesthey tendthemselves. However, the techniques available touse packetsuch networks, like inspection ofapplicationflow identifiers or deeper inspection of applicationsignatures. Thus, nowadays, Diffserv doesn'tsignatures, do not always sit well with encryption of the layers above IP <xref target="RFC8404" format="default"/>.SoIn these cases, users can haveto choose betweeneither privacyand QoS.</t>or Quality of Service (QoS), but not both.</t> <t>As with Diffserv, the L4S identifier is in the IP header. But, in contrast to Diffserv, the L4S identifier does not convey a want or a need for a certain level of quality. Rather, it promises a certain behaviour(scalable(Scalable congestion response), which networks can objectively verify if they need to. This is because low delay depends on collective host behaviour, whereas bandwidth priority depends on network behaviour.</t> </dd> <dt>State-of-the-art AQMs:</dt> <dd>AQMs for Classic traffic, such as PIE andFQ-CoDelFQ-CoDel, give a significant reduction in queuing delay relative to no AQM at all. L4S is intended to complement theseAQMs,AQMs and should not distract from the need to deploy them as widely as possible. Nonetheless, AQMs alone cannot reduce queuing delay too far without significantly reducing link utilization, because the root cause of the problem is on the host--- where Classic congestion controls use largesaw-toothingsawtoothing rate variations. The L4S approach resolves this tension between delay and utilization by enabling hosts to minimize the amplitude of their sawteeth. A single-queue Classic AQM is not sufficient to allow hosts to use small sawteeth for two reasons: i) smaller sawteeth would not get lower delay in an AQM designed for larger amplitude Classic sawteeth, because a queue can only have one length at atime;time and ii) much smaller sawteeth implies much more frequent sawteeth, so L4S flows would drive a Classic AQM into a high level of ECN-marking, which would appear as heavy congestion to Classic flows, which in turn would greatly reduce their rate as a result (see <xref target="l4sarch_sec_classic-ecn-neck" format="default"/>).</dd> <dt>Per-flow queuing or marking:</dt> <dd> <t>Similarly, per-flowapproachesapproaches, such as FQ-CoDel or Approx FairCoDel <xrefCoDel <xref target="AFCD"format="default"/>format="default"/>, are not incompatible with the L4S approach. However, per-flow queuing alone is not enough--- it only isolates the queuing of one flow fromothers;others, not from itself. Per-flow implementations need to have support forscalableScalable congestion control added, which has already been done for FQ-CoDel in Linux (seeSec.5.2.7 of<xref target="RFC8290"format="default"/>sectionFormat="of" section="5.2.7"/> and <xref target="FQ_CoDel_Thresh" format="default"/>). Without this simple modification, per-flowAQMsAQMs, likeFQ-CoDelFQ-CoDel, would still not be able to support applications that need both very low delay and high bandwidth,e.g. video-basede.g., video-based control of remoteprocedures,procedures or interactive cloud-based video (seeNote <xrefNote <xref format="counter" target="l4sarch_note_app_shuffle"/> below).</t> <t>Although per-flow techniques are not incompatible with L4S, it is important to have the DualQ alternative. This is because handling end-to-end (layer 4) flows in the network (layer 3 or 2) precludes some important end-to-end functions. For instance:</t> <ol spacing="normal"type="a"><li>type="A"><li> <t>Per-flow forms ofL4SL4S, likeFQ-CoDelFQ-CoDel, are incompatible with full end-to-end encryption of transport layer identifiers for privacy and confidentiality(e.g. IPSec(e.g., IPsec or encrypted VPN tunnels, as opposed to DTLS over UDP), because they require packet inspection to access the end-to-end transport flow identifiers. </t> <t>In contrast, the DualQ form of L4S requires no deeper inspection than the IP layer.So,So as long as operators take the DualQ approach, their users can have both very low queuing delay and full end-to-endencryption <xrefencryption <xref target="RFC8404" format="default"/>.</t> </li> <li> <t>With per-flow forms of L4S, the network takes over control of the relative rates of each application flow. Some see it as an advantage that the network will prevent some flows running faster than others. Others consider it an inherent part of the Internet's appeal that applications can control their rate while taking account of the needs of others via congestion signals. They maintain that this has allowed applications with interesting rate behaviours to evolve, forinstance,instance: i) a variable bit-rate video that varies around an equalshareshare, rather than being forced to remain equal at everyinstant,instant ore2eii) end-to-end scavengerbehaviours <xrefbehaviours <xref target="RFC6817" format="default"/> that use less than an equal share ofcapacity <xrefcapacity <xref target="LEDBAT_AQM" format="default"/>.</t> <t>The L4S architecture does not require the IETF to commit to one approach over the other, because it supportsboth,both so that the 'market' can decide. Nonetheless, in the spirit of 'Do one thing and do itwell' <xrefwell' <xref target="McIlroy78" format="default"/>, the DualQ option provides low delay without prejudging the issue of flow-rate control. Then, flow rate policing can be added separately if desired.This allowsIn contrast to scheduling, a policer would allow application control up to a point, but the networkcanwould stillchoosebe able to set the point at which itintervenesintervened to prevent one flow completely starving another.</t> </li><!-- <t>fq prevents any one flow from consuming more than 1/N of the capacity at any instant, where N is the number of flows. This is fine if all flows are elastic, but it does not sit well with a variable bit rate real-time multimedia flow, which requires wriggle room to sometimes take more and other times less than a 1/N share.<vspace blankLines="1"/>It might seem that an fq scheduler offers the benefit that it prevents individual flows from hogging all the bandwidth. However, L4S has been deliberately designed so that policing of individual flows can be added as a policy choice, rather than requiring one specific policy choice as the mechanism itself. {ToDo: refer to paper on FQ+LEDBAT rather than explain it here - but we might end up removing this whole bullet} On the other other end of the spectrum, fq also prevent a flow from using less than 1/N (otherwise the flow would equally underutilize the link when N=1). In a shared queue, all flows get equal congestion signal feedback, which allows less-than-best-effort-flows to use a lower rate to probability ratio than Reno-friendly traffic. With fq, the capacity is split by N equal parts, and congestion feedback is only valid for the 1/N capacity partition ocupied by the less-than best-effort flow as if the flow is alone (N=1) and would then try to fully utilize the available capacity.{/ToDo} A scheduler (like fq) has to decide packet-by-packet which flow to schedule without knowing application intent. Whereas a separate policing function can be configured less strictly, so that senders can still control the instantaneous rate of each flow dependent on the needs of each application (e.g. variable rate video), giving more wriggle-room before a flow is deemed non-compliant. Also policing of queuing and of flow-rates can be applied independently.</t> --></ol> <t>Note: </t> <ol spacing="normal"type="1"><litype="1"> <li anchor="l4sarch_note_app_shuffle">It might seem that self-inflicted queuing delay within a per-flow queue should not be counted, because if the delay wasn't in thenetworknetwork, it would just shift to the sender. However, modern adaptive applications,e.g. HTTP/2 <xrefe.g., HTTP/2 <xref target="RFC9113" format="default"/> or some interactive media applications (see <xref target="l4sarch_apps" format="default"/>), can keep low latency objects at the front of their local send queue by shuffling priorities of other objects dependent on the progress of other transfers (forexampleexample, see <xref target="lowat" format="default"/>). They cannot shuffle objects once they have released them into the network.</li> </ol> </dd> <dt>Alternative Back-off ECN (ABE):</dt> <dd>Here again, L4S is not an alternative to ABE but a complement that introduces much lower queuing delay.ABE <xrefABE <xref target="RFC8511" format="default"/> alters the host behaviour in response to ECN marking to utilize a link better and give ECN flows faster throughput. It uses ECT(0) and assumes the network still treats ECN and drop the same. Therefore, ABE exploits any lower queuing delay that AQMs can provide. But, as explained above, AQMs still cannot reduce queuing delay toofarmuch without losing link utilization (to allow for other, non-ABE, flows).</dd> <dt>BBR:</dt> <dd> <t>Bottleneck Bandwidth and Round-trip propagation time(BBR <xref(BBR) <xref target="I-D.cardwell-iccrg-bbr-congestion-control"format="default"/>)format="default"/> controls queuing delay end-to-end without needing any special logic in the network, such as an AQM. So it workspretty-muchpretty much on any path. BBR keeps queuing delay reasonably low, but perhaps not quite as low as with state-of-the-artAQMsAQMs, such as PIE or FQ-CoDel, and certainly nowhere near as low as with L4S. Queuing delay is also not consistently low, due to BBR's regular bandwidth probing spikes and its aggressive flow start-up phase.</t> <t>L4S complements BBR. Indeed, BBRv2 can use L4S ECN where available and ascalableScalable L4S congestion control behaviour in response to any ECN signalling from thepath <xrefpath <xref target="BBRv2" format="default"/>. The L4S ECN signal complements thedelay baseddelay-based congestion control aspects of BBR with an explicit indication that hosts can use, both to converge on a fair rate and to keep below a shallow queue target set by the network. Without L4S ECN, both these aspects need to be assumed or estimated.</t> </dd> </dl> </section> </section> <section anchor="l4sarch_applicability" numbered="true" toc="default"> <name>Applicability</name> <section anchor="l4sarch_apps" numbered="true" toc="default"> <name>Applications</name> <t>A transport layer that solves the current latency issues will provide new service,productproduct, and application opportunities.</t> <t>With the L4S approach, the following existing applications also experience significantly better quality of experience under load: </t> <ul spacing="normal"><li>Gaming,<li>gaming, includingcloud basedcloud-based gaming;</li> <li>VoIP;</li><li>Video<li>video conferencing;</li><li>Web<li>web browsing;</li><li>(Adaptive)<li>(adaptive) videostreaming;</li> <li>Instantstreaming; and</li> <li>instant messaging.</li> </ul> <t>The significantly lower queuing latency also enables some interactive application functions to be offloaded to the cloud that would hardly even be usabletoday: </t>today, including:</t> <ul spacing="normal"><li>Cloud based<li>cloud-based interactivevideo;</li> <li>Cloud basedvideo and</li> <li>cloud-based virtual and augmented reality.</li> </ul> <t>The above two applications have been successfully demonstrated with L4S, both running together over a40 Mb/s40 Mb/s broadband access link loaded up with the numerous otherlatency sensitivelatency-sensitive applications in the previouslistlist, as well as numerousdownloads -downloads, with all sharing the same bottleneck queuesimultaneously <xrefsimultaneously <xref target="L4Sdemo16" format="default"/> <xref target="L4Sdemo16-Video" format="default"/>. For the former, a panoramic video of a football stadium could be swiped and pinched so that, on the fly, a proxy in the cloud could generate a sub-window of the match video under the finger-gesture control of each user. For the latter, a virtual reality headset displayed a viewport taken from a 360-degree camera in a racing car. The user's head movements controlled the viewport extracted by a cloud-based proxy. In both cases, with7 msa 7 ms end-to-end base delay, the additional queuing delay of roughly1 ms1 ms was so low that it seemed the video was generated locally. </t> <t>Using a swiping finger gesture or head movement to pan a video are extremely latency-demanding actions -- far more demanding thanVoIP. BecauseVoIP -- because human vision can detect extremely low delays of the order of single milliseconds when delay is translated into a visual lag between a video and a reference point (the finger or the orientation of the head sensed by the balance system in the innerear --ear, i.e., the vestibular system). With an alternative AQM, the video noticeably lagged behind the finger gestures and head movements.</t> <t>Without the low queuing delay of L4S, cloud-based applications like these would not be credible without significantly moreaccessaccess-network bandwidth (to deliver all possible areas of the video that might be viewed) and more local processing, which would increase the weight and power consumption of head-mounted displays. When all interactive processing can be done in the cloud, only the data to be rendered for the end user needs to be sent.</t> <t>Other low latency high bandwidthapplicationsapplications, such as:</t> <ul spacing="normal"><li>Interactive<li>interactive remotepresence;</li> <li>Video-assistedpresence and</li> <li>video-assisted remote control of machinery or industrialprocesses.</li>processes</li> </ul> <t>are not credible at all without very low queuing delay. No amount of extra access bandwidth or local processing can make up for lost time.</t> </section> <section numbered="true" toc="default"> <name>Use Cases</name> <t>The followinguse-casesuse cases for L4S are being considered by various interested parties:</t> <ul spacing="normal"><li>Where<li>where the bottleneck is one of various types of accessnetwork: e.g. DSL,network, e.g., DSL, Passive Optical Networks(PON),(PONs), DOCSIS cable, mobile,satellitesatellite; or where it's a Wi-Fi link (see <xref target="l4sarch_link-specifics" format="default"/> for some technology-specific details)</li> <li><t>Private<t>private networks of heterogeneous data centres, where there is no single administrator that can arrange for all the simultaneous changes to senders,receiversreceivers, andnetworknetworks needed to deploy DCTCP:</t> <ul spacing="normal"> <li>a set of private data centres interconnected over a wide area with separateadministrations,administrations but within the same company</li> <li>a set of data centres operated by separate companies interconnected by a community of interest network(e.g. for(e.g., for the finance sector)</li> <li>multi-tenant (cloud) data centres where tenants choose their operating system stack (Infrastructure as a Service- IaaS)</li>(IaaS))</li> </ul> </li> <li><t>Different<t>different types of transport (or application) congestion control:</t> <ul spacing="normal"> <li>elastic (TCP/SCTP);</li> <li>real-time (RTP,RMCAT);</li> <li>queryRMCAT); and</li> <li>query-response (DNS/LDAP).</li> </ul> </li> <li><t>Where<t>where low delayquality of serviceQoS isrequired,required but without inspecting or intervening above the IPlayer <xreflayer <xref target="RFC8404" format="default"/>:</t> <ul spacing="normal"><li>mobile<li>Mobile and other networks have tended to inspect higher layers in order to guess application QoS requirements. However, with growing demand for support of privacy and encryption, L4S offers an alternative. There is no need to select which traffic to favour forqueuing,queuing when L4S can give favourable queuing to all traffic.</li> </ul> </li> <li>If queuing delay is minimized, applications with a fixed delay budget can communicate over longerdistances,distances or viaamore circuitous paths, e.g., longerchainchains of servicefunctions <xreffunctions <xref target="RFC7665" format="default"/> or of onion routers.</li> <li>If delay jitter is minimized, it is possible to reduce the dejitter buffers on thereceivereceiving end of video streaming, which should improve the interactiveexperience</li>experience.</li> </ul> </section> <section anchor="l4sarch_link-specifics" numbered="true" toc="default"> <name>Applicability with Specific Link Technologies</name> <t>Certain link technologies aggregate data from multiple packets intobursts,bursts and buffer incoming packets while building each burst. Wi-Fi,PONPON, and cable all involve such packet aggregation, whereas fixed Ethernet and DSL do not. No sender, whether L4S or not, can do anything to reduce the buffering needed for packet aggregation. So an AQM should not count this buffering as part of the queue that it controls, given no amount of congestion signals will reduce it.</t> <t>Certain link technologies also add buffering for other reasons, specifically:</t> <ul spacing="normal"> <li>Radio links (cellular, Wi-Fi, or satellite) that are distant from the source are particularly challenging. The radio link capacity can vary rapidly by orders of magnitude, so it is considered desirable to hold a standing queue that can utilize sudden increases ofcapacity;</li>capacity.</li> <li>Cellular networks are further complicated by a perceived need to buffer in order to make hand-oversimperceptible;</li>imperceptible.</li> </ul> <t>L4S cannot remove the need for all these different forms of buffering. However, by removing 'the longest pole in the tent' (buffering for the large sawteeth of Classic congestion controls), L4S exposes all these 'shorter poles' to greater scrutiny.</t> <t>Until now, the buffering needed for these additional reasons tended to be over-specified--- with the excuse that none were 'the longest pole in the tent'. But having removed the 'longest pole', it becomes worthwhile to minimize them, forinstanceinstance, reducing packet aggregation burst sizes and MAC scheduling intervals.</t><t>Also<t>Also, certain link types, particularly radio-based links, are far more prone to transmission losses. <xref target="l4sarch_sec_non-l4s-neck" format="default"/> explains how an L4S response to loss has to be as drastic as a Classic response. Nonetheless, research referred to in the same section has demonstrated potential for considerably more effective loss repair at the link layer, due to the relaxed ordering constraints of L4S packets.</t> </section> <section numbered="true" toc="default"> <name>Deployment Considerations</name> <t>L4S AQMs, whetherDualQ <xref target="I-D.ietf-tsvwg-aqm-dualq-coupled"DualQ <xref target="RFC9332" format="default"/> orFQ, e.g. <xrefFQ <xref target="RFC8290"format="default"/> are,format="default"/>, are inthemselves,themselves an incremental deployment mechanism for L4S--- so that L4S traffic can coexist with existing Classic (Reno-friendly) traffic. <xref target="l4sarch_deploy_top" format="default"/> explains why only deploying an L4S AQM in one node at each end of the access link will realize nearly all the benefit of L4S.</t> <t>L4S involves bothend systems andthenetwork,network and end systems, so <xref target="l4s_arch_deploy_seq" format="default"/> suggests some typical sequences to deploy eachpart,part and why there will be an immediate and significant benefit after deploying just one part.</t><t><xref<t>Sections <xref target="l4sarch_sec_non-l4s-neck"format="default"/>format="counter"/> and <xref target="l4sarch_sec_classic-ecn-neck"format="default"/>format="counter"/> describe the converse incremental deployment case where there is no L4S AQM at the network bottleneck, so any L4S flow traversing this bottleneck has to take care in case it is competing with Classic traffic.</t> <section anchor="l4sarch_deploy_top" numbered="true" toc="default"> <name>Deployment Topology</name> <t>L4S AQMs will not have to be deployed throughout the Internet before L4S can benefit anyone. Operators of public Internet access networks typically design their networks so that the bottleneck will nearly always occur at one known (logical) link. This confines the cost of queue management technology to one place.</t> <t>The case of mesh networks is different and will be discussed later in this section.ButHowever, theknown bottleneckknown-bottleneck case is generally true for Internet access to all sorts of different 'sites', where the word 'site' includes home networks, small- to medium-sized campus or enterprise networks and even cellular devices (<xref target="l4sarch_fig_access_topology" format="default"/>). Also, this known-bottleneck case tends to be applicable whatever the access linktechnology;technology, whether xDSL, cable, PON, cellular, line of sightwirelesswireless, or satellite.</t> <t>Therefore, the full benefit of the L4S service should be available in the downstream direction when an L4S AQM is deployed at the ingress to this bottleneck link. And similarly, the full upstream service will typically be available once an L4S AQM is deployed at the ingress into the upstream link. (Of course,multi-homedmultihomed sites would only see the full benefit once all their access links were covered.)</t> <figure anchor="l4sarch_fig_access_topology"> <name>LikelylocationLocation of DualQ (DQ) Deployments incommon access topologies</name>Common Access Topologies</name> <artwork name="" type="" align="left" alt=""><![CDATA[ ______ ( ) __ __ ( ) |DQ\________/DQ|( enterprise ) ___ |__/ \__| ( /campus ) ( ) (______) ( ) ___||_ +----+ ( ) __ __ / \ | DC |-----( Core )|DQ\_______________/DQ|| home | +----+ ( ) |__/ \__||______| (_____) __ |DQ\__/\ __ ,===. |__/ \ ____/DQ||| ||mobile \/ \__|||_||device | o | `---' ]]></artwork> </figure> <t>Deployment in mesh topologies depends on how overbooked the core is. If the core is non-blocking, or at least generously provisioned so that the edges are nearly always the bottlenecks, it would only be necessary to deploy an L4S AQM at the edge bottlenecks. For example, some data-centre networks are designed with the bottleneck in the hypervisor or hostNICs,Network Interface Controllers (NICs), while others bottleneck at the top-of-rack switch (both the output ports facing hosts and those facing the core).</t> <t>An L4S AQM would often next be needed where the Wi-Fi links in a home sometimes become the bottleneck.AndAlso an L4S AQM would eventuallyalsoneed to be deployed at any other persistentbottlenecksbottlenecks, such as network interconnections,e.g. somee.g., some public Internet exchange points and the ingress and egress to WAN links interconnectingdata-centres.</t>data centres.</t> </section> <section anchor="l4s_arch_deploy_seq" numbered="true" toc="default"> <name>Deployment Sequences</name> <t>For any one L4S flow to provide benefit, it requires three (or sometimes two) parts to have been deployed: i) the congestion control at the sender; ii) the AQM at the bottleneck; and iii) older transports (namely TCP) need upgraded receiver feedback too. This was the same deployment problem that ECNfaced <xreffaced <xref target="RFC8170"format="default"/>format="default"/>, so we have learned from that experience.</t> <t>Firstly, L4S deployment exploits the fact that DCTCP already exists on many Internet hosts(Windows, FreeBSD(e.g., Windows, FreeBSD, andLinux);Linux), both servers and clients. Therefore, an L4S AQM can be deployed at a network bottleneck to immediately give a working deployment of all the L4S parts for testing, as long as the ECT(0) codepoint is switched to ECT(1). DCTCP needs some safety concerns to be fixed for general use over the public Internet (seeSection 4.3 of the<xref target="RFC9331" format="default" sectionFormat="of" section="4.3">the L4S ECNspec <xref target="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/>),spec</xref>), but DCTCP is not on by default, so these issues can be managed within controlled deployments or controlled trials.</t> <t>Secondly, the performance improvement with L4S is so significant that it enables new interactive services and products that were not previously possible. It is much easier for companies to initiate new work on deployment if there is budget for a new product trial.If, inIn contrast, if there were only an incremental performance improvement (as with Classic ECN), spending on deployment tends to be much harder to justify.</t> <t>Thirdly, the L4S identifier is defined so thatinitiallynetwork operators can initially enable L4S exclusively for certain customers or certain applications.ButHowever, this is carefully defined so that it does not compromise future evolution towards L4S as an Internet-wide service. This is because the L4S identifier is defined not only as the end-to-end ECN field, but it can also optionally be combined with any other packet header or some status of a customer or their access link (seesection 5.4 of <xref target="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/>).<xref target="RFC9331" format="default" sectionFormat="of" section="5.4"/>). Operators could do this anyway, even if it were not blessed by the IETF. However, it is best for the IETF to specify that, if they use their own local identifier, it must be in combination with the IETF'sidentifier.identifier, ECT(1). Then, if an operator has opted for an exclusive local-use approach,laterthey only have to remove this extra rule later to make the service workInternet-wide -across the Internet -- it will already traverse middleboxes, peerings, etc.<!--{K: Review up to here}--></t> <figure anchor="l4s_arch_fig_deploy_seq"> <name>Example L4S Deployment Sequence</name> <artwork name="" type="" align="left"alt=""><![CDATA[+-+--------------------+----------------------+---------------------+alt=""><![CDATA[ +-+--------------------+----------------------+---------------------+ | | Servers or proxies | Access link | Clients | +-+--------------------+----------------------+---------------------+ |0| DCTCP (existing) | | DCTCP (existing) | +-+--------------------+----------------------+---------------------+ |1| |Add L4S AQM downstream| | | | WORKS DOWNSTREAM FOR CONTROLLED DEPLOYMENTS/TRIALS | +-+--------------------+----------------------+---------------------+ |2| Upgrade DCTCP to | |Replace DCTCP feedb'k| | | TCP Prague | | with AccECN | | | FULLY WORKS DOWNSTREAM | +-+--------------------+----------------------+---------------------+ | | | | Upgrade DCTCP to | |3| | Add L4S AQM upstream | TCP Prague | | | | | | | | FULLY WORKS UPSTREAM AND DOWNSTREAM | +-+--------------------+----------------------+---------------------+ ]]></artwork> </figure> <t><xref target="l4s_arch_fig_deploy_seq" format="default"/> illustrates some example sequences in which the parts of L4S might be deployed. It consists of the following stages, preceded by a presumption that DCTCP is already installed at both ends:</t> <ol spacing="normal" type="1"><li> <t>DCTCP is not applicable for use over the public Internet, so it is emphasized here that any DCTCP flow has to be completely contained within a controlled trial environment. </t> <t>Within this trial environment, once an L4S AQM has been deployed, the trial DCTCP flow will experience immediate benefit, without any other deployment being needed. In thisexampleexample, downstream deployment is first, but in otherscenariosscenarios, the upstream might be deployed first. If no AQM at all was previously deployed for the downstream access, an L4S AQM greatly improves the Classic service (as well as adding the L4S service). If an AQM was already deployed, the Classic service will be unchanged (and L4S will add an improvement on top).</t> </li> <li> <t>In this stage, the name 'TCPPrague' <xrefPrague' <xref target="I-D.briscoe-iccrg-prague-congestion-control" format="default"/> is used to represent a variant of DCTCP that is designed to be used in a production Internet environment (that is, it has to comply with all the requirements inSection 4 of the<xref target="RFC9331" format="default" section="4" sectionFormat="of">the L4S ECNspec <xref target="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/>,spec</xref>, which then means it can be used over the public Internet). If the application is primarily unidirectional, 'TCP Prague' atonethe sending end will provide all the benefitneeded.</t> <t>For TCP transports,needed, as long as the receiving end supports Accurate ECN (AccECN) feedback(AccECN) <xref<xref target="I-D.ietf-tcpm-accurate-ecn"format="default"/>format="default"/>.</t> <t>For TCP transports, AccECN feedback is needed at the other end, but it is a generic ECN feedback facility that is already planned to be deployed for other purposes,e.g. DCTCP,e.g., DCTCP and BBR. The two ends can be deployed in eitherorder,order because, in TCP, an L4S congestion control only enables itself if it has negotiated the use of AccECN feedback with the other end during the connection handshake. Thus, deployment of TCP Prague on a server enables L4S trials to move to a production service in one direction, wherever AccECN is deployed at the other end. This stage might be further motivated by the performance improvements of TCP Prague relative to DCTCP (seeAppendix A.2 of the<xref target="RFC9331" format="default" sectionFormat="of" section="A.2">the L4S ECNspec <xref target="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/>).</t>spec</xref>).</t> <t>Unlike TCP, from the outset, QUIC ECNfeedback <xreffeedback <xref target="RFC9000" format="default"/> has supported L4S. Therefore, if the transport is QUIC, one-ended deployment of a Prague congestion control at this stage is simple and sufficient.</t> <t>For QUIC, if a proxy sits in the path between multiple origin servers and the access bottlenecks to multiple clients, then upgrading the proxy with a Scalable congestion control would provide the benefits of L4S over all the clients' downstream bottlenecks in one go----- whether or not all the origin servers were upgraded. Conversely, where a proxy has not been upgraded, the clients served by it will not benefit from L4S at all in the downstream, even when any origin server behind the proxy has been upgraded to support L4S.</t> <t>For TCP, a proxy upgraded to support 'TCP Prague' would provide the benefits of L4S downstream to all clients that support AccECN (whether or not they support L4S as well). And in the upstream, the proxy would also support AccECN as a receiver, so that any client deploying its own L4S support would benefit in the upstream direction, irrespective of whether any origin server beyond the proxy supported AccECN.</t> </li> <li>This is a two-move stage to enable L4S upstream. An L4S AQM or TCP Prague can be deployed in either order as already explained. To motivate the first of two independent moves, the deferred benefit of enabling new services after the second move has to be worth it to cover the first mover's investment risk. As explained already, the potential for new interactive services provides this motivation. An L4S AQM also improves the upstream Classic service-significantly if no other AQM has already been deployed.</li> </ol> <t>Note that other deployment sequences might occur. Forinstance:instance, the upstream might be deployed first; a non-TCP protocol might be usedend-to-end, e.g. QUIC,end to end, e.g., QUIC and RTP; abodybody, such as the3GPP3GPP, might require L4S to be implemented in 5G userequipment,equipment; or other random acts ofkindness.</t>kindness might arise.</t> </section> <section anchor="l4sarch_sec_non-l4s-neck" numbered="true" toc="default"> <name>L4S Flow but Non-ECN Bottleneck</name> <t>If L4S is enabled between two hosts, the L4S sender is required to coexist safely with Reno in response to any drop (seeSection 4.3 of the<xref target="RFC9331" format="default" sectionFormat="of" section="4.3">the L4S ECNspec <xref target="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/>).</t>spec</xref>).</t> <t>Unfortunately, as well as protecting Classic traffic, this rule degrades the L4S service whenever there is any loss, even if the cause is not persistent congestion at a bottleneck,e.g.:</t>for example:</t> <ul spacing="normal"> <li>congestion loss at other transient bottlenecks,e.g. duee.g., due to bursts in shallower queues;</li> <li>transmission errors,e.g. duee.g., due to electricalinterference;</li>interference; and</li> <li>rate policing.</li> </ul> <t>Three complementary approaches are in progress to address this issue, but they are all currently research:</t> <ul spacing="normal"> <li>In Prague congestion control, ignore certain losses deemed unlikely to be due to congestion (using some ideas fromBBR <xrefBBR <xref target="I-D.cardwell-iccrg-bbr-congestion-control" format="default"/> regarding isolated losses). This could mask any of the above types of loss while still coexisting with drop-based congestion controls.</li> <li>A combination ofRACK, L4SRecent Acknowledgement (RACK) <xref target="RFC8985" format="default"/>, L4S, and link retransmission without resequencing could repair transmission errors without the head of line blocking delay usually associated with link-layerretransmission <xrefretransmission <xref target="UnorderedLTE"format="default"/>,format="default"/> <xreftarget="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/>;</li>target="RFC9331" format="default"/>.</li> <li>Hybrid ECN/drop rate policers (see <xref target="l4s_arch_sec_policing" format="default"/>).</li> </ul> <t>L4S deployment scenarios that minimize these issues(e.g. over(e.g., over wireline networks) can proceed in parallel to this research, in the expectation that research success could continually widen L4S applicability.</t> </section> <section anchor="l4sarch_sec_classic-ecn-neck" numbered="true" toc="default"> <name>L4S Flow but Classic ECN Bottleneck</name> <t>Classic ECN support is starting to materialize on the Internet as an increased level of CE marking. It is hard to detect whether this is all due to the addition of support for ECN in implementations of FQ-CoDel and/or FQ-COBALT, which is not generally problematic, becauseflow-queueflow queue (FQ) scheduling inherently prevents a flow from exceeding the 'fair' rate irrespective of its aggressiveness. However, some of this Classic ECN marking might be due to single-queue ECN deployment. This case is discussed inSection 4.3 of<xref target="RFC9331" format="default" sectionFormat="of" section="4.3"> the L4S ECNspec <xref target="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/>.</t>spec</xref>.</t> </section> <section numbered="true" toc="default"> <name>L4S AQM Deployment within Tunnels</name> <t>An L4S AQM uses the ECN field to signal congestion.So,So in common with Classic ECN, if the AQM is within a tunnel or at a lower layer, correct functioning of ECN signalling requirescorrectstandards-compliant propagation of the ECN field up thelayers <xreflayers <xref target="RFC6040"format="default"/>,format="default"/> <xref target="I-D.ietf-tsvwg-rfc6040update-shim"format="default"/>,format="default"/> <xref target="I-D.ietf-tsvwg-ecn-encap-guidelines" format="default"/>.</t> </section> </section> </section> <section anchor="l4sps_IANA" numbered="true" toc="default"> <name>IANAConsiderations (to be removed by RFC Editor)</name>Considerations</name> <t>Thisspecification containsdocument has no IANAconsiderations.</t>actions.</t> </section> <section anchor="l4sps_Security_Considerations" numbered="true" toc="default"> <name>Security Considerations</name> <section numbered="true" toc="default"> <name>Traffic Rate (Non-)Policing</name><t/><section numbered="true" toc="default"> <name>(Non-)Policing Rate per Flow</name> <t>In the current Internet, ISPs usually enforce separation between the capacity of shared links assigned to different 'sites'(e.g. households, businesses(e.g., households, businesses, or mobile users--- see terminology in <xref target="l4sps_Terminology" format="default"/>) using some form ofscheduler <xrefscheduler <xref target="RFC0970" format="default"/>. And they use varioustechniquestechniques, like redirection to traffic scrubbingfacilitiesfacilities, to deal with flooding attacks. However, there has never been a universal need to police the rate of individual application flows--- the Internet has generally always relied on self-restraint of congestion controls at senders for sharing intra-'site' capacity.</t> <t>L4S has been designed not to upset this status quo. If a DualQ is used to provide L4S service,section 4.2 of<xreftarget="I-D.ietf-tsvwg-aqm-dualq-coupled" format="default"/>target="RFC9332" format="default" sectionFormat="of" section="4.2"/> explains how it is designed to give no more rate advantage to unresponsive flows than a single-queue AQM would, whether or not there is traffic overload.</t> <t>Also, in case per-flow rate policing is ever required, it can be added because it is orthogonal to the distinction between L4S and Classic. As explained in <xref target="l4sps_why-not" format="default"/>, the DualQ variant of L4S provides low delay without prejudging the issue of flow-rate control.So,So if flow-rate control is needed,per-flow-queuingper-flow queuing (FQ) with L4S support can be used instead, or flow rate policing can be added as a modular addition to a DualQ. However, per-flow rate control is not usually deployed as a security mechanism, because an active attacker can just shard its traffic over more flowIDsidentifiers if the rate of each is restricted.</t> </section> <section numbered="true" toc="default"> <name>(Non-)Policing L4S Service Rate</name> <t><xref target="l4sps_why-not" format="default"/> explains how Diffserv only makes a difference if some packets get less favourable treatment than others, which typically requires traffic rate policing for a low latency class. In contrast, it should not be necessary to rate-police access to the L4S service to protect the Classic service, because L4S is designed to reduce delay without harming the delay or rate of any Classic traffic. </t> <t>During early deployment (and perhaps always), some networks will not offer the L4S service. In general, these networks should not need to police L4S traffic. They are required (by both the ECNspec <xrefspec <xref target="RFC3168" format="default"/> and the L4S ECNspec <xref target="I-D.ietf-tsvwg-ecn-l4s-id"spec <xref target="RFC9331" format="default"/>) not to change the L4S identifier, which would interfere with end-to-end congestion control. If they already treat ECN traffic as Not-ECT, they can merely treat L4S traffic as Not-ECT too. At a bottleneck, such networks will introduce some queuing and dropping. When ascalableScalable congestion control detects adropdrop, it will have to respond safely with respect to Classic congestion controls (as required inSection 4.3 of<xreftarget="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/>).target="RFC9331" format="default" sectionFormat="of" section="4.3"/>). This will degrade the L4S service to be no better (but never worse) than Classic bestefforts,efforts whenever a non-ECN bottleneck is encountered on a path (see <xref target="l4sarch_sec_non-l4s-neck" format="default"/>).</t> <t>In cases that are expected to be rare, networks that solely support ClassicECN <xrefECN <xref target="RFC3168" format="default"/> in a single queue bottleneck might opt to police L4S traffic so as to protect competing Classic ECN traffic (for instance, seeSection 6.1.3 of the<xref target="I-D.ietf-tsvwg-l4sops" format="default" sectionFormat="of" section="6.1.3">the L4S operationalguidance <xref target="I-D.ietf-tsvwg-l4sops" format="default"/>).guidance</xref>). However,Section 4.3 of<xref target="RFC9331" format="default" sectionFormat="of" section="4.3"> the L4S ECNspec <xref target="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/>spec</xref> recommends that the sender adapts its congestion response to properly coexist with Classic ECN flows,i.e. revertingi.e., reverting to the self-restraint approach.</t> <t>Certain network operators might choose to restrict access to the L4S service, perhaps only to selected premium customers as a value-added service. Their packet classifier (item 2 in <xref target="l4sps_fig_components" format="default"/>) could identify such customers against some other field(e.g. source(e.g., source addressrange)range), as well as classifying on the ECN field. If only the ECN L4S identifier matched, but not (say) the sourceaddress (say),address, the classifier could direct these packets (from non-premium customers) into the Classic queue. Explaining clearly how operators can use additional local classifiers (seesection 5.4 of the L4S ECN spec <xref target="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/>)<xref target="RFC9331" section="5.4" sectionFormat="of"/>) is intended to remove any motivation to clear the L4S identifier. Then at least the L4S ECN identifier will be more likely to surviveend-to-endend to end, even though the service may not be supported at every hop. Such local arrangements would only require simple registered/not-registered packet classification, rather than the managed, application-specific traffic policing against customer-specific traffic contracts that Diffserv uses.</t> </section> </section> <section numbered="true" toc="default"> <name>'Latency Friendliness'</name> <t>Like the Classic service, the L4S service relies on self-restraint- limitingto limit the rate in response to congestion. In addition, the L4S service requires self-restraint in terms of limiting latency (burstiness). It is hoped that self-interest and guidance on dynamic behaviour (especially flow start-up, which might need to be standardized) will be sufficient to prevent transports from sending excessive bursts of L4S traffic, given the application's own latency will suffer most from such behaviour.</t> <t>Because the L4S service can reduce delay without discernibly increasing the delay of any Classic traffic, it should not be necessary to police L4S traffic to protect the delay ofClassic.Classic traffic. However, whether burst policing becomes necessary to protect other L4S traffic remains to be seen. Without it, there will be potential for attacks on the low latency of the L4S service.</t> <t>If needed, various arrangements could be used to address this concern:</t> <dl newline="false" spacing="normal"> <dt>Local bottleneck queue protection:</dt> <dd>A per-flow (5-tuple) queue protectionfunction <xreffunction <xref target="I-D.briscoe-docsis-q-protection" format="default"/> has been developed for the low latency queue in DOCSIS, which has adopted the DualQ L4S architecture. It protects the low latency service from any queue-building flows that accidentally or maliciously classify themselves into the low latency queue. It is designed to score flows based solely on their contribution to queuing (not flow rate in itself). Then, if the shared low latency queue is at risk of exceeding a threshold, the function redirects enough packets of the highest scoring flow(s) into the Classic queue to preserve low latency.</dd> <dt>Distributed traffic scrubbing:</dt> <dd>Rather than policing locally at each bottleneck, it may only be necessary to address problems reactively,e.g. punitivelye.g., punitively target any deployments of new bursty malware, in a similar way to how traffic from flooding attack sources is rerouted via scrubbing facilities.</dd> <dt>Local bottleneck per-flow scheduling:</dt> <dd>Per-flow scheduling should inherently isolate non-bursty flows from bursty flows (see <xref target="l4sps_why-not" format="default"/> for discussion of the merits of per-flow scheduling relative to per-flow policing).</dd> <dt>Distributed access subnet queue protection:</dt> <dd>Per-flow queue protection could be arranged for a queue structure distributed across a subnet intercommunicating using lower layer control messages (see Section 2.1.4 of <xref target="QDyn" format="default"/>). For instance, in a radio access network, user equipment already sends regular buffer status reports to a radio network controller, which could use this information to remotely police individual flows.</dd> <dt>Distributed Congestion Exposure toIngress Policers:</dt>ingress policers:</dt> <dd>The Congestion Exposure (ConEx)architecture <xrefarchitecture <xref target="RFC7713" format="default"/> uses an egress audit to motivate senders to truthfully signal path congestionin-bandin-band, where it can be used by ingress policers. An edge-to-edge variant of this architecture is also possible.</dd> <dt>DistributedDomain-edgedomain-edge traffic conditioning:</dt> <dd>An architecture similar toDiffserv <xrefDiffserv <xref target="RFC2475" format="default"/> may be preferred, where traffic is proactively conditioned on entry to a domain, rather than reactively policed only if it leads to queuing once combined with other traffic at a bottleneck.</dd> <dt>Distributed core network queue protection:</dt> <dd>The policing function could be divided between per-flow mechanisms at the network ingress that characterize the burstiness of each flow into a signal carried with thetraffic,traffic and per-class mechanisms at bottlenecks that act on these signals if queuing actually occurs once the traffic converges. This would be somewhat similar to <xref target="Nadas20" format="default"/>, which is in turn similar to the idea behind core stateless fair queuing.</dd> </dl> <t>No single one of these possible queue protection capabilities is considered an essential part of the L4S architecture, which works without any of them under non-attack conditions (much as the Internet normally works without per-flow rate policing). Indeed, even where latency policers are deployed, under normalcircumstancescircumstances, they would not intervene, and if operators found they were notnecessarynecessary, they could disable them. Part of the L4S experiment will be to see whether such a function isnecessary,necessary and which arrangements are most appropriate to the size of the problem.</t> </section> <section anchor="l4s_arch_sec_policing" numbered="true" toc="default"> <name>Interaction between Rate Policing and L4S</name> <t>As mentioned in <xref target="l4sps_why-not" format="default"/>, L4S should remove the need for low latency Diffserv classes. However, those Diffserv classes that give certain applications or users priority overcapacity,capacity would still be applicable in certain scenarios(e.g. corporate(e.g., corporate networks). Then, within such Diffserv classes, L4S would often be applicable to give traffic low latency and low loss as well. Within such a Diffserv class, the bandwidth available to a user or application is often limited by a rate policer. Similarly, in the default Diffserv class, rate policers are sometimes used to partition shared capacity.</t> <t>AclassicClassic rate policer drops any packets exceeding a set rate, usually also giving a burst allowance (variants exist where the policer re-marksnon-compliantnoncompliant traffic to a discard-eligible Diffserv codepoint, so they can be dropped elsewhere during contention). Whenever L4S traffic encounters one of these rate policers, it will experience drops and the source will have to fall back to a Classic congestion control, thus losing the benefits of L4S (<xref target="l4sarch_sec_non-l4s-neck" format="default"/>).So,So in networks that already use rate policers and plan to deploy L4S, it will be preferable to redesign these rate policers to be more friendly to the L4S service.</t> <t>L4S-friendly rate policing is currently a research area (note that this is not the same as latency policing). It might be achieved by setting a threshold where ECN marking is introduced, such that it is just under the policed rate or just under the burst allowance where drop is introduced. Forinstanceinstance, thetwo-ratetwo-rate, three-colourmarker <xrefmarker <xref target="RFC2698" format="default"/> or a PCN threshold and excess-ratemarker <xrefmarker <xref target="RFC5670" format="default"/> could mark ECN at the lower rate and drop at the higher. Or an existing rate policer could have congestion-rate policing added,e.g. usinge.g., using the 'local' (non-ConEx) variant of the ConEx aggregate congestionpolicer <xrefpolicer <xref target="I-D.briscoe-conex-policing" format="default"/>. It might also be possible to designscalableScalable congestion controls to respond less catastrophically to loss that has not been preceded by a period of increasing delay.</t> <t>The design of L4S-friendly rate policers will require aseparateseparate, dedicated document. For further discussion of the interaction between L4S and Diffserv, see <xref target="I-D.briscoe-tsvwg-l4s-diffserv" format="default"/>.</t> </section> <section numbered="true" toc="default"> <name>ECN Integrity</name> <t>Various ways have been developed to protect the integrity of the congestion feedback loop (whether signalled by loss, ClassicECNECN, or L4S ECN) against misbehaviour by the receiver,sendersender, or network (or all three). Brief details ofeacheach, including applicability,prospros, andcons iscons, are given inAppendix C.1 of the<xref target="RFC9331" format="default" sectionFormat="of" section="C.1">the L4S ECNspec <xref target="I-D.ietf-tsvwg-ecn-l4s-id" format="default"/>.</t>spec</xref>.</t> </section> <section numbered="true" toc="default"> <name>Privacy Considerations</name> <t>As discussed in <xref target="l4sps_why-not" format="default"/>, the L4S architecture does not preclude approaches that inspect end-to-end transport layer identifiers. For instance, L4S support has been added to FQ-CoDel, which classifies by application flowIDidentifier in the network. However, the main innovation of L4S is the DualQ AQM framework that does not need to inspect any deeper than the outermost IP header, because the L4S identifier is in the IP-ECN field.</t> <t>Thus, the L4S architecture enables very low queuing delay without <em>requiring</em> inspection of information above the IP layer. This means that users who want to encrypt application flow identifiers,e.g. in IPSece.g., in IPsec or other encrypted VPN tunnels, don't have to sacrifice lowdelay <xrefdelay <xref target="RFC8404" format="default"/>.</t> <t>Because L4S can provide low delay for a broad set of applications that choose to use it, there is no need for individual applications or classes within that broad set to be distinguishable in any way while traversing networks. This removes much of the ability to correlate between the delay requirements of traffic and other identifyingfeatures <xreffeatures <xref target="RFC6973" format="default"/>. There may be some types of traffic that prefer not to use L4S, but the coarse binary categorization of traffic reveals very little that could be exploited to compromise privacy.</t> </section> </section> </middle><!-- *****BACK MATTER ***** --><back> <displayreference target="I-D.ietf-tcpm-accurate-ecn" to="ACCECN"/> <displayreference target="I-D.ietf-tsvwg-nqb" to="NQB-PHB"/> <displayreference target="I-D.briscoe-conex-policing" to="CONG-POLICING"/> <displayreference target="I-D.stewart-tsvwg-sctpecn" to="ECN-SCTP"/> <displayreference target="I-D.sridharan-tcpm-ctcp" to="CTCP"/> <displayreference target="I-D.ietf-tsvwg-rfc6040update-shim" to="ECN-SHIM"/> <displayreference target="I-D.ietf-tsvwg-ecn-encap-guidelines" to="ECN-ENCAP"/> <displayreference target="I-D.ietf-tsvwg-l4sops" to="L4SOPS"/> <displayreference target="I-D.briscoe-tsvwg-l4s-diffserv" to="L4S-DIFFSERV"/> <displayreference target="I-D.briscoe-docsis-q-protection" to="DOCSIS-Q-PROT"/> <displayreference target="I-D.cardwell-iccrg-bbr-congestion-control" to="BBR-CC"/> <displayreference target="I-D.briscoe-iccrg-prague-congestion-control" to="PRAGUE-CC"/> <displayreference target="I-D.morton-tsvwg-codel-approx-fair" to="CODEL-APPROX-FAIR"/> <displayreference target="I-D.mathis-iccrg-relentless-tcp" to="RELENTLESS"/> <references> <name>Informative References</name> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.0970.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2475.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2698.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2884.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"/> <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.4340.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.5033.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.5670.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.6040.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6817.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6973.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.7665.xml"/> <xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.ietf-tcpm-accurate-ecn.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.7567.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.8034.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8170.xml"/> <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.8290.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8298.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8311.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.8404.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.8888.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.9000.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.9113.xml"/> <referenceanchor="RFC0970" target="https://www.rfc-editor.org/info/rfc970" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.0970.xml"> <front> <title>On Packet Switches With Infinite Storage</title> <author initials="J." surname="Nagle" fullname="J. Nagle"> <organization/> </author> <date year="1985" month="December"/> <abstract> <t>The purpose of this RFC is to focus discussion on a particular problem in the ARPA-Internet and possible methods of solution. Most prior work on congestion in datagram systems focuses on buffer management. In this memo the case of a packet switch with infinite storage is considered. Such a packet switch can never run out of buffers. It can, however, still become congested. The meaning of congestion in an infinite-storage system is explored. An unexpected result is found that shows a datagram network with infinite storage, first-in-first-out queuing, at least two packet switches, and a finite packet lifetime will, under overload, drop all packets. By attacking the problem of congestion for the infinite-storage case, new solutions applicable to switches with finite storage may be found. No proposed solutions this document are intended as standards for the ARPA-Internet at this time.</t> </abstract> </front> <seriesInfo name="RFC" value="970"/> <seriesInfo name="DOI" value="10.17487/RFC0970"/> </reference> <reference anchor="RFC2475" target="https://www.rfc-editor.org/info/rfc2475" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.2475.xml"> <front> <title>An Architecture for Differentiated Services</title> <author initials="S." surname="Blake" fullname="S. Blake"> <organization/> </author> <author initials="D." surname="Black" fullname="D. Black"> <organization/> </author> <author initials="M." surname="Carlson" fullname="M. Carlson"> <organization/> </author> <author initials="E." surname="Davies" fullname="E. Davies"> <organization/> </author> <author initials="Z." surname="Wang" fullname="Z. Wang"> <organization/> </author> <author initials="W." surname="Weiss" fullname="W. Weiss"> <organization/> </author> <date year="1998" month="December"/> <abstract> <t>This document defines an architecture for implementing scalable service differentiation in the Internet. This memo provides information for the Internet community.</t> </abstract> </front> <seriesInfo name="RFC" value="2475"/> <seriesInfo name="DOI" value="10.17487/RFC2475"/> </reference> <reference anchor="RFC2698" target="https://www.rfc-editor.org/info/rfc2698" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.2698.xml"> <front> <title>A Two Rate Three Color Marker</title> <author initials="J." surname="Heinanen" fullname="J. Heinanen"> <organization/> </author> <author initials="R." surname="Guerin" fullname="R. Guerin"> <organization/> </author> <date year="1999" month="September"/> <abstract> <t>This document defines a Two Rate Three Color Marker (trTCM), which can be used as a component in a Diffserv traffic conditioner. This memo provides information for the Internet community.</t> </abstract> </front> <seriesInfo name="RFC" value="2698"/> <seriesInfo name="DOI" value="10.17487/RFC2698"/> </reference> <reference anchor="RFC2884" target="https://www.rfc-editor.org/info/rfc2884" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.2884.xml"> <front> <title>Performance Evaluation of Explicit Congestion Notification (ECN) in IP Networks</title> <author initials="J." surname="Hadi Salim" fullname="J. Hadi Salim"> <organization/> </author> <author initials="U." surname="Ahmed" fullname="U. Ahmed"> <organization/> </author> <date year="2000" month="July"/> <abstract> <t>This memo presents a performance study of the Explicit Congestion Notification (ECN) mechanism in the TCP/IP protocol using our implementation on the Linux Operating System. This memo provides information for the Internet community.</t> </abstract> </front> <seriesInfo name="RFC" value="2884"/> <seriesInfo name="DOI" value="10.17487/RFC2884"/> </reference> <reference anchor="RFC3168" target="https://www.rfc-editor.org/info/rfc3168" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.3168.xml"> <front> <title>The Addition of Explicit Congestion Notification (ECN) to IP</title> <author initials="K." surname="Ramakrishnan" fullname="K. Ramakrishnan"> <organization/> </author> <author initials="S." surname="Floyd" fullname="S. Floyd"> <organization/> </author> <author initials="D." surname="Black" fullname="D. Black"> <organization/> </author> <date year="2001" month="September"/> <abstract> <t>This memo specifies the incorporation of ECN (Explicit Congestion Notification) to TCP and IP, including ECN's use of two bits in the IP header. [STANDARDS-TRACK]</t> </abstract> </front> <seriesInfo name="RFC" value="3168"/> <seriesInfo name="DOI" value="10.17487/RFC3168"/> </reference> <reference anchor="RFC4774" target="https://www.rfc-editor.org/info/rfc4774" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.4774.xml"> <front> <title>Specifying Alternate Semantics for the Explicit Congestion Notification (ECN) Field</title> <author initials="S." surname="Floyd" fullname="S. Floyd"> <organization/> </author> <date year="2006" month="November"/> <abstract> <t>There have been a number of proposals for alternate semantics for the Explicit Congestion Notification (ECN) field in the IP header RFC 3168. This document discusses some of the issues in defining alternate semantics for the ECN field, and specifies requirements for a safe coexistence in an Internet that could include routers that do not understand the defined alternate semantics. This document evolved as a result of discussions with the authors of one recent proposal for such alternate semantics. This document specifies an Internet Best Current Practices for the Internet Community, and requests discussion and suggestions for improvements.</t> </abstract> </front> <seriesInfo name="BCP" value="124"/> <seriesInfo name="RFC" value="4774"/> <seriesInfo name="DOI" value="10.17487/RFC4774"/> </reference> <reference anchor="RFC6679" target="https://www.rfc-editor.org/info/rfc6679" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.6679.xml"> <front> <title>Explicit Congestion Notification (ECN) for RTP over UDP</title> <author initials="M." surname="Westerlund" fullname="M. Westerlund"> <organization/> </author> <author initials="I." surname="Johansson" fullname="I. Johansson"> <organization/> </author> <author initials="C." surname="Perkins" fullname="C. Perkins"> <organization/> </author> <author initials="P." surname="O'Hanlon" fullname="P. O'Hanlon"> <organization/> </author> <author initials="K." surname="Carlberg" fullname="K. Carlberg"> <organization/> </author> <date year="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> </front> <seriesInfo name="RFC" value="6679"/> <seriesInfo name="DOI" value="10.17487/RFC6679"/> </reference> <reference anchor="RFC3540" target="https://www.rfc-editor.org/info/rfc3540" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.3540.xml"> <front> <title>Robust Explicit Congestion Notification (ECN) Signaling with Nonces</title> <author initials="N." surname="Spring" fullname="N. Spring"> <organization/> </author> <author initials="D." surname="Wetherall" fullname="D. Wetherall"> <organization/> </author> <author initials="D." surname="Ely" fullname="D. Ely"> <organization/> </author> <date year="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> </front> <seriesInfo name="RFC" value="3540"/> <seriesInfo name="DOI" value="10.17487/RFC3540"/> </reference> <reference anchor="RFC3246" target="https://www.rfc-editor.org/info/rfc3246" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.3246.xml"> <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> <author initials="K." surname="Benson" fullname="K. Benson"> <organization/> </author> <author initials="J.Y." surname="Le Boudec" fullname="J.Y. Le Boudec"> <organization/> </author> <author initials="W." surname="Courtney" fullname="W. Courtney"> <organization/> </author> <author 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"> <organization/> </author> <date year="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> </front> <seriesInfo name="RFC" value="3246"/> <seriesInfo name="DOI" value="10.17487/RFC3246"/> </reference> <reference anchor="RFC3649" target="https://www.rfc-editor.org/info/rfc3649" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.3649.xml"> <front> <title>HighSpeed TCP for Large Congestion Windows</title> <author initials="S." surname="Floyd" fullname="S. Floyd"> <organization/> </author> <date year="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 modification to TCP's congestion control mechanism for use with TCP connections with large congestion windows. The congestion control mechanisms of the 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 packets and a 100 ms round-trip time, achieving a steady-state throughput of 10 Gbps would require an average congestion window of 83,333 segments, and a 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 limitation of TCP, this document proposes HighSpeed TCP, and solicits experimentation and feedback from the wider community.</t> </abstract> </front> <seriesInfo name="RFC" value="3649"/> <seriesInfo name="DOI" value="10.17487/RFC3649"/> </reference> <reference anchor="RFC4340" target="https://www.rfc-editor.org/info/rfc4340" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.4340.xml"> <front> <title>Datagram Congestion Control Protocol (DCCP)</title> <author initials="E." surname="Kohler" fullname="E. Kohler"> <organization/> </author> <author initials="M." surname="Handley" fullname="M. Handley"> <organization/> </author> <author initials="S." surname="Floyd" fullname="S. Floyd"> <organization/> </author> <date year="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> </front> <seriesInfo name="RFC" value="4340"/> <seriesInfo name="DOI" value="10.17487/RFC4340"/> </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="RFC5033" target="https://www.rfc-editor.org/info/rfc5033" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.5033.xml"> <front> <title>Specifying New Congestion Control Algorithms</title> <author initials="S." surname="Floyd" fullname="S. Floyd"> <organization/> </author> <author initials="M." surname="Allman" fullname="M. Allman"> <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 of this 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> </front> <seriesInfo name="BCP" value="133"/> <seriesInfo name="RFC" value="5033"/> <seriesInfo name="DOI" value="10.17487/RFC5033"/> </reference> <reference anchor="RFC5348" target="https://www.rfc-editor.org/info/rfc5348" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.5348.xml"> <front> <title>TCP Friendly Rate Control (TFRC): Protocol Specification</title> <author initials="S." surname="Floyd" fullname="S. Floyd"> <organization/> </author> <author initials="M." surname="Handley" fullname="M. Handley"> <organization/> </author> <author initials="J." surname="Padhye" fullname="J. Padhye"> <organization/> </author> <author initials="J." surname="Widmer" fullname="J. Widmer"> <organization/> </author> <date year="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> </front> <seriesInfo name="RFC" value="5348"/> <seriesInfo name="DOI" value="10.17487/RFC5348"/> </reference> <reference anchor="RFC5670" target="https://www.rfc-editor.org/info/rfc5670" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.5670.xml"> <front> <title>Metering and Marking Behaviour of PCN-Nodes</title> <author initials="P." surname="Eardley" fullname="P. Eardley" role="editor"> <organization/> </author> <date year="2009" month="November"/> <abstract> <t>The objective of Pre-Congestion Notification (PCN) is to protect the quality of service (QoS) of inelastic flows within a Diffserv domain in a simple, scalable, and robust fashion. This document defines the two metering and marking behaviours of PCN-nodes. Threshold-metering and -marking marks all PCN-packets if the rate of PCN-traffic is greater than a configured rate ("PCN-threshold-rate"). Excess- traffic-metering and -marking marks a proportion of PCN-packets, such that the amount marked equals the rate of PCN-traffic in excess of a configured rate ("PCN-excess-rate"). The level of marking allows PCN-boundary-nodes to make decisions about whether to admit or terminate PCN-flows. [STANDARDS-TRACK]</t> </abstract> </front> <seriesInfo name="RFC" value="5670"/> <seriesInfo name="DOI" value="10.17487/RFC5670"/> </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="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="RFC6817" target="https://www.rfc-editor.org/info/rfc6817" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.6817.xml"> <front> <title>Low Extra Delay Background Transport (LEDBAT)</title> <author initials="S." surname="Shalunov" fullname="S. Shalunov"> <organization/> </author> <author initials="G." surname="Hazel" fullname="G. Hazel"> <organization/> </author> <author initials="J." surname="Iyengar" fullname="J. Iyengar"> <organization/> </author> <author initials="M." surname="Kuehlewind" fullname="M. Kuehlewind"> <organization/> </author> <date year="2012" month="December"/> <abstract> <t>Low Extra Delay Background Transport (LEDBAT) is an experimental delay-based congestion control algorithm that seeks to utilize the available bandwidth on an end-to-end path while limiting the consequent increase in queueing delay on that path. LEDBAT uses changes in one-way delay measurements to limit congestion that the flow itself induces in the network. LEDBAT is designed for use by background bulk-transfer applications to be no more aggressive than standard TCP congestion control (as specified in RFC 5681) and to yield in the presence of competing flows, thus limiting interference with the network performance of competing flows. This document defines an Experimental Protocol for the Internet community.</t> </abstract> </front> <seriesInfo name="RFC" value="6817"/> <seriesInfo name="DOI" value="10.17487/RFC6817"/> </reference> <reference anchor="RFC6973" target="https://www.rfc-editor.org/info/rfc6973" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.6973.xml"> <front> <title>Privacy Considerations for Internet Protocols</title> <author initials="A." surname="Cooper" fullname="A. Cooper"> <organization/> </author> <author initials="H." surname="Tschofenig" fullname="H. Tschofenig"> <organization/> </author> <author initials="B." surname="Aboba" fullname="B. Aboba"> <organization/> </author> <author initials="J." surname="Peterson" fullname="J. Peterson"> <organization/> </author> <author initials="J." surname="Morris" fullname="J. Morris"> <organization/> </author> <author initials="M." surname="Hansen" fullname="M. Hansen"> <organization/> </author> <author initials="R." surname="Smith" fullname="R. Smith"> <organization/> </author> <date year="2013" month="July"/> <abstract> <t>This document offers guidance for developing privacy considerations for inclusion in protocol specifications. It aims to make designers, implementers, and users of Internet protocols aware of privacy-related design choices. It suggests that whether any individual RFC warrants a specific privacy considerations section will depend on the document's content.</t> </abstract> </front> <seriesInfo name="RFC" value="6973"/> <seriesInfo name="DOI" value="10.17487/RFC6973"/> </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="RFC7665" target="https://www.rfc-editor.org/info/rfc7665" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.7665.xml"> <front> <title>Service Function Chaining (SFC) Architecture</title> <author initials="J." surname="Halpern" fullname="J. Halpern" role="editor"> <organization/> </author> <author initials="C." surname="Pignataro" fullname="C. Pignataro" role="editor"> <organization/> </author> <date year="2015" month="October"/> <abstract> <t>This document describes an architecture for the specification, creation, and ongoing maintenance of Service Function Chains (SFCs) in a network. It includes architectural concepts, principles, and components used in the construction of composite services through deployment of SFCs, with a focus on those to be standardized in the IETF. This document does not propose solutions, protocols, or extensions to existing protocols.</t> </abstract> </front> <seriesInfo name="RFC" value="7665"/> <seriesInfo name="DOI" value="10.17487/RFC7665"/> </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">anchor="RFC9332" target="https://www.rfc-editor.org/info/rfc9332"> <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<title>Dual-Queue Coupled Active Queue Management (AQM) forTCP 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) orLowLatencyLatency, LowLossLoss, and 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="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>(L4S)</title> <authorinitials="M." surname="Mathis" fullname="M. Mathis"> <organization/>initials="K" surname="De Schepper" fullname="Koen De Schepper"> <organization>Nokia Bell Labs</organization> </author> <authorinitials="B."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="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"fullname="Bob Briscoe" 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="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/><organization>Independent</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="RFC8034" target="https://www.rfc-editor.org/info/rfc8034" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8034.xml"> <front> <title>Active Queue Management (AQM) Based on Proportional Integral Controller Enhanced PIE) for Data-Over-Cable Service Interface Specifications (DOCSIS) Cable Modems</title><authorinitials="G."initials="G" surname="White"fullname="G.fullname="Greg White"><organization/> </author> <author initials="R." surname="Pan" fullname="R. Pan"> <organization/> </author> <date year="2017" month="February"/> <abstract> <t>Cable modems based on Data-Over-Cable Service Interface Specifications (DOCSIS) provide broadband Internet access to over one hundred million users worldwide. In some cases, the cable modem connection is the bottleneck (lowest speed) link between the customer and the Internet. As a result, the impact of buffering and bufferbloat in the cable modem can have a significant effect on user experience. The CableLabs DOCSIS 3.1 specification introduces requirements for cable modems to support an Active Queue Management (AQM) algorithm that is intended to alleviate the impact that buffering has on latency-sensitive traffic, while preserving bulk throughput performance. In addition, the CableLabs DOCSIS 3.0 specifications have also been amended to contain similar requirements. This document describes the requirements on AQM that apply to DOCSIS equipment, including a description of the "DOCSIS-PIE" algorithm that is required on DOCSIS 3.1 cable modems.</t> </abstract> </front> <seriesInfo name="RFC" value="8034"/> <seriesInfo name="DOI" value="10.17487/RFC8034"/> </reference> <reference anchor="RFC8170" target="https://www.rfc-editor.org/info/rfc8170" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8170.xml"> <front> <title>Planning for Protocol Adoption and Subsequent Transitions</title> <author initials="D." surname="Thaler" fullname="D. Thaler" role="editor"> <organization/> </author> <date year="2017" month="May"/> <abstract> <t>Over the many years since the introduction of the Internet Protocol, we have seen a number of transitions throughout the protocol stack, such as deploying a new protocol, or updating or replacing an existing protocol. Many protocols and technologies were not designed to enable smooth transition to alternatives or to easily deploy extensions; thus, some transitions, such as the introduction of IPv6, have been difficult. This document attempts to summarize some basic principles to enable future transitions, and it also summarizes what makes for a good transition plan.</t> </abstract> </front> <seriesInfo name="RFC" value="8170"/> <seriesInfo name="DOI" value="10.17487/RFC8170"/> </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/><organization>CableLabs</organization> </author> <dateyear="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>month="January" year="2023"/> </front> <seriesInfo name="RFC"value="8257"/>value="9332"/> <seriesInfo name="DOI"value="10.17487/RFC8257"/>value="10.17487/RFC9332"/> </reference> <referenceanchor="RFC8290" target="https://www.rfc-editor.org/info/rfc8290" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8290.xml">anchor="RFC9331" target="https://www.rfc-editor.org/info/rfc9331"> <front> <title>TheFlow 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="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="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="RFC8404" target="https://www.rfc-editor.org/info/rfc8404" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8404.xml"> <front> <title>Effects of Pervasive Encryption on Operators</title> <author initials="K." surname="Moriarty" fullname="K. Moriarty" role="editor"> <organization/> </author> <author initials="A." surname="Morton" fullname="A. Morton" role="editor"> <organization/> </author> <date year="2018" month="July"/> <abstract> <t>Pervasive monitoring attacks on the privacy of Internet users are of serious concern to both user and operator communities. RFC 7258 discusses the critical need to protect users' privacy when developing IETF specifications and also recognizes that making networks unmanageable to mitigate pervasive monitoring is not an acceptable outcome: an appropriate balance is needed. This document discusses current security and network operations as well as management practices that may be impacted by the shift to increased use of encryption to help guide protocol development in support of manageable and secure networks.</t> </abstract> </front> <seriesInfo name="RFC" value="8404"/> <seriesInfo name="DOI" value="10.17487/RFC8404"/> </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="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 ControlProtocol(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="RFC9113" target="https://www.rfc-editor.org/info/rfc9113" xml:base="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.9113.xml"> <front> <title>HTTP/2</title> <author initials="M." surname="Thomson" fullname="M. Thomson" role="editor"> <organization/> </author> <author initials="C." surname="Benfield" fullname="C. Benfield" role="editor"> <organization/> </author> <date year="2022" month="June"/> <abstract> <t>This specification describes an optimized expression of the semantics of the Hypertext Transfer Protocol (HTTP), referred to as HTTP version 2 (HTTP/2). HTTP/2 enables a more efficient use of network resources and a reduced latency by introducing field compression and allowing multiple concurrent exchanges on the same connection.</t> <t>This document obsoletes RFCs 7540 and 8740.</t> </abstract> </front> <seriesInfo name="RFC" value="9113"/> <seriesInfo name="DOI" value="10.17487/RFC9113"/> </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 AQMsfor Low Latency, LowLossLoss, and Scalable Throughput (L4S)</title> <author initials="K" surname="De Schepper" fullname="Koen DeSchepper"/> <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.ietf-tsvwg-ecn-l4s-id" target="https://datatracker.ietf.org/api/v1/doc/document/draft-ietf-tsvwg-ecn-l4s-id/" xml:base="https://bib.ietf.org/public/rfc/bibxml-ids/reference.I-D.ietf-tsvwg-ecn-l4s-id.xml"> <front> <title>Explicit Congestion Notification (ECN) Protocol for Very Low Queuing Delay (L4S)</title> <author fullname="Koen De Schepper"/> <author fullname="Bob Briscoe"/> <date day="8" month="August" year="2022"/> <abstract> <t>This specification defines the protocol to be used for a new network service called low latency, low loss and scalable 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' congestion control, which induces much more frequent control signals from the network and it responds to them with much more fine-grained adjustments, so that very low (typically sub-millisecond on average) and consistently low queuing delay becomes possible for L4S traffic without compromising link utilization. Thus 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) traffic. Then, network bottlenecks can be incrementally modified to distinguish and isolate existing traffic that still follows the Classic behaviour, to prevent it degrading the low queuing delay and low loss of L4S traffic. This experimental track 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 new active queue management (AQM) marking algorithms and examples of new transports (whether TCP- like or real-time) are specified separately.</t> </abstract> </front> <seriesInfo name="Internet-Draft" value="draft-ietf-tsvwg-ecn-l4s-id-28"/> </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.briscoe-conex-policing" target="https://www.ietf.org/archive/id/draft-briscoe-conex-policing-01.txt" xml:base="https://bib.ietf.org/public/rfc/bibxml-ids/reference.I-D.briscoe-conex-policing.xml"> <front> <title>Network Performance Isolation using Congestion Policing</title>Schepper"> <organization>Nokia Bell Labs</organization> </author> <author initials="B" surname="Briscoe" fullname="BobBriscoe"/> <date day="14" month="February" year="2014"/> <abstract> <t>This document describes why policing using congestion information can isolate users from network performance degradation due to each other's usage, but without losing the multiplexing benefits of a LAN- style network where anyone can use any amount of any resource. Extensive numerical examples and diagrams are given. The document is agnostic to how the congestion information reaches the policer. The congestion exposure (ConEX) protocol is recommended, but other tunnel feedback mechanisms have been proposed.</t> </abstract> </front> <seriesInfo name="Internet-Draft" value="draft-briscoe-conex-policing-01"/> </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"/>Briscoe" role="editor"> <organization>Independent</organization> </author> <dateday="15"month="January"year="2014"/> <abstract> <t>This document describes the addition of the ECN to the Stream Control Transmission Protocol (SCTP).</t> </abstract>year="2023"/> </front> <seriesInfoname="Internet-Draft" value="draft-stewart-tsvwg-sctpecn-05"/> </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.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-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>name="RFC" value="9331"/> <seriesInfoname="Internet-Draft" value="draft-ietf-tsvwg-ecn-encap-guidelines-17"/>name="DOI" value="10.17487/RFC9331"/> </reference> <xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.ietf-tsvwg-nqb.xml"/> <xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.briscoe-conex-policing.xml"/> <xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.stewart-tsvwg-sctpecn.xml"/> <xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.sridharan-tcpm-ctcp.xml"/> <xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.ietf-tsvwg-rfc6040update-shim.xml"/> <xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.ietf-tsvwg-ecn-encap-guidelines.xml"/> <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">target="https://datatracker.ietf.org/doc/html/draft-ietf-tsvwg-l4sops-03"> <front><title>Operational<title> Operational Guidance for Deployment of L4S in theInternet</title>Internet </title> <author fullname="GregWhite"/>White" initials="G." surname="White" role="editor"> <organization>CableLabs</organization> </author> <dateday="28"month="April" day="28" 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"/> <format type="TXT" target="https://www.ietf.org/archive/id/draft-ietf-tsvwg-l4sops-03.txt"/> </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><xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.briscoe-tsvwg-l4s-diffserv.xml"/> <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">target="https://datatracker.ietf.org/doc/html/draft-briscoe-docsis-q-protection-06"> <front> <title>TheDOCSIS(r)DOCSIS(R) Queue Protection Algorithm to Preserve Low Latency</title> <author initials="B" surname="Briscoe" fullname="BobBriscoe"/>Briscoe" role="editor"> <organization>Independent</organization> </author> <author initials="G" surname="White" fullname="GregWhite"/>White"> <organization>CableLabs</organization> </author> <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.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><xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.cardwell-iccrg-bbr-congestion-control.xml"/> <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">target="https://datatracker.ietf.org/doc/html/draft-briscoe-iccrg-prague-congestion-control-01"> <front> <title>Prague Congestion Control</title> <authorfullname="Koen De Schepper"/>initials="K" surname="De Schepper"> <organization>Nokia Bell Labs</organization> </author> <author fullname="OlivierTilmans"/>Tilmans"> <organization>Nokia Bell Labs</organization> </author> <author fullname="BobBriscoe"/>Briscoe" role="editor"> <organization>Independent</organization> </author> <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.morton-tsvwg-codel-approx-fair" target="https://www.ietf.org/archive/id/draft-morton-tsvwg-codel-approx-fair-01.txt" xml:base="https://bib.ietf.org/public/rfc/bibxml-ids/reference.I-D.morton-tsvwg-codel-approx-fair.xml"> <front> <title>Controlled Delay Approximate Fairness AQM</title> <author fullname="Jonathan Morton"/> <author fullname="Peter G. Heist"/> <date day="9" month="March" year="2020"/> <abstract> <t>This note presents CodelAF, or Controlled Delay Approximate Fairness in full, as an alternative to single-queue AQM or Fair Queue implementations in the low-cost or high-speed network hardware spaces. It builds on the seminal work in Codel [RFC8289], and guides multiple competing flows towards similar throughputs by differential congestion signalling, whilst requiring only a single FIFO queue. It may also be combined with CNQ [I-D.morton-tsvwg-cheap-nasty-queueing] to provide a latency optimisation for sparse flows.</t> </abstract> </front> <seriesInfo name="Internet-Draft" value="draft-morton-tsvwg-codel-approx-fair-01"/> </reference><xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.morton-tsvwg-codel-approx-fair.xml"/> <reference anchor="Hohlfeld14" target="https://doi.acm.org/10.1145/2663716.2663730"> <front> <title>A QoE Perspective on Sizing Network Buffers</title> <author fullname="Oliver Hohlfeld" initials="O." surname="Hohlfeld "> <organization/> </author> <author fullname="Enric Pujol" initials="E." surname="Pujol"> <organization/><address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address></author> <author fullname="Florin Ciucu" initials="F." surname="Ciucu"> <organization/><address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address></author> <author fullname="Anja Feldmann" initials="A." surname="Feldmann"> <organization/><address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address></author> <author fullname="Paul Barford" initials="P." surname="Barford"> <organization/><address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address></author> <date month="November" year="2014"/> </front> <seriesInfoname="Proc. ACMname="DOI" value="10.1145/2663716.2663730"/> <refcontent>IMC '14: Proceedings of the 2014 Conference on InternetMeasurement Conf (IMC'14)" value="hmm"/> </reference> <reference anchor="Mathis09" target="https://www.gdt.id.au/~gdt/presentations/2010-07-06-questnet-tcp/reference-materials/papers/mathis-relentless-congestion-control.pdf"> <front> <title>Relentless Congestion Control</title> <author fullname="Matt Mathis" initials="M." surname="Mathis"> <organization>PSC</organization> </author> <date month="May" year="2009"/> </front> <seriesInfo name="PFLDNeT'09" value=""/>Measurement, pp. 333-346</refcontent> </reference><!--{ToDo: DCttH ref will need to be updated, once stable}--><xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.mathis-iccrg-relentless-tcp.xml"/> <referenceanchor="DCttH19" target="https://bobbriscoe.net/pubs.html#DCttH_TR">anchor="L4Seval22" target="https://arxiv.org/abs/2209.01078"> <front><title>`Data Centre to the Home': Ultra-Low Latency<title>Dual Queue Coupled AQM: Deployable Very Low Queuing Delay for All</title> <author fullname="Koen De Schepper" initials="K." surname="De Schepper"> <organization>Nokia Bell Labs</organization> </author> <author fullname="OlgaBondarenko"Albisser" initials="O."surname="Bondarenko">surname="Albisser"> <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> <datemonth="July" year="2019"/>month="September" year="2022"/> </front> <seriesInfoname="Updated RITE project Technical Report" value=""/>name="DOI" value="10.48550/arXiv.2209.01078"/> <refcontent>TR-BB-2022-001, arXiv:2209.01078 [cs.NI]</refcontent> <formattarget="https://bobbriscoe.net/projects/latency/dctth_journal_draft20190726.pdf"target="https://arxiv.org/pdf/2209.01078" type="PDF"/> </reference> <reference anchor="L4Sdemo16"target="https://dl.acm.org/citation.cfm?doid=2910017.2910633 (videos of demos: https://riteproject.eu/dctth/#1511dispatchwg )">target="https://dl.acm.org/citation.cfm?doid=2910017.2910633"> <front> <title>Ultra-Low Delay for All: Live Experience, Live Analysis</title> <author fullname="Olga Bondarenko" initials="O." surname="Bondarenko"> <organization>Simula Research Lab</organization> </author> <author fullname="Koen De Schepper" initials="K." surname="De Schepper"> <organization>Bell Labs</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> <author fullname="Andreas Petlund" initials="A." surname="Petlund"> <organization></organization> </author> <author fullname="Carstan Griwodz" initials="C." surname="Griwodz"> <organization></organization> </author> <date month="May" year="2016"/> </front> <seriesInfoname="Proc. MMSYS'16" value="pp33:1--33:4"/> <format target="https://dl.acm.org/citation.cfm?doid=2910017.2910633" type="PDF"/>name="DOI" value="10.1145/2910017.2910633"/> <refcontent>Proceedings of the 7th International Conference on Multimedia Systems, Article No. 33, pp. 1-4</refcontent> </reference> <reference anchor="L4Sdemo16-Video" target="https://riteproject.eu/dctth/#1511dispatchwg"> <front> <title>Videos used in IETF dispatch WG 'Ultra-Low Queuing Delay for All Apps' slot</title> <author> </author> </front> </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."initials="M." 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=""/> <format target="https://ee.lbl.gov/papers/congavoid.pdf" type="PDF"/> </reference> <reference anchor="UnorderedLTE"> <front> <title>Implementing immediate forwarding for 4G in a network simulator</title> <author fullname="Magnus Vevik Austrheim"initials="M.V."initials="M." surname="Austrheim"> <organization/> </author> <datemonth="June" year="2019"/>year="2018"/> </front><seriesInfo name="Master's<refcontent>Master's Thesis,Uni Oslo" value=""/>University of Oslo</refcontent> </reference> <reference anchor="PragueLinux" target="https://www.netdevconf.org/0x13/session.html?talk-tcp-prague-l4s"> <front> <title>Implementing the`TCP'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 Research Lab</organization> </author> <author fullname="Joakim Misund" initials="J." surname="Misund"> <organization>Simula Research Lab</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>ETH Zurich</organization> </author> <author fullname="Asad Sajjad Ahmed" initials="A.S." surname="Ahmed"> <organization>Simula Research Lab</organization> </author> <date month="March" year="2019"/> </front><seriesInfo name="Proc.<refcontent>Proceedings Linux Netdev0x13" value=""/>0x13</refcontent> <format target="https://www.files.netdevconf.org/f/4d6939d5f1fb404fafd1/?dl=1" type="PDF"/> </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.<refcontent>Proceedings of Linux Netdev0x13" value=""/>0x13</refcontent> <format target="https://www.files.netdevconf.org/f/febbe8c6a05b4ceab641/?dl=1" type="PDF"/> </reference> <reference anchor="DOCSIS3.1" target="https://specification-search.cablelabs.com/CM-SP-MULPIv3.1"> <front> <title>MAC and Upper Layer Protocols Interface (MULPI) Specification, CM-SP-MULPIv3.1</title> <author fullname="" surname=""> <organization>CableLabs</organization> </author> <date day="21" month="January" year="2019"/> </front> <seriesInfo name="Data-Over-Cable Service Interface SpecificationsDOCSIS®DOCSIS 3.1" value="Version i17 or later"/> </reference> <reference anchor="AFCD" target="https://doi.org/10.1016/j.jnca.2016.03.021"> <front> <title>Towards fair and low latency next generation high speed networks: AFCD queuing</title> <author fullname="Lin Xue" initials="L." surname="Xue"> <organization/> </author> <author fullname="Suman Kumar" initials="S." surname="Kumar"> <organization/><address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address></author> <author fullname="Cheng Cui" initials="C." surname="Cui"> <organization/><address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address></author> <author fullname="Praveenkumar Kondikoppa" initials="P." surname="Kondikoppa"> <organization/><address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address></author> <author fullname="Chui-Hui Chiu" initials="C-H." surname="Chiu"> <organization/><address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address></author> <author fullname="Seung-Jong Park" initials="S-J." surname="Park"> <organization/><address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address></author> <date month="July" year="2016"/> </front> <seriesInfoname="Journalname="DOI" value="10.1016/j.jnca.2016.03.021"/> <refcontent>Journal of Network and ComputerApplications" value="70:183--193"/>Applications, Volume 70, pp. 183-193</refcontent> </reference> <reference anchor="Nadas20" target="https://doi.org/10.1145/3404868.3406669"> <front> <title>A Congestion Control Independent L4S Scheduler</title> <author fullname="Szilveszter Nádas" initials="S." surname="Nádas"> <organization/> </author> <authorfullname="Gergofullname="Gergő Gombos" initials="G." surname="Gombos"> <organization/><address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address></author> <author fullname="Ferenc Fejes" initials="F." surname="Fejes"> <organization/><address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address></author> <author fullname="Sándor Laki" initials="S." surname="Laki"> <organization/><address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address></author> <date month="July" year="2020"/> </front> <seriesInfoname="Proc.name="DOI" value="10.1145/3404868.3406669"/> <refcontent>ANRW '20: Proceedings of the Applied Networking ResearchWorkshop (ANRW '20)" value="45--51"/>Workshop, pp. 45-51</refcontent> </reference> <reference anchor="QDyn" target="https://arxiv.org/abs/1904.07044"> <front> <title>Rapid Signalling of Queue Dynamics</title> <author fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization>bobbriscoe.net Ltd</organization> </author> <datemonth="September" year="2017"/>month="April" year="2019"/> </front> <seriesInfoname="bobbriscoe.net Technical Report" value="TR-BB-2017-001;name="DOI" value="10.48550/arXiv.1904.07044"/> <refcontent>TR-BB-2017-001, arXiv:1904.07044[cs.NI]"/>[cs.NI]</refcontent> <format target="https://arxiv.org/pdf/1904.07044" type="PDF"/> </reference> <reference anchor="McIlroy78" target="https://archive.org/details/bstj57-6-1899"> <front> <title>UNIX Time-Sharing System: Foreword</title> <author fullname="Doug McIlroy" initials="M.D." surname="McIlroy"> <organization/> </author> <author initials="E. N." surname="Pinson"> <organization/><address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address></author> <author initials="B. A." surname="Tague"> <organization/><address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address></author> <date month="July" year="1978"/> </front> <seriesInfoname="Thename="DOI" value="10.1002/j.1538-7305.1978.tb02135.x"/> <refcontent>The Bell System TechnicalJournal" value="57:6(1902--1903)"/>Journal 57: 6, pp. 1899-1904</refcontent> </reference> <reference anchor="LEDBAT_AQM" target="https://ieeexplore.ieee.org/document/8109367"> <front> <title>Characterising LEDBAT Performance Through Bottlenecks Using PIE, FQ-CoDel and FQ-PIE Active Queue Management</title> <author fullname="Rasool Al-Saadi" initials="R." surname="Al-Saadi"> <organization/> </author> <author fullname="Grenville Armitage" initials="G." surname="Armitage"> <organization/><address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address></author> <author fullname="Jason But" initials="J." surname="But"> <organization/><address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address></author> <date month="October" year="2017"/> </front> <seriesInfoname="Proc. IEEEname="DOI" value="10.1109/LCN.2017.22"/> <refcontent>IEEE 42nd Conference on Local Computer Networks(LCN)" value="278--285"/>(LCN)</refcontent> </reference> <reference anchor="BBRv2"target="https://github.com/google/bbr/blob/v2alpha/README.md">target="https://github.com/google/bbr"> <front> <title>TCP BBR v2 Alpha/Preview Release</title><author fullname="Neal Cardwell" initials="N" surname="Cardwell"><author> <organization/> </author><date/><date month="June" year="2022" /> </front><seriesInfo name="GitHub repository;" value="Linux congestion control module"/><refcontent>commit 17700ca</refcontent> </reference> <referenceanchor="SCReAM" target="https://github.com/EricssonResearch/scream/blob/master/README.md">anchor="SCReAM-L4S" target="https://github.com/EricssonResearch/scream"> <front> <title>SCReAM</title><author fullname="Ingemar Johansson" initials="I" surname="Johansson"><author> <organization/> </author><date/><date month="June" year="2022"/> </front><seriesInfo name="GitHub repository;" value=""/> <format target="https://github.com/google/bbr/blob/v2alpha/README.md" type="Source code"/><refcontent>commit fda6c53</refcontent> </reference> <reference anchor="BDPdata" target="https://arxiv.org/abs/2107.01003"> <front> <title>PI2 Parameters</title> <author fullname="Bob Briscoe" initials="B." surname="Briscoe"> <organization/> </author> <datemonth="July"month="October" year="2021"/> </front> <seriesInfoname="Technical Report" value="TR-BB-2021-001name="DOI" value="10.48550/arXiv.2107.01003"/> <refcontent>TR-BB-2021-001, arXiv:2107.01003[cs.NI]"/> <format target="https://arxiv.org/pdf/2107.01003" type="PDF"/>[cs.NI]</refcontent> </reference> <reference anchor="BufferSize" target="https://doi.org/10.1145/1015467.1015499"> <front> <title>Sizing Router Buffers</title> <author fullname="Guido Appenzeller" initials="G." surname="Appenzeller"> <organization>StanfordUni</organization>University</organization> </author> <author fullname="Isaac Keslassy" initials="I." surname="Keslassy"> <organization>StanfordUni</organization>University</organization> </author> <author fullname="Nick McKeown" initials="N." surname="McKeown"> <organization>StanfordUni</organization>University</organization> </author> <datemonth="September"month="October" year="2004"/> </front> <seriesInfoname="In Proc. SIGCOMM'04" value="34(4):281--292"/> <format target="http://yuba.stanford.edu/~nickm/papers/sigcomm2004.pdf" type="PDF"/>name="DOI" value="10.1145/1015467.1015499"/> <refcontent>SIGCOMM '04: Proceedings of the 2004 conference on Applications, technologies, architectures, and protocols for computer communications, pp. 281-292</refcontent> </reference> <reference anchor="COBALT" target="https://ieeexplore.ieee.org/abstract/document/8847054"> <front> <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="Gupta"> <organization/> </author> <author fullname="Pasquale Imputato" initials="P." surname="Imputato"> <organization/> </author> <author fullname="Jonathan Morton" initials="J." surname="Morton"> <organization/> </author> <author 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> <date month="July" year="2019"/> </front> <seriesInfoname="In Proc. IEEE Int'l Symp.name="DOI" value="10.1109/LANMAN.2019.8847054"/> <refcontent>IEEE International Symposium on Local and Metropolitan Area Networks(LANMAN'19)" value="2019:1-6"/> <format target="https://www.researchgate.net/publication/336087375_Design_and_Evaluation_of_COBALT_Queue_Discipline" type="PDF"/>(LANMAN)</refcontent> </reference> <reference anchor="DOCSIS3AQM"target="{https://www.cablelabs.com/wp-content/uploads/2013/11/Active_Queue_Management_Algorithms_DOCSIS_3_0.pdf">target="https://www.cablelabs.com/wp-content/uploads/2013/11/Active_Queue_Management_Algorithms_DOCSIS_3_0.pdf"> <front> <title>Active Queue Management Algorithms for DOCSIS3.0;3.0: A Simulation Study of CoDel, SFQ-CoDel and PIE in DOCSIS 3.0 Networks</title> <author fullname="Greg White" initials="G." surname="White"> <organization/> </author> <date month="April" year="2013"/> </front><seriesInfo name="CableLabs<refcontent>CableLabs TechnicalReport" value=""/>Report</refcontent> </reference> <reference anchor="FQ_CoDel_Thresh" target="https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next.git/commit/?id=dfcb63ce1de6b10b"> <front> <title>fq_codel: generalise ce_threshold marking for subset of traffic</title><author fullname="Toke Høiland-Jørgensen" initials="T." surname="Høiland-Jørgensen"><author> <organization/> </author> <dateday="20"month="October" year="2021"/> </front><seriesInfo name="Linux Patch" value="Commit ID: dfcb63ce1de6b10b"/><refcontent>commit dfcb63ce1de6b10b</refcontent> </reference> <reference anchor="Dukkipati06" target="https://dl.acm.org/doi/10.1145/1111322.1111336"> <front> <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="Rajiullah15" target="https://www.diva-portal.org/smash/get/diva2:846109/FULLTEXT01.pdf"> <front> <title>Towards a Low Latency Internet: Understanding and Solutions</title> <author fullname="Mohammad Rajiullah" initials="M." surname="Rajiullah"> <organization>KarlstadUni</organization>University Studies</organization> </author> <date year="2015"/> </front><seriesInfo name="Master's Thesis;<refcontent>Dissertation, KarlstadUni, Dept of Maths & CS " value="2015:41"/>University</refcontent> </reference> <reference anchor="lowat" target="https://blog.cloudflare.com/http-2-prioritization-with-nginx/"> <front> <title>Optimizing HTTP/2 prioritization with BBR and tcp_notsent_lowat</title> <author fullname="Patrick Meenan" initials="P." surname="Meenan"> <organization>Cloudflare</organization> </author> <dateday="12"month="October" year="2018"/> </front><seriesInfo name="Cloudflare Blog" value=""/><refcontent>Cloudflare Blog</refcontent> </reference> <reference anchor="NASA04" target="https://ntrs.nasa.gov/api/citations/20120009198/downloads/20120009198.pdf?attachment=true"> <front> <title>Latency Requirements for Head-Worn Display S/EVS Applications</title> <author fullname="Randall E. Bailey"initials="R.R."initials="R." surname="Bailey"><organization>NASA</organization><organization>NASA Langley Research Center</organization> </author> <author fullname="J.J. (Trey) Arthur III"initials="J.J."initials="J." surname="Trey Arthur III"><organization>NASA</organization> <address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address><organization>NASA Langley Research Center</organization> </author> <author fullname="Steven P. Williams"initials="S.P."initials="S." surname="Williams"><organization>NASA</organization> <address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address><organization>NASA Langley Research Center</organization> </author> <date month="April" year="2004"/> </front> <seriesInfoname="SPIE Defense and Security Symposium" value="LF99-1955"/>name="DOI" value="10.1117/12.554462"/> <refcontent>Proceedings of SPIE 5424</refcontent> </reference> <reference anchor="Raaen14" target="http://ojs.bibsys.no/index.php/NIK/article/view/9/6"> <front> <title>LatencythresholdsThresholds forusabilityUsability ingames:Games: Asurvey</title>Survey</title> <author fullname="Kjetil Raaen" initials="K." surname="Raaen"> <organization/> </author> <author fullname="Tor-Morten Grønli" initials="T-M." surname="Grønli"> <organization/><address> <postal> <street/> <city/> <region/> <code/> <country/> </postal> <phone/> <email/> <uri/> </address></author> <date year="2014"/> </front><seriesInfo name="Norsk<refcontent>Norsk IKT-konferanse for forskning ogutdanning" value=""/>utdanning (Norwegian ICT conference for research and education)</refcontent> </reference> </references><!-- <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, Wes Eddy, Karen Nielsen, David Black, Jake Holland, Vidhi Goel, Ermin Sakic, Praveen Balasubramanian, Gorry Fairhurst, Mirja Kuehlewind, Philip Eardley, Neal Cardwell, Pete Heist and Martin Duke<contact fullname="Richard Scheffenegger"/>, <contact fullname="Wes Eddy"/>, <contact fullname="Karen Nielsen"/>, <contact fullname="David Black"/>, <contact fullname="Jake Holland"/>, <contact fullname="Vidhi Goel"/>, <contact fullname="Ermin Sakic"/>, <contact fullname="Praveen Balasubramanian"/>, <contact fullname="Gorry Fairhurst"/>, <contact fullname="Mirja Kuehlewind"/>, <contact fullname="Philip Eardley"/>, <contact fullname="Neal Cardwell"/>, <contact fullname="Pete Heist"/>, and <contact fullname="Martin Duke"/> for their useful review comments. Thanks also to the area reviewers:Marco Tiloca, Lars Eggert, Roman Danyliw and Eric Vyncke.</t> <t>Bob Briscoe<contact fullname="Marco Tiloca"/>, <contact fullname="Lars Eggert"/>, <contact fullname="Roman Danyliw"/>, and <contact fullname="Éric Vyncke"/>.</t> <t><contact fullname="Bob Briscoe"/> andKoen<contact fullname="Koen DeSchepperSchepper"/> 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"/> was alsopart-fundedpartly funded by the Research Council of Norway through the TimeIn project, partly byCableLabsCableLabs, and partly by the Comcast Innovation Fund. The views expressed here are solely those of the authors.</t> </section> </back> </rfc>