<?xmlversion="1.0" encoding="US-ASCII"?>version='1.0' encoding='utf-8'?> <!DOCTYPE rfc SYSTEM"rfc2629.dtd" [ <!-- [rfced] updated by Chris /09/04/19 --> <!ENTITY RFC7230 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7230.xml"> <!ENTITY RFC7231 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7231.xml"> <!ENTITY RFC7232 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7232.xml"> <!ENTITY RFC7233 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7233.xml"> <!ENTITY RFC7234 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7234.xml"> <!ENTITY RFC7235 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7235.xml"> <!ENTITY RFC3550 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3550.xml"> <!ENTITY RFC4566 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4566.xml"> <!ENTITY RFC2119 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2119.xml"> <!ENTITY RFC3986 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3986.xml"> <!ENTITY RFC3264 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3264.xml"> <!ENTITY RFC4634 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4634.xml"> <!ENTITY RFC8017 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8017.xml"> <!ENTITY RFC0793 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.0793.xml"> <!ENTITY RFC0768 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.0768.xml"> <!ENTITY RFC3550 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3550.xml"> <!ENTITY RFC3629 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3629.xml"> <!ENTITY RFC5322 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5322.xml"> <!ENTITY RFC8174 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8174.xml"> <!ENTITY RFC3261 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3261.xml"> <!ENTITY RFC3649 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3649.xml"> <!ENTITY RFC4656 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4656.xml"> <!ENTITY RFC5357 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5357.xml"> ]>"rfc2629-xhtml.ent"> <rfc xmlns:xi="http://www.w3.org/2001/XInclude" submissionType="independent" category="info"number="XXXX" ipr="trust200902">number="8802" docName="draft-aranda-dispatch-q4s-10" ipr="trust200902" obsoletes="" updates="" xml:lang="en" sortRefs="false" symRefs="true" tocDepth="4" tocInclude="true" version="3"> <!-- xml2rfc v2v3 conversion 2.40.1 --> <!-- Generated by id2xml 1.4.4 on 2019-08-20T02:21:18Z --><?rfc compact="yes"?> <?rfc text-list-symbols="oo*+-"?> <?rfc subcompact="no"?> <?rfc sortrefs="no"?> <?rfc symrefs="yes"?> <?rfc strict="yes"?> <?rfc toc="yes"?><front> <title>The Quality for Service (Q4S) Protocol</title> <seriesInfo name="RFC" value="8802"/> <author fullname="Jose Javier Garcia Aranda"initials="J."initials="J.J." surname="Aranda"> <organization>Nokia</organization><address><postal><street>C/Maria<address> <postal> <street>María Tubau 9</street><street>28050 Madrid</street> <street>Spain</street><code>28050</code> <city>Madrid</city> <country>Spain</country> </postal> <phone>+34 91 330 4348</phone> <email>jose_javier.garcia_aranda@nokia.com</email> </address> </author> <authorfullname="Monica Cortes"fullname="Mónica Cortés" initials="M."surname="Cortes"> <organization abbrev="Univ. Politecnica de Madrid">Universidad Politecnica de Madrid</organization> <address><postal><street>Avenida Complutense 30</street> <street>28040 Madrid</street> <street>Spain</street>surname="Cortés"> <organization>Nokia</organization> <address> <postal> <street>María Tubau 9</street> <code>28050</code> <city>Madrid</city> <country>Spain</country> </postal><email>cortesm@dit.upm.es</email><email>monica.cortes_sack@nokia.com</email> </address> </author> <authorfullname="Joaquin Salvachua"fullname="Joaquín Salvachúa" initials="J."surname="Salvachua">surname="Salvachúa"> <organization abbrev="Univ. Politecnica de Madrid">Universidad Politecnica de Madrid</organization><address><postal><street>Avenida<address> <postal> <street>Avenida Complutense 30</street><street>28040 Madrid</street> <street>Spain</street><code>28040</code> <city>Madrid</city> <country>Spain</country> </postal> <phone>+34 91 0672134</phone><email>jsalvachua@dit.upm.es</email><email>Joaquin.salvachua@upm.es</email> </address> </author> <author fullname="Maribel Narganes" initials="M." surname="Narganes"> <organization abbrev="Tecnalia">Tecnalia Research & Innovation</organization><address><postal><street>Parque Cientifico<address> <postal> <extaddr>Parque Científico yTecnologicoTecnológico deBizkaia</street> <street>Geldo Auzoa,Bizkaia</extaddr> <street>Astondo Bidea, Edificio 700</street><street>E-48160 Derio (Bizkaia)</street> <street>Spain</street><code>E-48160</code> <city>Derio</city> <region>Bizkaia</region> <country>Spain</country> </postal> <phone>+34 946 430 850</phone> <email>maribel.narganes@tecnalia.com</email> </address> </author> <authorfullname="Inaki Martinez Sarriegui"fullname="Iñaki Martínez-Sarriegui" initials="I."surname="Sarriegui">surname="Martínez-Sarriegui"> <organization>Optiva Media</organization><address><postal><street>Edificio<address> <postal> <street>Edificio Europa II,</street> <street>Calle Musgo 2, 1G,</street> <street>28023 Madrid</street> <street>Spain</street> </postal> <phone>+34 91 297 7271</phone> <email>inaki.martinez@optivamedia.com</email> </address> </author> <datemonth="September" year="2019"/> <!-- [rfced] Please insert any keywords (beyond those that appear in the title) for use on https://www.rfc-editor.org/search. --> <keyword>example</keyword> <abstract><t>month="July" year="2020"/> <keyword>quality measurement</keyword> <keyword>measurement protocol</keyword> <keyword>latency</keyword> <keyword>jitter</keyword> <keyword>bandwidth</keyword> <keyword>packet-loss</keyword> <abstract> <t> This memo describes anapplication levelapplication-level protocol for the communication of end-to-end QoS compliance information based on theHypertextHyperText Transfer Protocol (HTTP) and the Session Description Protocol (SDP). The Quality for ServiceProtocol(Q4S) protocol provides a mechanism to negotiate and monitor latency, jitter, bandwidth, andpacket,packet loss, and to alert whenever one of the negotiated conditions is violated.</t> <t> Implementation details on the actions to be triggered upon reception/detection of QoS alerts exchanged by the protocol are out of scope of thisdocument,document; it is either application dependent (e.g., act to increase quality or reduce bit-rate) or network dependent (e.g., change connection's quality profile).</t> <t> This protocol specification is the product of research conducted over a number ofyears, andyears; it is presented here as a permanent record and to offer a foundation for future similar work. It does not represent a standard protocol and does not have IETF consensus.</t> </abstract> </front> <middle> <sectiontitle="Introduction" anchor="section-1"><t>anchor="sec-1" numbered="true" toc="default"> <name>Introduction</name> <t> The World Wide Web (WWW) is a distributed hypermedia systemwhichthat has gained widespread acceptance among Internet users. Although WWW browsers support other, preexisting Internet application protocols, the primary protocol used between WWW clients and servers became the HyperText Transfer Protocol (HTTP)(RFC 7230(<xref target="RFC7230" format="default"/>, <xreftarget="ref-1"/>, RFC 7231target="RFC7231" format="default"/>, <xreftarget="ref-2"/>, RFC 7232target="RFC7232" format="default"/>, <xreftarget="ref-3"/>, RFC 7233target="RFC7233" format="default"/>, <xreftarget="ref-4"/>, RFC 7234 <xref target="ref-5"/>,target="RFC7234" format="default"/>, andRFC 7235<xreftarget="ref-6"/>).target="RFC7235" format="default"/>). Since then, HTTP over TLS (known as HTTPS and described inRFC 2818 [7])<xref target="RFC2818" format="default"/>) has become an imperative for providing secure and authenticated WWW access. The mechanisms described in this document are equally applicable to HTTP and HTTPS.</t> <t> The ease of use of the Web has prompted its widespread employment as a client/server architecture for many applications. Many of such applications require the client and the server to be able to communicate with each other and exchange information with certain quality constraints.</t> <t> Quality in communications at the application level consists of four measurable parameters:<list style="hanging" hangIndent="6"> <t hangText="Latency:">The</t> <dl newline="false" spacing="normal" indent="6"> <dt>Latency:</dt> <dd>The time a message takes to travel from source to destination. It may be approximatedtoas RTT/2(Round trip(round-trip time), assuming the networks are symmetrical. In thiscontextcontext, we will consider the statistical medianformula.</t> <t hangText="Jitter:">latencyformula.</dd> <dt>Jitter:</dt> <dd>Latency variation. There are some formulas to calculateJitter,jitter, and in thiscontextcontext, we will consider the arithmetic meanformula.</t> <t hangText="Bandwidth:">bitformula.</dd> <dt>Bandwidth:</dt> <dd>Bit rate of communication. Toassureensure quality, a protocol mustassureensure the availability of the bandwidth needed by theapplication.</t> <t hangText="Packet loss:">Theapplication.</dd> <dt>Packet loss:</dt> <dd>The percentage of packet loss is closely related to bandwidth and jitter.AffectsPacket loss affects bandwidth because a high packet lossimpliessometimes implies retransmissions that also consumes extra bandwidth, other times the retransmissions are not achieved (forexampleexample, in video streaming overUDP)UDP), and the information received is less than the required bandwidth. In terms of jitter, a packet loss sometimes is seen by the destinationlikeas a larger time between arrivals, causing a jittergrowth.</t> </list> </t>growth.</dd> </dl> <t> Any other communicationparameterparameter, such as throughput, is not a network parameter because it depends on protocol window size and other implementation-dependent aspects.</t> <t> TheQuality for Service Protocol (Q4S)Q4S protocol provides a mechanism for quality monitoring based on an HTTP syntax and the Session DescriptionprotocolProtocol (SDP) in order to be easily integrated in the WWW, but it may be used by any type of application, not only those based on HTTP. Quality requirements may be needed by any type of application that communicates using any kind of protocol, especially those with real-time constraints. Depending on the nature of eachapplicationapplication, the constraints may bedifferentdifferent, leading to different parameter thresholds that need to be met.</t> <t> Q4S is anapplication level Client/Serverapplication-level client/server protocol that continuously measures session quality for a given flow (or set of flows), end-to-end (e2e) and inreal-time;real time; raising alerts if quality parameters are below a givenpre-negotiatednegotiated threshold and sending recoveries when quality parameters are restored. Q4S describes when these notifications,alertsalerts, andrecoveries,recoveries need to be sent and the entity receiving them. The actions undertaken by the receiver of the alert are out of scope of the protocol.</t> <t> Q4S is session-independent from the applicationflows,flows to minimize the impact on them. To perform the measurements, two control flows are created on both communication paths (forward and reverse directions).</t> <t> This protocol specification is the product of research conducted over a number ofyears,years and is presented here as a permanent record and to offer a foundation for future similar work. It does not represent a standard protocol and does not have IETF consensus.</t> <sectiontitle="Scope" anchor="section-1.1"><t>anchor="sec-1.1" numbered="true" toc="default"> <name>Scope</name> <t> The purpose of Q4S is to measure end-to-end network quality inreal-time.real time. Q4S does not transport any application data.ItThis means that Q4S is designed to be used jointly with other transport protocols such asReal TimeReal-time Transport Protocol(RTP)(RFC 3550(RTP) <xreftarget="ref-8"/>),target="RFC3550" format="default"/>, Transmission Control Protocol (TCP)(RFC 793<xreftarget="ref-16"/>), Quick UDP Internet Connections (QUIC)<xref target="ref-9"/> ,target="RFC0793" format="default"/>, QUIC <xref target="I-D.ietf-quic-transport" format="default"/>, HTTP <xreftarget="ref-1"/>,target="RFC7230" format="default"/>, etc.</t> <t> Some existent transport protocols are focusedinon real-time media transport and certain connection metrics are available, which is the case of RTP andReal TimeRTP Control Protocol(RTCP)<xref target="ref-8"/>.(RTCP) <xref target="RFC3550" format="default"/>. Other protocols such as QUIC provide low connection latencies as well as advanced congestion control. These protocols transport data efficiently and provide a lot of functionalities. However, there are currently no other quality measurement protocols offering the same level of function as Q4S. See <xreftarget="section-1.4"/>target="sec-1.4" format="default"/> for a discussion of the IETF'sOWAMP and TWAMPquality measurementprotocols.</t>protocols, One-Way Active Measurement Protocol (OWAMP) and Two-Way Active Measurement Protocol (TWAMP).</t> <t> Q4Senableenables applications to become reactive under e2e network quality changes. To achieve it, an independent Q4S stack application must run in paralleltowith the target application. Then, Q4S metrics may be used to trigger actions on the targetapplicationapplication, such as speed adaptation to latency in multiuser games, bitrate control at streaming services, intelligent commutation of delivery node at Content Delivery Networks, and whatever the target applicationallow.</t>allows.</t> </section> <sectiontitle="Motivation" anchor="section-1.2"><t>anchor="sec-1.2" numbered="true" toc="default"> <name>Motivation</name> <t> Monitoring quality of service (QoS) in computer networks is useful for several reasons:</t><t><list style="symbols"><t>Enable<ul spacing="normal"> <li>It enables real-time services and applications to verify whether network resources achieve a certain QoS level. This helps real-time services and applications to runthroughover the Internet, allowing the existence of Application Content Providers(ACPs)(ACPs), which offer guaranteed real-time services to thefinal users.</t> <t>Real-timeend users.</li> <li>Real-time monitoring allows applications to adapt themselves to network conditions(Application-based(application-based QoS) and/or request more network qualitytofrom the Internet Service Provider (ISP) (if the ISP offers thispossibility).</t> <t>Monitoringpossibility).</li> <li>Monitoring may also be required byPeer to Peerpeer-to-peer (P2P) real-time applications for which Q4S can beused</t> <t>Enableused.</li> <li>Monitoring enables ISPs to offer QoS to any ACP orfinalend user application in an accountableway</t> <t>Enableway.</li> <li> <t>Monitoring enables e2e negotiation of QoS parameters, independently of the ISPs of bothendpoints.<vspace blankLines="1"/>endpoints.</t> </li> </ul> <t> A protocol to monitor QoS must address the following issues: </t><t>Must<ul spacing="normal"> <li>Must be ready to be used in conjunction with current standard protocols and applications, without forcing a change onthem.</t> <t>Mustthem.</li> <li>Must have a formal and compact way to specify quality constraints desired by the application torun.</t> <t>Mustrun.</li> <li>Must have measurement mechanismsavoidingthat avoid application disruption andminimizingminimize network resourcesconsumption.</t> <t>Mustconsumption.</li> <li>Must have specific messages to alert about the violation of quality constraints in different directions (forward andreverse),reverse) because network routing may not be symmetrical, and of course, quality constraints may not besymmetrical.</t> <t>Aftersymmetrical.</li> <li>After having alerted about the violation of quality constraints, must have specific messages to inform about the recovery of quality constraints in corresponding directions (forward andreverse).</t> <t>Mustreverse).</li> <li>Must protect the data(constrains,(constraints, measurements, QoS levels demanded from the network) in order to avoid the injection of malicious data in themeasurements.</t> </list> </t>measurements.</li> </ul> </section> <sectiontitle="Summaryanchor="sec-1.3" numbered="true" toc="default"> <name>Summary ofFeatures" anchor="section-1.3"><t>Features</name> <t> The Quality for ServiceProtocol(Q4S) protocol is a message-oriented communication protocol that can be used in conjunction with any other application-level protocol. Q4S is a measurement protocol. Any action taken derived from its measurements are out of scope of the protocol. These actions depend on the application provider and may be application-level adaptive reactions, may involve requests to the ISP, or whatever the application providerdecide.</t>decides.</t> <t> The benefits in quality measurements provided by Q4S can be used by any type of application that uses any type of protocol for data transport. It provides a quality monitoring scheme for any communication that takes place between the client and the server, not only for the Q4S communication itself.</t> <t> Q4S does not establish multimediasessionssessions, and it does not transport application data. It monitors the fulfillment of the quality requirements of the communication between the client and theserver, and thereforeserver; therefore, it does not impose any restrictions on the type of application,protocolprotocol, orthe type ofusage of the monitored quality connection.</t> <t> Some applications may vary their quality requirements dynamically for any given quality parameter. Q4S is able to adapt to the changing applicationneedsneeds, modifying the parameter thresholds to the new values and monitoring the network quality according to the new quality constraints. It will raise alerts if the new constraints are violated.</t> <t> The Q4S session lifetime is composed of four phases with different purposes: Handshake, Negotiation,ContinuityContinuity, and Termination. Negotiation and Continuity phases perform network parameter measurementsasper a negotiated measurement procedure. Different measurement procedures could be used inside Q4S, although one default measurement mechanism is needed for compatibility reasons and is the one defined in this document. Basically, Q4S defines how to transport application quality requirements and measurement results between a client and server andprovidinghow to provide monitoring andalertingalerting, too.</t> <t> Q4S must be executed just before starting a client-server applicationwhichthat needs a quality connection in terms of latency, jitter,bandwidthbandwidth, and/or packet loss. Once the client and server have succeeded in establishing communication under quality constraints, the application can start, and Q4S continues measuring and alerting if necessary.</t> <t> The quality parameters can be suggested by the client in the first message of thehandshakeHandshake phase, butit'sit is the server that accepts these parameter values or forces others. The server is in charge of deciding the final values of quality connection.</t> </section> <sectiontitle="Differences with OWAMP/TWAMP" anchor="section-1.4"><t>anchor="sec-1.4" numbered="true" toc="default"> <name>Differences from OWAMP/TWAMP</name> <t> OWAMP(RFC 4656)<xreftarget="ref-27"/>target="RFC4656" format="default"/> and TWAMP(RFC 5357)<xreftarget="ref-28"/>target="RFC5357" format="default"/> are two protocols to measure network quality in terms of RTT, buthasthey have a different goal than Q4S. The main difference is the scope: Q4S is designed to assist reactive applications,whilewhereas OWAMP/TWAMP is designedjustto measure just network delay.</t> <t>DifferencesThe differences can be summarized in the following points:</t><t><list style="symbols"><t>OWAMP/TWAMP is<ul spacing="normal"> <li>OWAMP and TWAMP are not intended for measuring availability of resources (certainBandwidth availabilitybandwidth availability, for example) but only RTT. However, Q4S is intended for measuring required bandwidth,packet-loss, jitterpacket loss, jitter, and latency in both directions. Available bandwidth is not measured by Q4S, butrequiredbandwidth required for a specificapplication.</t> <t>OWAMP/TWAMP doesapplication is.</li> <li>OWAMP and TWAMP do not have responsivity control (which defines the speed of protocol reactions under network qualitychanges),changes) becausethis protocol isthese protocols are designed to measure network performance, not to assist reactiveapplicationsapplications, anddoesdo not detect the fluctuations of qualityinwithin certain time intervals to take reactive actions. However, responsivity control is a key feature ofQ4S.</t> <t>OWAMP/TWAMP isQ4S.</li> <li>OWAMP and TWAMP are not intended to run in parallel with reactive applications, butQ4S'the Q4S protocol's goal is to run in parallel and assist reactive applicationsto takein making decisions based onQ4S ALERT packetsQ4S-ALERT packets, which may triggeractions.</t> </list> </t>actions.</li> </ul> </section> </section> <sectiontitle="Terminology" anchor="section-2"><t>anchor="sec-2" numbered="true" toc="default"> <name>Terminology</name> <t> The key words"MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY","<bcp14>MUST</bcp14>", "<bcp14>MUST NOT</bcp14>", "<bcp14>REQUIRED</bcp14>", "<bcp14>SHALL</bcp14>", "<bcp14>SHALL NOT</bcp14>", "<bcp14>SHOULD</bcp14>", "<bcp14>SHOULD NOT</bcp14>", "<bcp14>RECOMMENDED</bcp14>", "<bcp14>NOT RECOMMENDED</bcp14>", "<bcp14>MAY</bcp14>", and"OPTIONAL""<bcp14>OPTIONAL</bcp14>" in this document are to be interpreted as described inBCP 14 RFC 2119BCP 14 <xreftarget="ref-11"/> RFC 8174target="RFC2119"/> <xreftarget="ref-21"/>target="RFC8174"/> when, and only when, they appear in all capitals, as shownhere.</t>here. </t> </section> <sectiontitle="Overviewanchor="sec-3" numbered="true" toc="default"> <name>Overview ofOperation" anchor="section-3"><t>Operation</name> <t> This section introduces the basic operation of Q4S using simple examples. This section is of a tutorial nature and does not contain any normative statements.</t> <t> The first example shows the basic functions ofaQ4S: communication establishment between a client and a server, quality requirement negotiations for the requested application, application start and continuous quality parameter measurements, and finally communication termination.</t> <t> The client triggers the establishment of the communication by requesting a specific service or application from the server. This first message must have a special URI(RFC 3986)<xref target="ref-12"/>,<xref target="RFC3986" format="default"/>, which may force the use of the Q4S protocol if it is implemented in a standard web browser. This message consists of a Q4S BEGIN method, which can optionally include a proposal for the communication quality requirements in an SDP body. This option gives the client a certain negotiation capacity about quality requirements, but it will be the server who finally decidesaboutthe stated requirements.</t> <t> This request is answered by the server with a Q4S 200 OK response letting the client know that it accepts the request. This response message must contain an SDP bodywith:</t> <t><list style="symbols"><t>Thewith the following:</t> <ul spacing="normal"> <li>The assigned Q4Ssession id.</t> <t>Thesess-id.</li> <li>The quality constraints required by the requestedapplication.</t> <t>Theapplication.</li> <li>The measurement procedure touse.</t> <t>The alerting mode: thereuse.</li> <li> <t>"alerting-mode" attribute: There are two different scenarios for sending alerts that trigger actions either on the network or in the application when measurements identify violated quality constraints. In both cases, alerts are triggered by the server.<list style="format (%c)"> <t> Q4S-aware-network</t> <ol spacing="normal" type="(%c)"> <li> <t>Q4S-aware-network scenario:theThe network is Q4Saware,aware and reacts by itself to these alerts. In thisscenario Q4S ALERTscenario, Q4S-ALERT messages are sent by the server to the client, and network elements inspect and process these alert messages. The alerting mode in this scenario is called Q4S-aware-network alerting mode.</t> </li> <li> <t>Reactive scenario: As shown inFigure 1,<xref target="ref-reactive-scenario" format="default"/>, the network is not Q4S aware. In thisscenarioscenario, alert notifications are sent to a specific node, called an Actuator, which is in charge oftakingmaking decisions regarding what actions to trigger: either to change application behavior to adapt it to network conditions and/or invoke a network policy server in order to reconfigure the network and requestmorebetter quality for application flows.</t></list> </t> </list> </t><figuretitle="Reactive scenario" anchor="ref-reactive-scenario"><artwork><![CDATA[anchor="ref-reactive-scenario"> <name>Reactive Scenario</name> <artwork name="" type="" align="left" alt=""><![CDATA[ +------+ +-----------+ | App |<----- app flows---------->|Application| |Client| +-----------+ +------+ A | +------+ +------+ +--------+ | Q4S |<----Q4S---->| Q4S |<----->|Actuator| |Client| |Server| +--------+ +------+ +------+ | V +-------------+ |policy server| +-------------+ ]]></artwork> </figure><t><list style="symbols"><t>The<t>The format of messages exchanged between the server stack and theActuator,Actuator doesn't follow Q4S codificationrules, butrules; their format will be implementation dependent. In this way, we will call the messages sent from the server stack to the Actuator "notifications" (e.g., alertnotifications),notifications) and the messages sent from the Actuator to the server stack in response to notifications "acknowledges" (e.g., alert acknowledges).</t><t>alert-pause:</li> </ol></li> </ul> <ul spacing="normal"> <li>"alert-pause" attribute: The amount of time between consecutive alerts. In the Q4S-aware-network scenario, the server has to wait this period of time betweenQ4S ALERTQ4S-ALERT messages sent to the client. In the Reactive scenario, the server stack has to wait this period of time between alert notifications sent to the Actuator. Measurements are not stopped in Negotiation or ContinuityPhasesphases during this period of time, but no alerts aresentsent, even with violated network qualityconstraintsconstraints, in order to leave time for network reconfiguration or for applicationadjustments.</t> <t>recovery-pause:adjustments.</li> <li>"recovery-pause" attribute: The amount of time the Q4S server waits before trying to recover the initialqos-level."qos-level" (<xref target="sec-7.2.1" format="default"/>). After having detected violation of quality constraints several times, theqos-level"qos-level" will have been increased accordingly. If this violation detection finally stops, the server waits for a period of time (recoverytime)time), and if the situation persists, it tries to recover to previousqos-level"qos-level" values gradually by sendingQ4S RECOVERYQ4S-RECOVERY messages to theclient,client in the Q4S-aware-network scenario, or recovery notifications to theActuator,Actuator in the Reactivescenario.</t> </list> </t>scenario (<xref target="sec-7.9" format="default"/>).</li> </ul> <t> It is important to highlight that any Q4S 200 OK response sent by the server to the client at any time during the life of a quality session may contain an SDP body with new values of quality constraints required by the application. Depending on the phase and the state of the measurement procedure within the specific phase, the client will react accordinglyso asto take into account the new quality constraints in the measurement procedure.</t> <t> Once the communication has been established(handshake(i.e., the Handshake phase is finished), the protocol will verify that the communication path between the client and the server meets the quality constraintsonin both directions, from and to the server(negotiation(Negotiation phase). ThisnegotiationNegotiation phase requires taking measurements of the quality parameters: latencies, jitter,bandwidthbandwidth, and packet loss. This phase is initiated with a client message containing a Q4S READY method, which will be answered by the server with a Q4S 200 OK response.</t> <t> Negotiation measurements are achieved in two sequential stages:<list style="hanging" hangIndent="6"> <t hangText="Stage 0:"></t> <dl newline="false" spacing="normal" indent="6"> <dt>Stage 0:</dt> <dd> latency and jittermeasurements</t> <t hangText="Stage 1:">measurements</dd> <dt>Stage 1:</dt> <dd> bandwidth and packet lossmeasurements</t> </list> </t>measurements</dd> </dl> <t> Stage 0 measurements arebeingtaken through Q4S PING messages sentbothfrom both the client and the server. All Q4S PING requests will be answered by Q4S 200 OK messages to allow for bidirectional measurements.</t> <t> Different client and server implementations may send a different number of PING messages for measuring, although at least 255 messages should be considered to perform the latency measurement. The Stage 0 measurements only may be considered ended when neither client nor server receive new PING messages after an implementation-dependent guard time. Onlyafter, clientafter Stage 0 has ended, can the client send a "READY 1" message.</t> <t> After a pre-agreed number of measurements have been performed, determined by the measurement procedure sent by the server, three scenarios may be possible:<list style="format (%c)"> <t></t> <ol spacing="normal" type="(%c)"> <li> Measurements do not meet the requirements: in thiscasecase, the stage 0 is repeated after sending an alert from the server to the client or from the server stack to the Actuator, depending on the alerting mode defined in the Handshake phase. Notice that measurements continue to be taken but no alerts are sent during thealert-pause"alert-pause" time. In the Reactive scenario, the Actuator will decide either to forward the alert notification to the network policy server or to the application, depending on where reconfiguration actions have to be taken.</t> <t></li> <li> Measurements do meet the requirements: in thiscasecase, client moves to stage 1 by sending a new READY message.</t> <t></li> <li> At any time during the measurement procedure, the Q4S 200 OK message sent by the server to the client, in response to a Q4S PING message, contains an SDP body with new values of quality constraints required by theapplication; thisapplication. This means the application has varied their quality requirementsdynamically and thereforedynamically; therefore, quality thresholds used while monitoring quality parameters have to be changed to the new constraints. In thiscasecase, the client moves to the beginning oftheStage 0 for initiating the negotiation measurements again.</t> </list> </t></li> </ol> <t> Stage 1 is optional. Its purpose is to measure the availability ofapplication neededapplication-needed bandwidth.This stageIf the "bandwidth" attribute is set to zero kbps in the SDP, the client canbe skippedskip stage 1 byclientsending a "READY 2" message after completion of stage0 when bandwidth requirements is set to cero kbps in the SDP.0. Stage 1 measurements are achieved through Q4S BWIDTH messages sentbothfrom both the client and the server. Unlike PING messages, Q4S BWIDTH requests will not be answered.</t> <t> If Stage 0 and 1 meet the application quality constraints, the application may start. Q4S will enter thecontinuityContinuity phase by measuring the network quality parameters through the Q4S PING message exchange on both connectionpaths,paths and raising alerts in case of violation.</t> <t> Once the client wants to terminate the qualitysessionsession, it sends a Q4S CANCEL message, which will be acknowledged by the server with another Q4S CANCEL message. Termination of quality sessions are always initiated by the client because Q4S TCP requests follow theclient serverclient/server schema.</t><t><list style="hanging" hangIndent="-1"><t hangText="Figure 2<t><xref target="ref-successful-q4s-message-exchange" format="default"/> depicts the message exchange in a successfulscenario."> <vspace blankLines="0"/> </t> </list> </t>scenario.</t> <figuretitle="Successfulanchor="ref-successful-q4s-message-exchange"> <name>Successful Q4Smessage exchange" anchor="ref-successful-q4s-message-exchange"><artwork><![CDATA[Message Exchange</name> <artwork name="" type="" align="left" alt=""><![CDATA[ +-------------------------------------------+ | | | Client Server | | | Handshake | --------- Q4S BEGIN -----------> | | <-------- Q4S 200 OK ----------- | | | Negotiation | | (Stage 0) | --------- Q4S READY 0----------> | | <-------- Q4S 200 OK ----------- | | | | --------- Q4S PING ------------> | | <-------- Q4S 200 OK ----------- | | <-------- Q4S PING ------------- | | -------- Q4S 200 OK ----------> | | --------- Q4S PING ------------> | | <-------- Q4S PING ------------- | | --------- Q4S 200 OK ----------> | | <-------- Q4S 200 OK ----------- | | ... | Negotiation | | (Stage 1) | --------- Q4S READY 1----------> | | <-------- Q4S 200 OK ----------- | | | | --------- Q4S BWITDH ----------> | | <-------- Q4S BWIDTH------------ | | --------- Q4S BWITDH ----------> | | <-------- Q4S BWIDTH------------ | | ... | Continuity | --------- Q4S READY 2 ---------> | | <-------- Q4S 200 OK ----------- | app start | | | --------- Q4S PING ------------> | | <-------- Q4S 200 OK ----------- | | <-------- Q4S PING ------------- | | -------- Q4S 200 OK ----------> | | | Termination | --------- Q4S CANCEL ----------> | app end | <-------- Q4S CANCEL ----------- | | | +-------------------------------------------+ ]]></artwork> </figure> <t>ClientBoth client and server measurements are includedintoin the PING and BWIDTH messages, allowing both sides of the communication channel to beareaware of all measurements in both directions.</t> <t> The following two examples show the behavior of the Q4S protocolwhen:when quality constraints are violated, and alerts are generated; and, later on, when the violation of quality constraints stops leading to the execution of the recovery process. The first example(Figure 3)(<xref target="ref-q4s-aware-network-alerting-mode" format="default"/>) shows the Q4S-aware-network alerting mode scenario:</t> <figuretitle="Q4S-aware-network alerting mode" anchor="ref-q4s-aware-network-alerting-mode"><artwork><![CDATA[anchor="ref-q4s-aware-network-alerting-mode"> <name>Q4S-Aware-Network Alerting Mode</name> <artwork name="" type="" align="left" alt=""><![CDATA[ +-------------------------------------------+ | | | Client Server | | | Handshake | --------- Q4S BEGIN -----------> | | <-------- Q4S 200 OK ----------- | | | Negotiation | | (Stage 0) | --------- Q4S READY 0----------> | | <-------- Q4S 200 OK ----------- | | | | --------- Q4S PING ------------> | | <-------- Q4S 200 OK ----------- | | <-------- Q4S PING ------------- | | -------- Q4S 200 OK ----------> | | ... | | | | <--------Q4S ALERTQ4S-ALERT ------------ | | --------Q4S ALERTQ4S-ALERT ------------> | | (alert-pause start) | Repetition | | of Stage 0 | --------- Q4S READY 0----------> | | <-------- Q4S 200 OK ----------- | | | | --------- Q4S PING ------------> | | <-------- Q4S 200 OK ----------- | | <-------- Q4S PING ------------- | | ... | Negotiation | | (Stage 1) | --------- Q4S READY 1----------> | | <-------- Q4S 200 OK ----------- | | | | --------- Q4S BWITDH ----------> | | <-------- Q4S BWIDTH------------ | | ... | | | Continuity | --------- Q4S READY 2 ---------> | | <-------- Q4S 200 OK ----------- | app start | | | --------- Q4S PING ------------> | | <-------- Q4S 200 OK ----------- | | <-------- Q4S PING ------------- | | -------- Q4S 200 OK ----------> | | ... | |(alert-pause expires & | | violated constraints) | | <--------Q4S ALERTQ4S-ALERT ------------ | | ---------Q4S ALERTQ4S-ALERT -----------> | | | | (alert-pause start) | | --------- Q4S PING ------------> | | <-------- Q4S 200 OK ----------- | | <-------- Q4S PING ------------- | | --------- Q4S 200 OK ----------> | | ... | |(alert-pause expires & | | violated constraints) | | <--------Q4S ALERTQ4S-ALERT ------------ | | ---------Q4S ALERTQ4S-ALERT -----------> | | (alert-pause) | | --------- Q4S PING ------------> | | <-------- Q4S 200 OK ----------- | | <-------- Q4S PING ------------- | | -------- Q4S 200 OK ----------> | | ... | |(alert-pause expires & | |FullfilledFulfilled constraints) | | | | (recovery-pause start) | | | | --------- Q4S PING ------------> | | <-------- Q4S 200 OK ----------- | | <-------- Q4S PING ------------- | | -------- Q4S 200 OK ----------> | | ... | |(recovery-pause expires & | |FullfilledFulfilled constraints) | | <---------Q4S RECOVERYQ4S-RECOVERY --------- | | --------Q4S RECOVERYQ4S-RECOVERY -----------> | | | | (recovery-pause start) | | --------- Q4S PING ------------> | | <-------- Q4S 200 OK ----------- | | <-------- Q4S PING ------------- | | -------- Q4S 200 OK ----------> | | ... | | | Termination | --------- Q4S CANCEL ----------> | app end | <-------- Q4S CANCEL ----------- | | | +-------------------------------------------+ ]]></artwork> </figure> <t> In this Q4S-aware-network alerting mode scenario, the server may send Q4S alerts to the client at any timeonupon detection of violated quality constraints. This alerting exchange must not interrupt the continuity quality parameter measurements between client and server.</t> <t> The second example depicted inthe figure 4<xref target="ref-reactive-alerting-mode" format="default"/> represents the Reactive scenario, in which alert notifications are sent from the server stack to theActuatorActuator, which is in charge of decidingeitherto act over application behavior and/or to invoke a network policy server. The Actuator is an entity that has apre-defineddefined set of different quality levels and decides how to act depending on the actions stated for each of these levels; it can take actions for making adjustments on theapplicationapplication, or it can send a request to the policy server for acting on the network. The policy server also has apre-defineddefined set of different quality levelspre-agreedpreviously agreed upon between the Application Content Provider and the ISP. The Reactive alerting mode is the default mode.</t> <figuretitle="Reactive alerting mode" anchor="ref-reactive-alerting-mode"><artwork><![CDATA[anchor="ref-reactive-alerting-mode"> <name>Reactive Alerting Mode</name> <artwork name="" type="" align="left" alt=""><![CDATA[ +-------------------------------------------+ | | | Client Server Actuator | Handshake | ----- Q4S BEGIN -----> | | <---- Q4S 200 OK ----- | | | Negotiation | | (Stage 0) | ----- Q4S READY 0----> | | <---- Q4S 200 OK ----- | | | | ----- Q4S PING ------> | | <---- Q4S 200 OK ----- | | <---- Q4S PING ------- | | ---- Q4S 200 OK ----> | | ... | | (alert-pause start) | | --alert | | notification--> | | | | <--alert | | acknowledge--- | | | Repetition | | of Stage 0 | ----- Q4S READY 0----> | | <---- Q4S 200 OK ----- | | | | ----- Q4S PING ------> | | <---- Q4S 200 OK ----- | | <---- Q4S PING ------- | | ... | |(alert-pause expires & | | violated constraints) | | | | --alert | | notification--> | | | | <--alert | | acknowledge--- | | | | ----- Q4S PING ------> | | <---- Q4S 200 OK ----- | | <---- Q4S PING ------- | | ... | Negotiation | | (Stage 1) | ----- Q4S READY 1----> | | <---- Q4S 200 OK ----- | | | | ----- Q4S BWITDH ----> | | <---- Q4S BWIDTH------ | | ... | Continuity | ----- Q4S READY 2 ---> | | <---- Q4S 200 OK ----- | app start | | |(alert-pause expires & | | fulfilled constraints) | | | |(recovery-pause start) | | ----- Q4S PING ------> | | <---- Q4S 200 OK ----- | | <---- Q4S PING ------- | | ----- Q4S PING ------> | | | |(recovery-pause expires & | | fulfilled constraints) | | | | --recovery | | notification--> | | | | <--recovery | | acknowledge--- | | | |(recovery-pause start) | | <---- Q4S 200 OK ----- | | <---- Q4S PING ------- | | ----- Q4S 200 OK ----> | | ----- Q4S PING ------> | | ... | | | Termination | ----- Q4S CANCEL ----> | app end | --cancel | | notification--> | | | | <--cancel | | acknowledge-- | | <---- Q4S CANCEL ----- | | | +-------------------------------------------+ ]]></artwork> </figure> <t> At the end of any stage of the Negotiationphase stage,phase, the server sends an alert notification to the Actuator if quality constraints are violated. During the period of time defined by thealert-pause parameter,"alert-pause" attribute, no further alert notifications are sent, but measurements are not interrupted. This way, both the client and the server will detect network improvements as soon as possible. In a similarway,way during thecontinuityContinuity phase, the server may send alert notifications at any time to the Actuatoronupon detection of violated quality constraints. This alerting exchange must not interrupt the continuity measurements between client and server.</t> <t> Finally, in the Termination phase, Q4S CANCEL messages sent from the client to the server must be forwarded from the server to the Actuator in order to release possible assigned resources for the session.</t> </section> <sectiontitle="Q4S Messages" anchor="section-4"><t>anchor="sec-4" numbered="true" toc="default"> <name>Q4S Messages</name> <t> Q4S is a text-based protocol and uses the UTF-8 charset(RFC 3629<xreftarget="ref-19"/>).target="RFC3629" format="default"/>. A Q4S message is either a request or a response.</t> <t> BothRequestrequest andResponseresponse messages use the basic format of Internet Message Format(RFC 5322<xreftarget="ref-20"/>).target="RFC5322" format="default"/>. Both types of messages consist of a start-line, one or more header fields, an empty line indicating the end of the header fields, and an optionalmessage-body.</t>message-body. This document uses ABNF notation <xref target="RFC5234" format="default"/> for the definitions of the syntax of messages.</t> <t> The start-line, each message-header line, and the empty lineMUST<bcp14>MUST</bcp14> be terminated by a carriage-return line-feed sequence (CRLF). Note that the empty lineMUST<bcp14>MUST</bcp14> be present even if the message-body is not.</t><figure><artwork><![CDATA[<sourcecode type="abnf"><![CDATA[ generic-message = start-line CRLF *message-header CRLF CRLF [ message-body ] start-line = Request-Line / Status-Line]]></artwork> </figure>]]></sourcecode> <t> Much of Q4S's messages and header field syntax are identical to HTTP/1.1. However, Q4S is not an extension of HTTP.</t> <sectiontitle="Requests" anchor="section-4.1"><t>anchor="sec-4.1" numbered="true" toc="default"> <name>Requests</name> <t> Q4S requests are distinguished by having a Request-Line for a start-line. A Request-Line contains a method name, a Request-URI, and the protocol version separated by a single space (SP) character.</t> <t> The Request-Line ends with CRLF. No CR or LF are allowed except in the end-of-line CRLF sequence. No linear whitespace(LWS)(LWSP) is allowed in any of the elements.</t><t><sourcecode type="abnf"><![CDATA[ Request-Line = Method SP Request-URI SP Q4S-Version CRLF</t> <t><list style="hanging" hangIndent="6"> <t hangText="Method:">]]></sourcecode> <dl newline="false" spacing="normal" indent="6"> <dt>Method:</dt> <dd> This specification defines seven methods: BEGIN for starting andnegotiatenegotiating quality sessions, READY for synchronization of measurements, PING and BWIDTH for quality measurementspurpose,purposes, CANCEL for terminating sessions, Q4S-ALERT for reporting qualityviolations reporting,violations, and Q4S-RECOVERY for reporting qualityrecovery reporting. </t> <t hangText="Request-URI:">recovery. </dd> <dt>Request-URI:</dt> <dd> The Request-URI is a Q4S URI(RFC 2396)<xref target="RFC3986" format="default"/> as described in7.4.<xref target="sec-7.4" format="default"/>. The Request-URIMUST NOT<bcp14>MUST NOT</bcp14> contain unescaped spaces or control characters andMUST NOT<bcp14>MUST NOT</bcp14> be enclosed in "<>".</t> <t hangText="Q4S-Version:"></dd> <dt>Q4S-Version:</dt> <dd> Both request and response messages include the version of Q4S in use. To be compliant with this specification, applications sending Q4S messagesMUST<bcp14>MUST</bcp14> include a Q4S-Version of "Q4S/1.0". The Q4S-Version string iscase-insensitive,case insensitive, but implementationsMUST<bcp14>MUST</bcp14> sendupper-case.uppercase. Unlike HTTP/1.1, Q4S treats the version number as a literal string. In practice, this should make no difference.</t> </list> </t></dd> </dl> </section> <sectiontitle="Responses" anchor="section-4.2"><t>anchor="sec-4.2" numbered="true" toc="default"> <name>Responses</name> <t> Q4S responses are distinguished from requests by having a Status-Line as their start-line. A Status-Line consists of the protocol version followed by a numeric Status-Code and its associated textual phrase, with each element separated by a single SP character. No CR or LF is allowed except in the final CRLF sequence.</t><t><sourcecode type="abnf"><![CDATA[ Status-Line = Q4S-Version SP Status-Code SP Reason-Phrase CRLF</t>]]></sourcecode> <t> The Status-Code is a 3-digit integer result code that indicates the outcome of an attempt to understand and satisfy a request. The Reason-Phrase is intended to give a short textual description of the Status-Code. The Status-Code is intended for use by automata, whereas the Reason-Phrase is intended for the human user. A client is not required to examine or display the Reason-Phrase.</t> <t> While this specification suggests specific wording for thereason phrase,Reason-Phrase, implementationsMAY<bcp14>MAY</bcp14> choose other text, for example, in the language indicated in the Accept-Language header field of the request.</t> <t> The first digit of the Status-Code defines the class of response. The last two digits do not have any categorization role. For this reason, any response with a status code between 100 and 199 is referred to as a "1xx response", any response with a status code between 200 and 299 as a "2xx response", and so on. Q4S/1.0 allows following values for the first digit:<list style="hanging" hangIndent="6"> <t hangText="1xx:"></t> <dl newline="false" spacing="normal" indent="6"> <dt>1xx:</dt> <dd> Provisional -- request received, continuing to process the request;</t> <t hangText="2xx:"></dd> <dt>2xx:</dt> <dd> Success -- the action was successfully received, understood, and accepted;</t> <t hangText="3xx:"></dd> <dt>3xx:</dt> <dd> Redirection -- further action needs to be taken in order to complete the request;</t> <t hangText="4xx:"></dd> <dt>4xx:</dt> <dd> Request Failure -- the request contains bad syntax or cannot be fulfilled at this server;</t> <t hangText="5xx:"></dd> <dt>5xx:</dt> <dd> Server Error -- the server failed to fulfill an apparently validrequest" </t> <t hangText="6xx:">request; </dd> <dt>6xx:</dt> <dd> Global Failure -- the request cannot be fulfilled at anyserver" </t> </list></t>server. </dd> </dl> <t> The status codes are the same as described in HTTP(RFC 7231<xreftarget="ref-2"/>).target="RFC7231" format="default"/>. In the same way as HTTP, Q4S applications are not required to understand the meaning of all registered status codes, though such understanding is obviously desirable. However, applicationsMUST<bcp14>MUST</bcp14> understand the class of any status code, as indicated by the first digit, and treat any unrecognized response as being equivalent to the x00 status code of that class.</t> <t> The Q4S-ALERT,Q4S-RECOVERYQ4S-RECOVERY, and CANCEL requests do not have to beresponded.responded to. However, after receiving a Q4S-ALERT,Q4S-RECOVERYQ4S-RECOVERY, or CANCEL request, the serverSHOULD<bcp14>SHOULD</bcp14> send a Q4S-ALERT,Q4S-RECOVERYQ4S-RECOVERY, or CANCEL request to theclient</t>client.</t> </section> <sectiontitle="Header Fields" anchor="section-4.3"><t>anchor="sec-4.3" numbered="true" toc="default"> <name>Header Fields</name> <t> Q4S header fields are identical to HTTP header fields in both syntax and semantics.</t> <t> Some header fields only make sense in requests or responses. These are called request header fields and response header fields, respectively. If a header field appears in a message that does notmatchingmatch its category (such as a request header field in a response), itMUST<bcp14>MUST</bcp14> be ignored.</t> <sectiontitle="Commonanchor="sec-4.3.1" numbered="true" toc="default"> <name>Common Q4S HeaderFields" anchor="section-4.3.1">Fields</name> <t> These fields may appear inRequestrequest andResponseresponse messages.<list style="hanging" hangIndent="6"> <t hangText="Session-Id:">the</t> <dl newline="false" spacing="normal" indent="6"> <dt>Session-Id:</dt> <dd>the value for this header field is the samesession idsess-id used in SDP (embedded in"o"the SDPparameter)"o=" line) and is assigned by the server. The messages without SDPMUST<bcp14>MUST</bcp14> include thisheader.header field. If a message hasandan SDP body, this header field is optional. The method of<session id>sess-id allocation is up to the creating tool, but it is suggested that a UTC timestamp be used to ensureuniqueness.</t> <t hangText="Sequence-Number:">sequentialuniqueness.</dd> <dt>Sequence-Number:</dt> <dd>sequential and cyclic positive integer number assigned to PING and BWIDTHmessages,messages and acknowledged in 200 OKresponses.</t> <t hangText="Timestamp:">thisresponses.</dd> <dt>Timestamp:</dt> <dd>this optional header field contains the system time (with the best possible accuracy). It indicates the time in which the PING request was sent. If this header field is present in PING messages, then the 200 OK response messagesMUST<bcp14>MUST</bcp14> include thisvalue.</t> <t hangText="Stage:">thisvalue.</dd> <dt>Stage:</dt> <dd>this is used in the client's READY requests and the server's 200 OK responses during the Negotiation and Continuity phases in order to synchronize the initiation of the measurements. Example: Stage:0</t> </list> </t>0</dd> </dl> </section> <sectiontitle="Specificanchor="sec-4.3.2" numbered="true" toc="default"> <name>Specific Q4S Request HeaderFields" anchor="section-4.3.2"><t>Fields</name> <t> In addition to HTTP header fields, these are the specific Q4S request headerfields</t> <t><list style="symbols"><t>User-Agent: thisfields:</t> <dl newline="false" spacing="normal" indent="6"> <dt>User-Agent:</dt> <dd>this header field contains information about the implementation of the user agent. This is for statistical purposes, the tracing of protocol violations, and the automated recognition of user agents for the sake of tailoring responses to avoid particular user agent limitations. User agentsSHOULD<bcp14>SHOULD</bcp14> include this field with requests. The fieldMAY<bcp14>MAY</bcp14> contain multiple product tokens and comments identifying the agent and any sub-productswhichthat form a significant part of the user agent. By convention, the product tokens are listed in order of their significance for identifying theapplication.</t> <t>Signature: thisapplication.</dd> <dt>Signature:</dt> <dd><t>this header field contains a digital signature that can be used by the network,actuatorActuator, or policy server to validate the SDP, preventing security attacks. ThesignatureSignature is an optional header field generated by the server according to the pre-agreed security policies between the Application Content Provider and the ISP. For example, a hash algorithm and encryption method such as SHA256(RFC 4634<xreftarget="ref-14"/>)target="RFC6234" format="default"/> and RSA(RFC 8017<xreftarget="ref-15"/>)target="RFC8017" format="default"/> based on the server certificate could be used. This certificate is supposed to be delivered by a Certification Authority (CA) or policy owner to the server. The signature is applied to the SDPbody. <figure><artwork><![CDATA[body.</t> <sourcecode type="abnf"><![CDATA[ Signature= RSA ( SHA256 (<sdp>), <certificate> )]]></artwork> </figure> If]]></sourcecode> <t>If thesignatureSignature header field is not present, other validationmechanism MAYmechanisms <bcp14>MAY</bcp14> be implemented in order to provide assured quality with security and control. </t><t>Measurements: this</dd> <dt>Measurements:</dt> <dd><t>this header field carries the measurements of the quality parameters in PING and BWIDTH requests. The format is:<figure><artwork><![CDATA[</t> <sourcecode type="abnf"><![CDATA[ Measurements: "l=" " "|[0..9999] ", j=" " "|[0..9999] ", pl=" " "|[0.00 .. 100.00] ", bw=" " "|[0..999999]]]></artwork> </figure>]]></sourcecode> <t> Where "l" stands for latency followed by the measured value (inmilliseconds)ormilliseconds) or an empty space, "j" stands for jitter followed by the measured value (in milliseconds) or an empty space, "pl" stands forpacketlosspacket loss followed by the measured value (in percentage with two decimals) or an emptyspacespace, and "bw" stands for bandwidth followed by the measured value (in kbps) or an empty space.</t></list> </t></dd> </dl> </section> <sectiontitle="Specificanchor="sec-4.3.3" numbered="true" toc="default"> <name>Specific Q4S Response HeaderFields" anchor="section-4.3.3"><t><list style="symbols"><t>Expires: itsFields</name> <dl newline="false" spacing="normal" indent="6"> <dt>Expires:</dt> <dd><t>its purpose is to provide a sanity check and allow the server to close inactive sessions. If the client does not send a new request before the expiration time, the serverMAY<bcp14>MAY</bcp14> close the session. The valueMUST<bcp14>MUST</bcp14> be anintegerinteger, and the measurement units aremilliseconds.<vspace blankLines="1"/>milliseconds.</t> <t> In order to keep the sessionopenopen, the serverMUST<bcp14>MUST</bcp14> send a Q4S alert before the session expiration (Expiresheader),header field), with the same quality levels and an alert cause of "keep-alive". The purpose of this alert is to avoid TCPsockets (whichsockets, which were opened with READYmessage)message, from being closed, specially in NATscenarios. </t> </list> </t>scenarios.</t> </dd> </dl> </section> </section> <sectiontitle="Bodies" anchor="section-4.4"><t>anchor="sec-4.4" numbered="true" toc="default"> <name>Bodies</name> <t> Requests, including new requests defined in extensions to this specification,MAY<bcp14>MAY</bcp14> contain message bodies unless otherwise noted. The interpretation of the body depends on the request method.</t> <t> For response messages, the request method and the response status code determine the type and interpretation of any message body. All responsesMAY<bcp14>MAY</bcp14> include a body.</t> <t> The Internet media type of the message bodyMUST<bcp14>MUST</bcp14> be given by the Content-Type header field.</t> <sectiontitle="Encoding" anchor="section-4.4.1"><t>anchor="sec-4.4.1" numbered="true" toc="default"> <name>Encoding</name> <t> The bodyMUST NOT<bcp14>MUST NOT</bcp14> be compressed. This mechanism is valid for other protocols such as HTTP and SIP(RFC 3261<xreftarget="ref-22"/>),target="RFC3261" format="default"/>, but a compression/coding scheme will limitcertain logical implementations ofthe way the request isparsed, thus,parsed to certain logical implementations, thus making the protocol concept more implementation dependent. In addition, the bandwidth calculation may not be valid if compression is used. Therefore, the HTTP Accept-Encoding request header"Accept-Encoding"field cannot be used in Q4S withdifferentvaluesthan "identity"different from "identity", and if it is present in a request, the serverMUST<bcp14>MUST</bcp14> ignore it. In addition, the response header"Content-Encoding"field Content-Encoding is optional, but if present, the unique permitted value is "identity".</t> <t> The body length in bytesMUST<bcp14>MUST</bcp14> be provided by the Content-Length header field. The "chunked" transfer encoding of HTTP/1.1MUST NOT<bcp14>MUST NOT</bcp14> be used forQ4S (Note:Q4S.</t> <aside><t>Note: The chunked encoding modifies the body of a message in order to transfer it as a series of chunks, each one with its own sizeindicator.)</t>indicator.</t></aside> </section> </section> </section> <sectiontitle="Q4Sanchor="sec-5" numbered="true" toc="default"> <name>Q4S MethodDefinitions" anchor="section-5"><t>Definitions</name> <t> The Method token indicates the method to be performed on the resource identified by the Request-URI. The method iscase-sensitive.</t> <figure><artwork><![CDATA[case sensitive.</t> <sourcecode type="abnf"><![CDATA[ Method = "BEGIN" | "READY" | "PING" | "BWIDTH" | "Q4S-ALERT" | "Q4S-RECOVERY" | "CANCEL" | extension-method extension-method = token]]></artwork> </figure>]]></sourcecode> <t> The list of methods allowed by a resource can be specified in an"Allow"Allow header field(RFC 7231<xreftarget="ref-2"/>).target="RFC7231" format="default"/>. The return code of the response always notifies the client when a method is currently allowed on a resource, since the set of allowed methods can change dynamically. Any server applicationSHOULD<bcp14>SHOULD</bcp14> return the status code 405 (Method Not Allowed) if the method is known, but not allowed for the requested resource, and 501 (Not Implemented) if the method is unrecognized or not implemented by the server.</t> <sectiontitle="BEGIN" anchor="section-5.1"><t>anchor="sec-5.1" numbered="true" toc="default"> <name>BEGIN</name> <t> The BEGIN method requests information from a resource identified by a Q4S URI. Thesemanticspurpose of this method is to start thestarting of aquality session.</t> <t> This method isonlyused only during thehandshakeHandshake phase to retrieve the SDP containingsession idthe sess-id and all quality and operation parameters for the desired application to run.</t> <t> When a BEGIN message is received by the server, any current quality sessionMUST<bcp14>MUST</bcp14> becancelled,canceled, and a new session should be created.</t> <t> The response to a Q4S BEGIN request is not cacheable.</t> </section> <sectiontitle="READY" anchor="section-5.2"><t>anchor="sec-5.2" numbered="true" toc="default"> <name>READY</name> <t> The READY method is used to synchronize the starting time for the sending of PING and BWIDTH messages over UDP between clients and servers.The stageIncluding the Stage headerincludedfield in this method is mandatory.</t> <t> This message isonlyused only innegotiationNegotiation andcontinuityContinuity phases, and only just before making a measurement. Otherwise(out(outside of this context), the serverMUST<bcp14>MUST</bcp14> ignore this method.</t> </section> <sectiontitle="PING" anchor="section-5.3"><t>anchor="sec-5.3" numbered="true" toc="default"> <name>PING</name> <t> This message is used during thenegotiationNegotiation andcontinuityContinuity phases to measure the RTT and jitter of a session. The messageMUST<bcp14>MUST</bcp14> be sent only over UDP ports.</t> <t> The fundamental difference between the PING and BWIDTH requests is reflected in the different measurements achieved with them. PING is a short message, andMUSTit <bcp14>MUST</bcp14> be answered in order to measure RTT and jitter, whereas BWIDTH is a long message andMUST NOT<bcp14>MUST NOT</bcp14> be answered.</t> <t> PING is a request method that can be originated by either the clientbut also byor the server.Client MUSTThe client <bcp14>MUST</bcp14> also answer the server PING messages, assuming a "server role" for these messages during the measurement process.</t> <t>TheIncluding the Measurements headerincludedfield in this method is mandatory, and provides updated measurements values for latency,jitterjitter, and packet loss to the counterpart.</t> </section> <sectiontitle="BWIDTH" anchor="section-5.4"><t>anchor="sec-5.4" numbered="true" toc="default"> <name>BWIDTH</name> <t> This message is used only during the Negotiation phase to measure the bandwidth and packet loss of a session. The messageMUST<bcp14>MUST</bcp14> be sent only over UDP ports.</t> <t> BWIDTH is a request method that can be originated by either the clientbut also byor the server. Both(clientclient andserver) MUST NOTserver <bcp14>MUST NOT</bcp14> answer BWIDTH messages.</t> <t>TheIncluding the Measurements headerincludedfield in this method is mandatory and provides updated measurements values for bandwidth and packet loss to the counterpart.</t> </section> <sectiontitle="Q4S-ALERT" anchor="section-5.5"><t>anchor="sec-5.5" numbered="true" toc="default"> <name>Q4S-ALERT</name> <t> This is the request message that Q4S generates when the measurements indicate that quality constraints are being violated. It is used during thenegotiationNegotiation andcontinuityContinuity phases.</t> <t> This informative message indicates that the user experience is being degraded and includes the details of the problem (bandwidth, jitter, packet loss measurements). The Q4S-ALERT message does not contain any detail on the actions to be taken, whichdependsdepend on the agreements between all involved parties.</t> <t>Q4S-ALERT request does not have to be answered with a response message unlessUnless there is an error condition,but withan answer to a Q4S-ALERT request is optional and is formatted as a request Q4S-ALERT message. If there is an error condition, then a response message is sent. The response to a Q4S-ALERT request is not cacheable.</t> <t> This methodMUST<bcp14>MUST</bcp14> be initiated by the server in both alerting modes. In the Q4S-aware-network alerting mode, the Q4S-ALERT messages arefiredsent by the serverand sentto the client, advising the network to react by itself. In the Reactive alerting mode, alert notifications are triggered by the server stack and sent to theActuator(see Figure1Actuator (see <xref target="ref-reactive-scenario" format="default"/>, "Reactive Scenario").</t><figure><artwork><![CDATA[<artwork name="" type="" align="left" alt=""><![CDATA[ Client----q4s----SERVER STACK--->ACTUATOR-->APP OR POLICY SERVER ]]></artwork></figure><t> The way in which the server stack notifies the Actuator is implementation dependent, and the communication between the Actuator and the network policy server is defined by the protocol and API that the policy server implements.</t> </section> <sectiontitle="Q4S-RECOVERY" anchor="section-5.6"><t>anchor="sec-5.6" numbered="true" toc="default"> <name>Q4S-RECOVERY</name> <t> This is the request message that Q4S generates when the measurements indicate that qualityconstraints were being violated but theyconstraints, which had been violated, have been fulfilled during a period of timealready (recovery pause).("recovery-pause"). It is used during thenegotiationNegotiation andcontinuityContinuity phases.</t> <t> This informative message indicates that theqos-level"qos-level" could be increased gradually until the initialqos-level"qos-level" is recovered (theqos-level"qos-level" established at the beginningosof the sessionofthat was decreased during violation ofconstraints).constraints. See <xref target="sec-7.9" format="default"/>). The Q4S-RECOVERY message does not contain any detail on the actions to be taken, which depends on the agreements between all involved parties.</t> <t> The answer to a Q4S-RECOVERY request is formatted as a request Q4S-RECOVERY message. A Q4S-RECOVERY requestMUST NOT<bcp14>MUST NOT</bcp14> be answered with a response message unless there is an errorcondition, but with an answer formatted as a request Q4S-RECOVERY message.condition. The response to a Q4S-RECOVERY request is not cacheable.</t> <t>As forLike the Q4S-ALERT message, the Q4S-RECOVERY method is always initiated by the server in both alerting modes. In the Q4S-aware-network alerting mode, the Q4S-RECOVERY messages arefiredsent by the serverand sentto the client, advising the network to react by itself. In the Reactive alerting mode, recovery notifications are triggered by the server stack and sent to theActuator(see Figure1Actuator (see <xref target="ref-reactive-scenario" format="default"/>, "Reactive Scenario").</t> </section> <sectiontitle="CANCEL" anchor="section-5.7"><t>anchor="sec-5.7" numbered="true" toc="default"> <name>CANCEL</name> <t> Thesemanticspurpose of the CANCEL message is the release of the Q4Ssession idSession-Id and the possible resources assigned to the session. This message could be triggered by the Q4S stack or by the application using the stack (through animplementation dependentimplementation-dependent API).</t> <t> In the same way as Q4S-ALERT, CANCEL must not be answered with a response message, but with an answer formatted as a request Q4S-CANCEL message.</t> <t> In the Reactive scenario, the server stackMUST<bcp14>MUST</bcp14> react to the Q4S CANCEL messages received from the client by forwarding a cancel notification to the Actuator, in order to release possible assigned resources for the session (at the application or at the policy server). The ActuatorMUST<bcp14>MUST</bcp14> answer the cancel notification with a cancel acknowledge towards the server stack, acknowledging the reception.</t> </section> </section> <sectiontitle="Response Codes" anchor="section-6"><t>anchor="sec-6" numbered="true" toc="default"> <name>Response Codes</name> <t> Q4S response codes are used for TCP and UDP. However, inUDPUDP, only the response code 200 is used.</t> <t> The receiver of an unknown response code must take a generic action for the received error group(1XX, 2XX, 3XX, 4XX, 5XX, 6XX).(1xx, 2xx, 3xx, 4xx, 5xx, 6xx). In case of an unknown error group, the expected action should be the same as with6XXthe 6xx error group.</t> <sectiontitle="100 Trying" anchor="section-6.1"><t>anchor="sec-6.1" numbered="true" toc="default"> <name>100 Trying</name> <t> This response indicates that the request has been received by the next-hop server and that some unspecified action is being taken on behalf of this request (for example, a database is being consulted). This response, like all other provisional responses, stops retransmissions of a Q4S-ALERT during thealert-pause"alert-pause" time.</t> </section> <sectiontitle="Success 2xx" anchor="section-6.2"><t>anchor="sec-6.2" numbered="true" toc="default"> <name>Success 2xx</name> <t> 2xx responses give information about the success of a request.</t> <sectiontitle="200 OK" anchor="section-6.2.1"><t>anchor="sec-6.2.1" numbered="true" toc="default"> <name>200 OK</name> <t> The request has succeeded.</t> </section> </section> <sectiontitle="Redirection 3xx" anchor="section-6.3"><t>anchor="sec-6.3" numbered="true" toc="default"> <name>Redirection 3xx</name> <t> 3xx responses give information about the user's newlocation,location or about alternative services that might be able to satisfy the request.</t> <t> The requesting clientSHOULD<bcp14>SHOULD</bcp14> retry the request at the new address(es) given by the Location header field.</t> </section> <sectiontitle="Requestanchor="sec-6.4" numbered="true" toc="default"> <name>Request Failure4xx" anchor="section-6.4"><t>4xx</name> <t> 4xx responses are definite failure responses from a particular server. The clientSHOULD NOT<bcp14>SHOULD NOT</bcp14> retry the same request without modification (for example, adding appropriateheadersheader fields or SDP values). However, the same request to a different server might be successful.</t> <sectiontitle="400anchor="sec-6.4.1" numbered="true" toc="default"> <name>400 BadRequest" anchor="section-6.4.1"><t>Request</name> <t> The request could not be understood due to malformed syntax. The Reason-PhraseSHOULD<bcp14>SHOULD</bcp14> identify the syntax problem in more detail, for example, "Missing Sequence-Number header field".</t> </section> <sectiontitle="404anchor="sec-6.4.2" numbered="true" toc="default"> <name>404 NotFound" anchor="section-6.4.2"><t>Found</name> <t> The server has definitive information that the user does not exist at the domain specified in the Request-URI. This status is also returned if the domain in the Request-URI does not match any of the domains handled by the recipient of the request.</t> </section> <sectiontitle="405anchor="sec-6.4.3" numbered="true" toc="default"> <name>405 Method NotAllowed" anchor="section-6.4.3"><t>Allowed</name> <t> The method specified in the Request-Line is understood, but not allowed for the address identified by the Request-URI.</t> <t> The responseMUST<bcp14>MUST</bcp14> include an Allow header field containing a list of valid methods for the indicated address.</t> </section> <sectiontitle="406anchor="sec-6.4.4" numbered="true" toc="default"> <name>406 NotAcceptable" anchor="section-6.4.4"><t>Acceptable</name> <t> The resource identified by the request is only ableof generatingto generate response entities that have content characteristics that are not acceptable according to the Accept header field sent in the request.</t> </section> <sectiontitle="408anchor="sec-6.4.5" numbered="true" toc="default"> <name>408 RequestTimeout" anchor="section-6.4.5"><t>Timeout</name> <t> The server could not produce a response within a suitable amount of time, and the clientMAY<bcp14>MAY</bcp14> repeat the request without modifications at any latertime</t>time.</t> </section> <sectiontitle="413anchor="sec-6.4.6" numbered="true" toc="default"> <name>413 Request Entity TooLarge" anchor="section-6.4.6"><t>Large</name> <t> The server is refusing to process a request because the request entity-body is larger than the one that the server is willing or able to process. The serverMAY<bcp14>MAY</bcp14> close the connection to prevent the client from continuing the request.</t> </section> <sectiontitle="414anchor="sec-6.4.7" numbered="true" toc="default"> <name>414 Request-URI TooLong" anchor="section-6.4.7"><t>Long</name> <t> The server is refusing to process the request because the Request-URI is longer than the one that the server accepts.</t> </section> <sectiontitle="415anchor="sec-6.4.8" numbered="true" toc="default"> <name>415 Unsupported MediaType" anchor="section-6.4.8"><t>Type</name> <t> The server is refusing to process the request because the message body of the request is in a format not supported by the server for the requested method. The serverMUST<bcp14>MUST</bcp14> return a list of acceptable formats using the Accept, Accept-Encoding, or Accept-Language header field, depending on the specific problem with the content.</t> </section> <sectiontitle="416anchor="sec-6.4.9" numbered="true" toc="default"> <name>416 Unsupported URIScheme" anchor="section-6.4.9"><t>Scheme</name> <t> The server cannot process the request because the scheme of the URI in the Request-URI is unknown to the server.</t> </section> </section> <sectiontitle="Serveranchor="sec-6.5" numbered="true" toc="default"> <name>Server Failure5xx" anchor="section-6.5"><t>5xx</name> <t> 5xx responses are failure responses given when a server itself is having trouble.</t> <sectiontitle="500anchor="sec-6.5.1" numbered="true" toc="default"> <name>500 Server InternalError" anchor="section-6.5.1"><t>Error</name> <t> The server encountered an unexpected condition that prevented it from fulfilling the request. The clientMAY<bcp14>MAY</bcp14> display the specific error condition andMAY<bcp14>MAY</bcp14> retry the request after several seconds.</t> </section> <sectiontitle="501anchor="sec-6.5.2" numbered="true" toc="default"> <name>501 NotImplemented" anchor="section-6.5.2"><t>Implemented</name> <t> The server does not support the functionality required to fulfill the request. This is the appropriate response when aServerserver does not recognize the requestmethodmethod, and it is not capable of supporting it for any user.</t> <t> Note that a 405 (Method Not Allowed) is sent when the server recognizes the request method, but that method is not allowed or supported.</t> </section> <sectiontitle="503anchor="sec-6.5.3" numbered="true" toc="default"> <name>503 ServiceUnavailable" anchor="section-6.5.3"><t>Unavailable</name> <t> The server is temporarily unable to process the request due to a temporary overloading or maintenance of the server. The serverMAY<bcp14>MAY</bcp14> indicate when the client should retry the request in a Retry-After header field. If no Retry-After is given, the clientMUST<bcp14>MUST</bcp14> act as if it had received a 500 (Server Internal Error) response.</t> <t> A client receiving a 503 (Service Unavailable)SHOULD<bcp14>SHOULD</bcp14> attempt to forward the request to an alternate server. ItSHOULD NOT<bcp14>SHOULD NOT</bcp14> forward any other requests to that server for the duration specified in the Retry-After header field, if present.</t> <t> ServersMAY<bcp14>MAY</bcp14> refuse the connection or drop the request instead of responding with 503 (Service Unavailable).</t> </section> <sectiontitle="504anchor="sec-6.5.4" numbered="true" toc="default"> <name>504 ServerTime-out" anchor="section-6.5.4"><t>Time-Out</name> <t> The server did not receive a timely response from an external server it accessed in attempting to process the request.</t> </section> <sectiontitle="505anchor="sec-6.5.5" numbered="true" toc="default"> <name>505 Version NotSupported" anchor="section-6.5.5"><t>Supported</name> <t> The server does not support, or refuses to support, the Q4S protocol version that was used in the request. The server is indicating that it is unable or unwilling to complete the request using the same major version as the client, other than with this error message.</t> <t> In the case that the Q4S version is not supported, this error may be sent by the server inhandshakethe Handshake phase just after receiving the first BEGIN message from client.</t> </section> <sectiontitle="513anchor="sec-6.5.6" numbered="true" toc="default"> <name>513 Message TooLarge" anchor="section-6.5.6"><t>Large</name> <t> The server was unable to process the requestsincebecause the message length exceeded its capabilities.</t> </section> </section> <sectiontitle="Globalanchor="sec-6.6" numbered="true" toc="default"> <name>Global Failures6xx" anchor="section-6.6"><t>6xx</name> <t> 6xx responses indicate that a server has definitive information about a particular policy not satisfied for processing the request.</t> <sectiontitle="600 session does not exist" anchor="section-6.6.1"><t>anchor="sec-6.6.1" numbered="true" toc="default"> <name>600 Session Does Not Exist</name> <t> The Session-Id is notvalid</t>valid.</t> </section> <sectiontitle="601 quality level not allowed" anchor="section-6.6.2"><t>anchor="sec-6.6.2" numbered="true" toc="default"> <name>601 Quality Level Not Allowed</name> <t> TheQOS level"qos-level" requested is not allowed for thepair client/server</t>client/server pair.</t> </section> <sectiontitle="603anchor="sec-6.6.3" numbered="true" toc="default"> <name>603 Sessionnot allowed" anchor="section-6.6.3"><t>Not Allowed</name> <t> The session is not allowed due to some policy(number(the number of sessions allowed for the server is exceeded, or the time band of the Q4S-ALERT is not allowed for thepair client/server,client/server pair, etc.).</t> </section> <sectiontitle="604 authorization not allowed" anchor="section-6.6.4"><t>anchor="sec-6.6.4" numbered="true" toc="default"> <name>604 Authorization Not Allowed</name> <t> The policy server does not authorize the Q4S-ALERT quality session improvement operation due to an internal or external reason.</t> </section> </section> </section> <sectiontitle="Protocol" anchor="section-7"><t>anchor="sec-7" numbered="true" toc="default"> <name>Protocol</name> <t> This section describes the measurement procedures, the SDP structure of the Q4S messages, the different Q4S protocolphasesphases, and the messages exchanged in them.</t> <sectiontitle="Protocol Phases" anchor="section-7.1"><t>anchor="sec-7.1" numbered="true" toc="default"> <name>Protocol Phases</name> <t> All elements of the IP network contribute tothequality in terms of latency, jitter,bandwidthbandwidth, and packet loss. All these elements have their own quality policies in terms of priorities, traffic mode,etc.etc., and each element has its own way to manage the quality. The purpose of a quality connection is to establishanend-to-end communication with enough quality for the application to function flawlessly.</t> <t> To monitor quality constraints of the application, four phases are defined and can be seen infigure 5:</t><xref target="ref-session-lifetime-phases" format="default"/>:</t> <figuretitle="Session lifetime phases" anchor="ref-session-lifetime-phases"><artwork><![CDATA[anchor="ref-session-lifetime-phases"> <name>Session Lifetime Phases</name> <artwork name="" type="" align="left" alt=""><![CDATA[ +---------------------------------------------------------------+ | | | | | Handshake ---> Negotiation -+--> Continuity ----> Termination | | A | (app start) | (app end) | | | V A V A | | | violated | violated | | | | constraints | constraints | | | | | | |_______| ____| | | | | | +-------+ | | | | | | | | | +------+ +---------------------+ | | | +---------------------------------------------------------------+ ]]></artwork> </figure><t><list style="symbols"><t>Handshake phase: in<dl spacing="normal"> <dt>Handshake phase:</dt><dd>in which the server is contacted by theclientclient, and in the answermessagemessage, the quality constraints for the applicationisare communicatedembedded in an SDP.</t> <t>Negotiation phase:in the embedded SDP.</dd> <dt>Negotiation phase:</dt><dd>in which the quality of the connection is measured in both directions (latency, jitter,bandwidthbandwidth, and packet loss), and Q4S messages may be sent in order to alert if the measured quality does not meet the constraints. This phase is iterative until quality constraints are reached, or the session iscancelledcanceled after a number of measurement cycles with consistent violation of the quality constraints. The number of measurement cycles executed depends on theqos-level"qos-level", which is incremented in each cycle until a maximumqos-level"qos-level" value is reached. Just after reaching the quality requirements, Q4S provides a simple optional mechanism using HTTP to start theapplication.</t> <t>Continuity phase: inapplication.</dd> <dt>Continuity phase:</dt><dd>in which quality is continuously measured. In thisphasephase, the measurementsMUST<bcp14>MUST</bcp14> avoid disturbing the application by consuming network resources. If quality constraints are not met, the server stack will notify the Actuator with an alert notification. If later the quality improves, the server stack will notify the Actuator, in this case with a recovery notification. After several alert notifications with no quality improvements, the Q4S stackSHOULD<bcp14>SHOULD</bcp14> move to the Terminationphase.</t> <t>Termination phase: inphase.</dd> <dt>Termination phase:</dt><dd>in which the Q4S session is terminated. The application may be closedtooalso or may notstart.</t> </list> </t>start.</dd> </dl> </section> <sectiontitle="SDP Structure" anchor="section-7.2"><t>anchor="sec-7.2" numbered="true" toc="default"> <name>SDP Structure</name> <t> The original goal of SDP was to announce necessary information for the participants and multicast MBONE (Multicast Backbone) applications. Right now, its use has been extended to the announcement and the negotiation of multimedia sessions. The purpose of Q4S is not to establish media stream sessions, but to monitor a quality connection. This connection may be later used to establish any type of session including media sessions; Q4S does not impose any conditions on the type of communication requiring quality parameters.</t> <t> SDP will be used by Q4S to exchange quality constraints and will therefore always have all the mediaattributes ("m")descriptions ("m=") set to zero.</t> <t> The SDP embedded in the messages is the container of the quality parameters. As these may vary depending on the direction of the communication (to and from theclient)client), all quality parameters need to specify the uplink and downlink values: <uplink> /<downlink>.<downlink> (see <xref target="sec-7.5.3" format="default"/> for an example). When one or both of these values are empty, itMUST<bcp14>MUST</bcp14> be understood as needing no constraint on that parameter and/or that direction.</t> <t> The uplink directionMUST<bcp14>MUST</bcp14> be considered as being the communication from the client to the server. The downlink directionMUST<bcp14>MUST</bcp14> be considered as being the communication from the server to the client.</t> <t> The SDP information can comprise all or some of the following parameters shown in the example below. This is an example of an SDP message used by Q4S included in the 200 OK response to a Q4S BEGIN request.</t><figure><artwork><![CDATA[<sourcecode type="sdp"><![CDATA[ v=0 o=q4s-UA 53655765 2353687637 IN IP4 192.0.2.33 s=Q4S i=Q4S parameters t=0 0 a=qos-level:0/0 a=alerting-mode:Reactive a=alert-pause:5000 a=public-address:client IP4 198.51.100.51 a=public-address:server IP4 198.51.100.58 a=measurement:procedure default(50/50,75/75,5000,40/80,100/256) a=latency:40 a=jitter:10/10 a=bandwidth:20/6000 a=packetloss:0.50/0.50 a=flow:app clientListeningPort TCP/10000-20000 a=flow:app clientListeningPort UDP/15000-18000 a=flow:app serverListeningPort TCP/56000 a=flow:app serverListeningPort UDP/56000 a=flow:q4s clientListeningPort UDP/55000 a=flow:q4s clientListeningPort TCP/55001 a=flow:q4s serverListeningPort UDP/56000 a=flow:q4s serverListeningPort TCP/56001]]></artwork> </figure>]]></sourcecode> <t> As quality constraints may be changed by applications at any time during the Q4S session lifetime, any Q4S 200 OK response sent by the server to the client in the Negotiation and Continuity phases could also include an SDP body with the new quality requirements stated by the applications from then on. Therefore, in response to any PING request sent by the client to the server, the server could send a Q4S 200 OK with an embedded SDP messageembeddedthat specifies new quality constraints requested by the application.</t> <sectiontitle=""qos-level" attribute" anchor="section-7.2.1"><t>anchor="sec-7.2.1" numbered="true" toc="default"> <name>"qos-level" Attribute</name> <t> The "qos-level" attribute contains the QoS level for uplink and downlink. Default values are 0 for both directions. The meaning of each level is out of scope of Q4S, but a higher levelSHOULD<bcp14>SHOULD</bcp14> correspond to a better service quality.</t> <t> Appropriate attribute values: [0..9] "/" [0..9]</t> <t> The "qos-level" attribute may be changed during the sessionlifetimelifetime, raising or lowering the value as necessary following the network measurements and the application needs.</t> </section> <sectiontitle=""alerting-mode" attribute" anchor="section-7.2.2"><t>anchor="sec-7.2.2" numbered="true" toc="default"> <name>"alerting-mode" Attribute</name> <t> The "alerting-mode" attribute specifies the player in charge of triggering Q4S alerts in the case of constraint violation. There are two possible values:</t> <t> Appropriate attribute values: <"Q4S-aware-network"|"Reactive"></t><t><list style="format (%c)"> <t>Q4S-aware-network: Q4S ALERT<dl newline="false" spacing="normal" indent="6"> <dt>Q4S-aware-network:</dt> <dd>Q4S-ALERT messages are triggered by the server to the client. In thiscasecase, the network is supposed to be Q4S aware, and reacts by itself to thesealerts. </t> <t> Reactive: alertalerts.</dd> <dt> Reactive:</dt> <dd>alert notifications are sent by the server stack to the Actuator. In thiscasecase, the network is not Q4Sawareaware, and a specific node (Actuator) is in charge of triggering tuningmechanisms.,mechanisms, either on the network or in the application.</t> </list> </t></dd> </dl> <t> The "alerting-mode" attribute isoptionaloptional, and if notpresentpresent, Reactive alerting mode is assumed.</t> </section> <sectiontitle=""alert-pause" attribute" anchor="section-7.2.3"><t>anchor="sec-7.2.3" numbered="true" toc="default"> <name>"alert-pause" Attribute</name> <t> In the Q4S-aware-network scenario, the "alert-pause" attribute specifies the amount of time (in milliseconds) the server waits between consecutiveQ4S ALERTQ4S-ALERT messages sent to the client. In the Reactive scenario, the "alert-pause" attribute specifies the amount of time (in milliseconds) the server stack waits between consecutive alert notifications sent to the Actuator. Measurements are not stopped in Negotiation or ContinuityPhasesphases during this period of time, but noQ4S ALERTQ4S-ALERT messages or alert notifications are fired, even with violated quality constraints, allowing for either network reconfigurations or application adjustments.</t> <t> Appropriate attribute values: [0..60000]</t> </section> <sectiontitle=""recovery-pause" attribute" anchor="section-7.2.4"><t>anchor="sec-7.2.4" numbered="true" toc="default"> <name>"recovery-pause" Attribute</name> <t> In the Q4S-aware-network scenario, the "recovery-pause" attribute specifies the amount of time (in milliseconds) the server waits for initiating theqos-level"qos-level" recovery process. Once the recovery process has started, the "recovery-pause" attribute also states the amount of time (in milliseconds) between consecutive Q4S-RECOVERY messages sent by the server to the client (in the Q4S-aware-networkscenario),scenario) or between recovery notifications sent by the server stack to the Actuator (in the Reactive scenario).</t> <t> Appropriate attribute values: [0..60000]</t> </section> <sectiontitle=""public-address" attributes" anchor="section-7.2.5"><t>anchor="sec-7.2.5" numbered="true" toc="default"> <name>"public-address" Attributes</name> <t> This attribute contains the public IP address of the client and the server. The server fills these attributes withhisits own public IP address and the public IP address of the first message received from the client in thehandshakeHandshake phase.</t> <t> The purpose of these attributes is to make available the addressing information to the network policy server or other external entities in charge of processing Q4S-ALERT messages.</t> <t> Appropriate attributevalues:<"client"|"server"><"IP4"|"IP6">values: <"client"|"server"> <"IP4"|"IP6"> <value of IP address></t> </section> <sectiontitle=""latency" attribute" anchor="section-7.2.6"><t>anchor="sec-7.2.6" numbered="true" toc="default"> <name>"latency" Attribute</name> <t> The maximum latency (considered equal for uplink and downlink) toleranceareis specified in the "latency" attribute, expressed in milliseconds. In the Q4S-aware-network scenario, if the latency constraints are not met, a Q4S-ALERT method will be sent to the client. In the Reactive scenario, if the latency constraints are not met, an alert notification will be sent to the Actuator. If the "latency" attribute is not present or has a 0 value, no latency constraints need to bemetmet, and no measurementsMAY<bcp14>MAY</bcp14> be taken.</t> <t> Appropriate attribute values: [0..9999]</t> </section> <sectiontitle=""jitter" attribute" anchor="section-7.2.7"><t>anchor="sec-7.2.7" numbered="true" toc="default"> <name>"jitter" Attribute</name> <t> The maximum uplink and downlink jitter toleranceareis specified in the "jitter" attribute, expressed in milliseconds. In the Q4S-aware-network scenario, if the jitter constraints are not met, a Q4S-ALERT method will be sent to the client. In the Reactive scenario, if the latency constraints are not met, an alert notification will be sent to the Actuator. If the "jitter" attribute is not present or has a 0 value, no jitter constraints need to bemetmet, and no measurementsMAY<bcp14>MAY</bcp14> be taken.</t> <t> Appropriate attribute values: [0..9999] "/" [0..9999]</t> </section> <sectiontitle=""bandwidth" attribute" anchor="section-7.2.8"><t>anchor="sec-7.2.8" numbered="true" toc="default"> <name>"bandwidth" Attribute</name> <t> The minimum uplink and downlink bandwidthareis specified in the "bandwidth" attribute, expressed in kbps. In the Q4S-aware-network scenario, if the bandwidth constraints are not met, a Q4S-ALERT method will be sent to the client. In the Reactive scenario, an alert notification will be sent to the Actuator. If the "bandwidth" attribute is not present or has a 0 value, no bandwidth constraints need to bemetmet, and no measurementsMAY<bcp14>MAY</bcp14> be taken.</t> <t> Appropriate attribute values: [0..99999] "/" [0..99999]</t> </section> <sectiontitle=""packetloss" attribute" anchor="section-7.2.9"><t>anchor="sec-7.2.9" numbered="true" toc="default"> <name>"packetloss" Attribute</name> <t> The maximum uplink and downlink packet loss toleranceareis specified in the "packetloss" attribute expressed in percentage (two decimal accuracy). In the Q4S-aware-network scenario, if the packetloss constraints are not met, a Q4S-ALERT method will be sent to the client. In the Reactive scenario, an alert notification will be sent to the Actuator. If the "packetloss" attribute is not present or has a 0 value, nopacketlosspacket loss constraints need to bemetmet, and no measurementsMAY<bcp14>MAY</bcp14> be taken.</t> <t> Appropriate attribute values: [0.00 ..100.00] "/"[0.00 ..100.00]</t> </section> <sectiontitle=""flow" attributes" anchor="section-7.2.10"><t>anchor="sec-7.2.10" numbered="true" toc="default"> <name>"flow" Attributes</name> <t> These attributes specify the flows (protocol, destination IP/ports) of data over TCP and UDP ports to be used in uplink and downlink communications.</t> <t> Several "flow" attributes can be defined. These flows identify the listening port (client or server), the protocol (TCPor UDP) (RFC 793<xreftarget="ref-16"/> and RFC 768target="RFC0793" format="default"/> or UDP <xreftarget="ref-17"/>)target="RFC0768" format="default"/>) with the range of ports that are going to be used by the application and, of course, by the Q4S protocol (for quality measurements). All defined flows(app("app" andq4s)"q4s") will be considered within the same quality profile, which is determined by theqos-level"qos-level" attribute in each direction. This allows us to assume that measurements onq4s"q4s" flows are the sameexperimentedas experienced by theapplicationapplication, which is usingapp"app" flows.</t> <t> DuringnegotiationNegotiation andcontinuity phasesContinuity phases, the specified Q4S ports in the "flow:q4s" attributes of SDP will be used for Q4S messages.</t> <t> The Q4S flows comprise two UDP flows and two TCP flows (one uplink and one downlink for eachone)one), whereas application trafficMAY<bcp14>MAY</bcp14> consist of many flows, depending on its nature. ThehandshakeHandshake phase takes place through the Q4S Contact URI, using the standard Q4S TCP port. However, thenegotiationNegotiation andcontinuityContinuity phases will take place on thespecifiedQ4S ports (UDP and TCP) specified in the SDP.</t> <t> The "clientListeningPort" is a portinon which the client listens for server requests andMUST<bcp14>MUST</bcp14> be used as the origin port of client responses. The "serverListeningPort" is a portinon which the server is listening for incoming messages from the client. The origin port of server responses may be different than the "serverListeningPort" value.</t> <t> If "clientListeningPort" is zero(a=flow:q4s("a=flow:q4s clientListeningPortTCP/0),TCP/0"), the clientMAY<bcp14>MAY</bcp14> choose one randomlyasper OS standard rules. Client ports inside the SDP must always be matched against actual received port values on the server side in order to deal withNAT/NATPNAT/NAPT devices. If a zero value or incorrect value is present, the server must set the value to the received origin port in the next message with SDP (200 OK,ALERTALERT, and CANCEL messages).</t><figure><artwork><![CDATA[<sourcecode type="abnf"><![CDATA[ Attribute values: <"q4s"|"app"> <"serverListeningPort"|"clientListeningPort"> <"UDP"|"TCP"><0..65535>[<0..65535> [ "-" [0..65535]]]]></artwork> </figure>]]></sourcecode> </section> <sectiontitle=""measurement" attributes" anchor="section-7.2.11"><t>anchor="sec-7.2.11" numbered="true" toc="default"> <name>"measurement" Attributes</name> <t> These attributes contain the measurement procedure and the results of the quality measurements.</t> <t> Measurement parameters are included using the session attribute "measurement". The first measurement parameter is the procedure. Q4S provides a "default" procedure for measurements, but others like RTP/RTCP might be used and defined later. This document will only define and explain the "default" procedure.</t> <t> In the initial clientrequestrequest, a set of measurement procedures can be sent to the server for negotiation. One measurement procedure lineMUST<bcp14>MUST</bcp14> be included in the SDP message for each proposed method. The serverMUST<bcp14>MUST</bcp14> answer with only one line with the chosen procedure.</t> <t> For each procedure, a set of values of parameters separated by "," can be included in the same attribute line. The amount and type of parameters depends on the procedure type.</t> <t> In the followingexampleexample, the "default" procedure type is chosen:</t><figure><artwork><![CDATA[<sourcecode type="sdp"><![CDATA[ a=measurement:procedure default(50/50,75/75,5000,40/80,100/256)]]></artwork></figure>]]></sourcecode> <t> In the "default" procedure, the meaning of these parametersis:</t> <t><list style="symbols"><t>Theis the following:</t> <ul spacing="normal"> <li>The first parameter is the interval of time (in milliseconds) between PING requests during thenegotiationNegotiation phase. Uplink and downlink values from the client's point of view are separated by "/". This allowshavingdifferent responsiveness values depending on the control resources used in eachdirection.</t> <t>Thedirection.</li> <li>The second parameter is the time interval (in milliseconds) between PING requests during thecontinuityContinuity phase. Uplink and downlink values are separated by "/". This allowshavingtwo different responsiveness values depending on the control resources used in eachdirection.</t> <t>Thedirection.</li> <li>The third parameter is the time interval to be used to measure bandwidth during thenegotiation phase.</t> <t>TheNegotiation phase.</li> <li>The fourth parameter indicates the window size for jitter and latency calculations. Uplink and downlink values are separated by"/".</t> <t>The"/".</li> <li>The fifth parameter indicates the window size for packet loss calculations. Uplink and downlink values are separated by"/".</t> </list> </t>"/".</li> </ul> <t> There are four moremeasurement"measurement" attributes:</t><figure><artwork><![CDATA[<sourcecode type="sdp"><![CDATA[ a=measurement:latency 45 a=measurement:jitter 3/12 a=measurement:bandwidth 200/9800 a=measurement:packetloss 0.00/1.00]]></artwork> </figure>]]></sourcecode> <t> Thelatency, jitter, bandwidth"measurement:latency", "measurement:jitter", "measurement:bandwidth", andpacket-loss measurement"measurement:packetloss" attributes contain the values measured for each of these quality parameters in uplink and downlink directions. Notice that latency is considered equal for uplink and downlink directions. Quality parameter values in thesemeasurement"measurement" attributes provide a snapshot of the quality reached andMUST<bcp14>MUST</bcp14> only be included in Q4S-ALERT messages in the SDP body such that they can be protected from malicious attacks as these alerts include a signature of the SDP body in the header. The rest of the messages will include the measured values in the Measurementsheader.</t>header field.</t> <t> In the case ofprocedure "default",the "default" procedure, the valid valuesare:</t> <figure><artwork><![CDATA[are as follows:</t> <sourcecode type="abnf"><![CDATA[ a=measurement:procedure default,[0..999]"/" [0..999] "," [0..999] "/" [0..999] "," [0..9999] "," [0..999]/[0..999] "," [0..999]/[0..999]]]></artwork> </figure>]]></sourcecode> </section> <sectiontitle=""max-content-length" attribute" anchor="section-7.2.12"><t>anchor="sec-7.2.12" numbered="true" toc="default"> <name>"max-content-length" Attribute</name> <t> The adaptation of measurement traffic to approximate the actual data streams' characteristics is convenient to accurately estimate the expected QoS for applications. Particularly, packet size can have a remarkable effect on bandwidth estimations. Moreover, this can produce problems depending on the MTU of the end hosts and links along the path.</t> <t> Therefore, the maximum content lengthMAY<bcp14>MAY</bcp14> be set in an attribute denoted as "max-content-length". Its valueMUST<bcp14>MUST</bcp14> be given in bytes andMUST NOT<bcp14>MUST NOT</bcp14> include application, transport,networknetwork, or link layer headers, i.e., size of the content length at the application layer. If not set, the valueMUST<bcp14>MUST</bcp14> be 1000 bytes.</t> <t> Furthermore, this attributeMAY<bcp14>MAY</bcp14> be used toinform aboutcommunicate MTU limits inend points,endpoints, hence reducing possible bias as a result of network-layer fragmentation.</t> <t> For instance:</t> <t> a=max-content-length:1300</t> </section> </section> <sectiontitle="Measurements" anchor="section-7.3"><t>anchor="sec-7.3" numbered="true" toc="default"> <name>Measurements</name> <t> This section describes the way quality parameters are measured as defined by the "default" procedure. MeasurementsMUST<bcp14>MUST</bcp14> be taken for any quality parameter with constraints, that is, specified in the SDP attributes with non-zero values. Fornon-present attributesabsent attributes, measurementsMAY<bcp14>MAY</bcp14> be omitted.</t> <sectiontitle="Latency" anchor="section-7.3.1"><t>anchor="sec-7.3.1" numbered="true" toc="default"> <name>Latency</name> <t> Latency measurements will be performed if thelatency"latency" attribute and/or theapplication latency"a=measurement:latency" attribute are present andwithhave non-zero values.</t> <t> Q4S defines a PING method in order to exchange packets between the client and the server. Based on this PINGexchangeexchange, the client and the server are able to calculate the round-trip time (RTT). The RTT is the sum of downlink latency (normally named "reverse latency") and uplink latency (normally named "forward latency").</t> <t> At least 255 samples of RTTMUST<bcp14>MUST</bcp14> be taken by the client and server. As the forward and reverse latencies are impossible to measure, the client and server will assume that both latencies are identical (symmetric network assumption). The latency will therefore be calculated as the statistical median value of all the RTT samples divided by 2.</t> </section> <sectiontitle="Jitter" anchor="section-7.3.2"><t>anchor="sec-7.3.2" numbered="true" toc="default"> <name>Jitter</name> <t> Jitter measurements will be performed if thejitter"jitter" attribute and/or theapplication jitter"a=measurement:jitter" attribute are present andwithhave non-zero values.</t> <t> The jitter can be calculated independently by the client and by the server. The downlink jitter is calculated by the client taking into account the time interval between PING requests as defined by themeasurement procedure"measurement:procedure" attribute in the first or second parameter depending on the Q4S protocol phase. The client and the serverMUST<bcp14>MUST</bcp14> send these PING requests at the specified intervals. The client measures the downlinkjitterjitter, whereas the server measures the uplink jitter. Note that PING responses are not taken into account when calculating jitter values.</t> <t> Every time a PING requestmessageis received by an endpoint (either server or client), the corresponding jitter value is updatedusing the Statistical Jitter value calculated onwith thefirst 255 packets received usingstatistical jitter value, which is the arithmetic mean of the absolute values of elapsedtimes.</t>times calculated on the first 255 packets received. </t> <t> Each endpoint sends a PING periodically with a fixed interval, and each value of "elapsed time" (ET) should be very close to this interval. If a PING message is lost, theelapsed timeET value is doubled. Identifying lost PING messages, however, is not an issue because all PING messages are labeled with a Sequence-Numberheader. Thereforeheader field. Therefore, the receiver can discard thiselapsed timeET value.</t> <t> In order to have the first jitter sample, the receiverMUST<bcp14>MUST</bcp14> wait until it receives 3 PING requests, because each ET is the time between twoPINGsPINGs, and aJitterjitter measurement needs at least two ET.</t> <t> The client measures the values of RTT and downlinkjitterjitter, and the server measures RTT and uplink jitter, but all measurements are shared with the counterpart by means of"Measurements"the Measurements header field of the PING message.</t> </section> <sectiontitle="Bandwidth" anchor="section-7.3.3"><t>anchor="sec-7.3.3" numbered="true" toc="default"> <name>Bandwidth</name> <t> Bandwidth measurements will be performed if thebandwidth"bandwidth" attribute and/or theapplication bandwidth"a=measurement:bandwidth" attribute is present andwithhas non-zero values.</t> <t> In order to measure the available bandwidth, both the client and the serverMUST<bcp14>MUST</bcp14> start sending BWIDTH messages simultaneously using the UDP control ports exchanged during thehandshakeHandshake phase in the SDPmessage,message at the needed rate to verify the availability of the bandwidth constraint in each direction. The messages are sent during the period of time defined in the third parameter of the SDPmeasurement default procedure"measurement:procedure default" attribute inmillisecond units.</t>milliseconds.</t> <figuretitle="Bandwidthanchor="ref-bandwidth-and-packet-loss-measurements"> <name>Bandwidth andpacket loss measurements" anchor="ref-bandwidth-and-packet-loss-measurements"><artwork><![CDATA[Packet Loss Measurements</name> <artwork name="" type="" align="left" alt=""><![CDATA[ a=measurement:procedure default(50/50,75/75,5000,256/256,256/256) +------------------------------------------------+ | Rate | | A | | | | |downlink rate-|-------------------+ <-- traffic | | | | sent by | | | | server | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | uplink rate-|-------------------+ <-- traffic | | | | sent by | | | | client | | | | | | | | | | |---|---|---|---|---|----> time | | 0 1 2 3 4 5 (sec.) | | | +------------------------------------------------+ ]]></artwork> </figure> <t> The goal of these measurements is not to identify the available bandwidth of the communicationpathpath, but to determine if the required bandwidth is available, meeting the application's constraints. Therefore, the requested bandwidthMUST<bcp14>MUST</bcp14> be measured sending only the highestbit ratebitrate required by the bandwidth attribute. This is illustrated inFigure 6.</t><xref target="ref-bandwidth-and-packet-loss-measurements" format="default"/>.</t> <t>During bandwidth measurement time,ALERTS are notexpected,expected during bandwidth measurement, but only at the end of the measurement time.</t> <t> When measuring bandwidth, all BWIDTH requests sentMUST<bcp14>MUST</bcp14> be 1 kilobyte in length (UDP payload length by default),and MUSTthey <bcp14>MUST</bcp14> include a Sequence-Number header field with a sequential number starting at 0, and their contentMUST<bcp14>MUST</bcp14> consist of randomly generated values to minimize the effect of compression elements along the path. The Sequence-NumberMUST<bcp14>MUST</bcp14> be incremented by 1 with each BWIDTH packet sent. If any measurement stage needs to be repeated, the sequence numberMUST<bcp14>MUST</bcp14> start at zero again. BWIDTH requestsMUST NOT<bcp14>MUST NOT</bcp14> be answered. Examples:</t><figure><artwork><![CDATA[<artwork name="" type="" align="left" alt=""><![CDATA[ Client message: ========================= BWIDTH q4s://www.example.com Q4S/1.0 User-Agent: q4s-ua-experimental-1.0 Session-Id: 53655765 Sequence-Number: 0 Content-Type: text Content-Length: XXXX Measurements: l=22, j=10, pl=0.00, bw=3000 VkZaU1FrNVZNVlZSV0doT1ZrZ (to complete up to "max-content- length" bytes UDP payload length) ========================= ]]></artwork></figure><t> The clientMUST<bcp14>MUST</bcp14> send BWIDTH packets to the server to allow the server to measure the uplink bandwidth. The serverMUST<bcp14>MUST</bcp14> send BWIDTH packets to the client to allow the client to measure the downlink bandwidth.</t><figure><artwork><![CDATA[<artwork name="" type="" align="left" alt=""><![CDATA[ Server message: ========================= BWIDTH q4s://www.example.com Q4S/1.0 Session-Id: 53655765 Sequence-Number: 0 Content-Type: text Content-Length: XXXX Measurements: l=22, j=7, pl=0.00, bw=200 ZY0VaT1ZURlZVVmhyUFE9PQ (to complete up to max-content- length UDP payload length) ========================= ]]></artwork></figure></section> <sectiontitle="Packet loss" anchor="section-7.3.4"><t>anchor="sec-7.3.4" numbered="true" toc="default"> <name>Packet Loss</name> <t> Packet loss and bandwidth are measured simultaneously using the BWIDTH packets sent by both the client and the server. Because the BWIDTH packets contain a Sequence-Number header field incremented sequentially with each sent packet, lost packets can be easily identified. The lost packetsMUST<bcp14>MUST</bcp14> be counted during the measurement time.</t> </section> </section> <sectiontitle="Handshake Phase" anchor="section-7.4"><t>anchor="sec-7.4" numbered="true" toc="default"> <name>Handshake Phase</name> <t> The first phase consists of a Q4S BEGIN method issued from the client to the server as shown inFigure 7.</t><xref target="ref-handshake-phase" format="default"/>.</t> <t> The first Q4S messageMUST<bcp14>MUST</bcp14> have a special URI(RFC 3986<xreftarget="ref-12"/>),target="RFC3986" format="default"/>, which forces the use of the Q4S protocol if it is implemented in a standard web browser.</t> <t> This URI, named "Contact URI", is used to request the start of a session. Its schemeMUST<bcp14>MUST</bcp14> be:</t><figure><artwork><![CDATA[<sourcecode type="abnf"><![CDATA[ "q4s:" "//" host [":" port] [path["?" query]]]></artwork> </figure>]]></sourcecode> <t> Optionally, the client can send the desired quality parameters enclosed in the body of the message as an SDP document. The serverMAY<bcp14>MAY</bcp14> take them into account when building the answer message with the final values in the SDP body, following arequest / responserequest/response schema(RFC 3264<xreftarget="ref-13"/>).</t>target="RFC3264" format="default"/>.</t> <t> If the request is accepted, the serverMUST<bcp14>MUST</bcp14> answer it with a Q4S 200 OK message, whichMUST<bcp14>MUST</bcp14> contain an SDP body(RFC 4566<xreftarget="ref-10"/>)target="RFC4566" format="default"/> with the assignedsession idsess-id (embedded in the"o"SDPparameter),"o=" line), the IP addresses to be used, the flow ports to be used, the measurement procedure to befollowedfollowed, and information about the required quality constraints. Additionally, thealerting-mode"alerting-mode" andalert-pause"alert-pause" timeparametersattributes may be included. Q4S responses should use the protocol designator "Q4S/1.0".</t> <t> After these two messages are exchanged, the first phase is completed. The quality parameter thresholds have been sent to the client. The next step is to measure the actual quality of the communication path between the client and the server and alert if the Service Level Agreement (SLA) is being violated.</t> <figuretitle="Handshake phase" anchor="ref-handshake-phase"><artwork><![CDATA[anchor="ref-handshake-phase"> <name>Handshake Phase</name> <artwork name="" type="" align="left" alt=""><![CDATA[ +------------------------------------------------+ | | | Client Server | | | | ------- Q4S BEGIN ------------> | | | | <------ Q4S 200 OK ------------ | | | | | +------------------------------------------------+ ]]></artwork> </figure> <t>ExampleThe following is an example ofClient Requesta client request andServer Answer:</t> <figure><artwork><![CDATA[a server answer:</t> <artwork name="" type="" align="left" alt=""><![CDATA[ Client Request: ========================= BEGIN q4s://www.example.com Q4S/1.0 Content-Type: application/sdp User-Agent: q4s-ua-experimental-1.0 Content-Length: 142 (SDP not shown) ========================= Server Answer: ========================= Q4S/1.0 200 OK Date: Mon, 10 Jun 2010 10:00:01 GMT Content-Type: application/sdp Expires: 3000 Signature: 6ec1ba40e2adf2d783de530ae254acd4f3477ac4 Content-Length: 131 (SDP not shown) =========================The headers]]></artwork> <t>The header fields used are explained insection 4.3. ]]></artwork> </figure><xref target="sec-4.3" format="default"/>.</t> </section> <sectiontitle="Negotiation Phase" anchor="section-7.5"><t>anchor="sec-7.5" numbered="true" toc="default"> <name>Negotiation Phase</name> <t> ThenegotiationNegotiation phase is in charge of measuring the quality parameters and verifying that the communication paths meet the required quality constraintsonin both directions as specified in the SDP body.</t> <t> The measured parameters will be compared with the quality constraints specified in the SDP body. If the quality session is compliant with all the qualityconstraintsconstraints, the application can start.</t><t><list style="symbols"><t>If<t>If the quality constraints are not met, a higher quality service level will be demanded. Depending on the scenario, this quality upgrade will be managed as follows:In</t> <dl> <dt>In the Q4S-aware-networkscenario: ascenario:</dt><dd>a Q4S-ALERT method will be triggered by the server to theclientclient, and the client will answer with the same Q4S-ALERT method. After receiving the same Q4S-ALERT from the counterpart, no other alerts will be triggered by the server during the "alert-pause" period oftime,time in order to allow the network to react, but measurements will continue to be taken to achieve early detection of improved network quality conditions and a fast applicationstart.</t> <t>Instart.</dd> <dt>In the Reactivescenario: anscenario:</dt><dd>an alert notification will be sent by the server stack to the Actuator, and the Actuator will answer with an alert acknowledgement. After receiving the alert acknowledgement from the Actuator, the server stack will not send other alert notifications during the "alert-pause" period oftime,time in order to allow the Actuator to react and trigger actions on the application or on the policy server, but measurements will continue to be taken to achieve early detection of improved network quality conditions and a fast applicationstart.</t> </list> </t>start.</dd> </dl> <t> In both scenarios stated above, if after several measurement cycles, the network constraints cannot bemetmet, the quality session is terminated. Concretelywhenwhen, under all possible actions taken byActuatorActuator, the quality remains below requirements, the session must be terminated.</t> <t> The steps to be taken in this phase depend on the measurement procedure exchanged during thehandshakeHandshake phase. This document only describes the "default" procedure, but others can be used, like RTP/RTCP(RFC 3550<xreftarget="ref-18"/>).</t>target="RFC3550" format="default"/>.</t> <t> Measurements of latency and jitter aredonemade by calculating the differences in the arrival times of packets and can be achieved with little bandwidth consumption. The bandwidth measurement, on the other hand, involves higher bandwidth consumption in both directions (uplink and downlink).</t> <t> To avoid wasting unnecessary networkresourcesresources, these two types of measurements will be performed in two separate stages. If the required latencies and jitters cannot be reached, it makes no sense to waste network resources measuring bandwidth. In addition, if achieving the required latency and jitter thresholds implies upgrading the quality session level, the chance of obtaining compliant bandwidth measurements without retries is higher, saving network traffic again. Therefore, thedefault procedure,"default" procedure determines that the measurements are taken in two stages:<list style="hanging" hangIndent="6"> <t hangText="Stage 0:"></t> <dl newline="false" spacing="normal" indent="6"> <dt>Stage 0:</dt> <dd> Measurement of latencies,jittersjitters, and packetloss</t> <t hangText="Stage 1:">loss</dd> <dt>Stage 1:</dt> <dd> Measurement of bandwidths and packetloss</t> </list> </t>loss</dd> </dl> <t> Notice that packet loss can be measured in both stages, as all messages exchanged include asequence-numberSequence-Number header field that allows for easy packet loss detection.</t> <t> The client starts thenegotiationNegotiation phase by sending a READY request using the TCP Q4S ports defined in the SDP. This READY request includes a"Stage"Stage header field that indicates the measurement stage.</t> <t> If either jitter,latencylatency, or both are specified, thenegotiationNegotiation phase begins with the measurement of latencies and jitters (stage 0). If none of those attributesareis specified, stage 0 is skipped.</t> <sectiontitle="Stageanchor="sec-7.5.1" numbered="true" toc="default"> <name>Stage 0: Measurement of Latencies andJitter" anchor="section-7.5.1"><t>Jitter</name> <t> The Stage 0MUST<bcp14>MUST</bcp14> start with a synchronization message exchange initiated with the client's READY message.</t><figure><artwork><![CDATA[<artwork name="" type="" align="left" alt=""><![CDATA[ Clientrequest,Request, READY message: ========================= READY q4s://www.example.com Q4S/1.0 Stage: 0 Session-Id: 53655765 User-Agent: q4s-ua-experimental-1.0 Content-Length: 0 ========================= Server Response: ========================= Q4S/1.0 200 OK Session-Id: 53655765 Stage:0 Content-Length: 0 ========================= ]]></artwork></figure><t> This triggers the exchange of a sequence of PING requests and responses that will lead to the calculation of RTT (latency),jitterjitter, and packet loss.</t> <t> After receiving a 200 OK, the client must send the first PINGmessagemessage, and the server will wait to send PINGs until the reception of this first client PING.</t> <t>ClientThe client and serverMUST<bcp14>MUST</bcp14> send PING requests to each other. The Sequence-Number header field of the first PINGMUST<bcp14>MUST</bcp14> be set to 0.ClientThe client and server will manage their own sequence numbers.</t> <figuretitle="Simultaneous exchangeanchor="ref-simultaneous-exchange-of-ping-request-and-responses"> <name>Simultaneous Exchange of PINGrequestRequest andresponses" anchor="ref-simultaneous-exchange-of-ping-request-and-responses"><artwork><![CDATA[Responses</name> <artwork name="" type="" align="left" alt=""><![CDATA[ +------------------------------------------------+ | | | Client Server | | | | --------- Q4S READY 0 ---------> | | <-------- Q4S 200 OK ----------- | | | | --------- Q4S PING ------------> | | <-------- Q4S 200 OK ----------- | | <-------- Q4S PING ------------- | | -------- Q4S 200 OK ----------> | | --------- Q4S PING ------------> | | <-------- Q4S PING ------------- | | --------- Q4S 200 OK ----------> | | <-------- Q4S 200 OK ----------- | | ... | | | +------------------------------------------------+ ]]></artwork> </figure> <t>Figure 8 showsThe following is an example of the PING request sent from the client and the server's response:</t><figure><artwork><![CDATA[<artwork name="" type="" align="left" alt=""><![CDATA[ Client Request: ========================= PING q4s://www.example.com Q4S/1.0 Session-Id: 53655765 Sequence-Number: 0 User-Agent: q4s-ua-experimental-1.0 Measurements: l=22, j=12, pl=0.20, bw= Content-Length: 0 ========================= Server Response: ========================= Q4S/1.0 200 OK Session-Id: 53655765 Sequence-Number: 0 Content-Length: 0 ========================= ]]></artwork></figure><t> The function of the PING method is similar to the ICMP echo requestmessage.message <xref target="RFC0792" format="default"/>. The serverMUST<bcp14>MUST</bcp14> answer as soon as it receives the message.</t> <t> Both endpointsMUST<bcp14>MUST</bcp14> send Q4S PING messages with the periodicity specified in the first parameter of SDPmeasurement procedure"measurement:procedure" attribute,usingalways using the same UDP ports and incrementing the Sequence-Number with each message.</t> <t> In the following example, theSDP measurement procedure attribute, thisvalue of the first parameter of the SDP "measurement:procedure" attribute is 50 milliseconds (from the client to the server) and60ms60 ms (from the server to theclient).</t> <figure><artwork><![CDATA[client):</t> <sourcecode type="sdp"><![CDATA[ a=measurement:procedure default(50/60,50/50,5000,256/256,256/256)]]></artwork> </figure>]]></sourcecode> <t> TheyMUST NOT<bcp14>MUST NOT</bcp14> wait for a response to send the next PING request. The"Sequence-Number"Sequence-Number header field value is incremented sequentially andMUST<bcp14>MUST</bcp14> start at zero. If this stage is repeated, the initial Sequence-NumberMUST<bcp14>MUST</bcp14> start at zero again.</t> <t> All PING requestsMUST<bcp14>MUST</bcp14> contain a"Measurements" header,Measurements header field with the values of the latency,jitterjitter, and packet loss measured by each entity up to that moment. The client will send its measurements to theserverserver, and the serverhiswill send its measurements to the client. Example:</t><figure><artwork><![CDATA[<artwork name="" type="" align="left" alt=""><![CDATA[ Measurements: l=22, j=13, pl=0.10,bw=</t>bw= ]]></artwork></figure><t> Wherel"l" stands for latency,j"j" for jitter,pl"pl" forpacketlosspacket loss, andbw"bw" for bandwidth. The bandwidth value is omitted, as it is not measured at this stage.</t> <t> Optionally the PING request can include a"Timestamp" header,Timestamp header field with the time in which the message has been sent. In the case that the header field is present, the serverMUST<bcp14>MUST</bcp14> include the header field in the response without changing the value.</t> <t> A minimum number of PING messagesMUST<bcp14>MUST</bcp14> be exchanged in order to be able to measure latency,jitterjitter, andpacket-losspacket loss with certain accuracy (at least 256 samples areRECOMMENDED<bcp14>RECOMMENDED</bcp14> to getaan accurate packet loss measurement). Both the client and the server calculate the respective measured parameter values. The mechanisms to calculate the different parameters are described insection 7.3.</t><xref target="sec-7.3" format="default"/>.</t> <t> At the end of this stage 0, there are three possibilities:</t><t><list style="symbols"><t>The<ul spacing="normal"> <li>The latency,jitterjitter, andpacket losspacketloss constraints are reached in bothdirections</t> <t>Thedirections</li> <li>The latency,jitterjitter, andpacket losspacketloss constraints are not reached in one or bothdirections</t> </list> </t>directions</li> </ul> <t> In the first case, Stage 0 is finished.ClientThe client and server are ready for Stage 1: bandwidth and packet loss measurement. The client moves to stage 1 by sending a READY messageincludingthat includes the header field, "Stage: 1".</t> <t> If the bandwidth constraints are either empty orwithhave a value of zero, thenegotiationNegotiation phaseMUST terminate<bcp14>MUST</bcp14> terminate, and both client and server may initiate the ContinuityPhase.phase. In thiscasecase, client moves to the Continuity phase by sending a READY messageincludingthat includes the header field, "Stage: 2".</t> <t> The second case, in which one or more quality constraints have not been met, is detailed insection 7.5.4.</t><xref target="sec-7.5.4" format="default"/>.</t> </section> <sectiontitle="Stageanchor="sec-7.5.2" numbered="true" toc="default"> <name>Stage 1: Measurement of Bandwidth and PacketLoss" anchor="section-7.5.2"><t>Loss</name> <t> This stage begins in a similar way to stage 0, sending a READY request over TCP.ThisThe value of the READYmessage "Stage"message's Stage headervaluefield is 1. The server answers with a Q4S 200 OK message to synchronize the initiation of the measurements as shown inFigure 9.</t><xref target="ref-starting-bandwidth-and-packet-loss-measurement" format="default"/>.</t> <figuretitle="Starting bandwidthanchor="ref-starting-bandwidth-and-packet-loss-measurement"> <name>Starting Bandwidth andpacket loss measurement" anchor="ref-starting-bandwidth-and-packet-loss-measurement"><artwork><![CDATA[Packet Loss Measurement</name> <artwork name="" type="" align="left" alt=""><![CDATA[ +------------------------------------------------+ | | | Client Server | | | | --------- Q4S READY 1 -----------> | | <-------- Q4S 200 OK ------------- | | | | --------- Q4S BWIDTH -----------> | | <-------- Q4S BWIDTH ------------ | | --------- Q4S BWIDTH -----------> | | <-------- Q4S BWIDTH ------------ | | ... | | | +------------------------------------------------+ ]]></artwork> </figure><figure><artwork><![CDATA[<artwork name="" type="" align="left" alt=""><![CDATA[ Client Request: ========================= READY q4s://www.example.com Q4S/1.0 User-Agent: q4s-ua-experimental-1.0 Stage: 1 Session-Id: 53655765 Content-Length: 0 ========================= Server Response: ========================= Q4S/1.0 200 OK Session-Id: 53655765 Stage: 1 Content-Length: 0 ========================= ]]></artwork></figure><t> Just after receiving the 200 OK, both the client and the serverMUST<bcp14>MUST</bcp14> start sending BWIDTH messages simultaneously using the UDPq4s"q4s" ports. <xreftarget="section-7.3.3"/>target="sec-7.3.3" format="default"/> describes the bandwidth measurement in detail.</t> <t> At the end of this stage 1, there are three possibilities:</t><t><list style="symbols"><t>The<ul spacing="normal"> <li>The bandwidth andpacket losspacketloss constraints are reached in bothdirections</t> <t>Thedirections.</li> <li>The bandwidth andpacket losspacketloss constraints are not reached in one or bothdirections.</t> </list> </t>directions.</li> </ul> <t> In the first case, Stage 1 is finished.ClientThe client and server are ready for the Continuity phase. The client moves to this phase by sending a READY messageincludingthat includes the header field, "Stage: 2". The server answerMUST<bcp14>MUST</bcp14> be 200 OK as shown inFigure 10.</t><xref target="ref-trigger-the-application-using-http-uri" format="default"/>.</t> <figuretitle="Triggeranchor="ref-trigger-the-application-using-http-uri"> <name>Trigger theapplication usingApplication Using HTTPURI" anchor="ref-trigger-the-application-using-http-uri"><artwork><![CDATA[URI</name> <artwork name="" type="" align="left" alt=""><![CDATA[ +------------------------------------------------+ | | | Client Server | | | | --------- Q4S READY 2 --------------> | | <---- Q4S 200 OK with trigger URI----- | | | | --------- HTTP GET ----------------> | | | | (Application starts) | | | +------------------------------------------------+ ]]></artwork> </figure><figure><artwork><![CDATA[<artwork name="" type="" align="left" alt=""><![CDATA[ Client Request: ========================= READY q4s://www.example.com Q4S/1.0 User-Agent: q4s-ua-experimental-1.0 Stage: 2 Session-Id: 53655765 Content-Length: 0 ========================= Server Answer: ========================= Q4S/1.0 200 OK Date: Mon, 10 Jun 2010 10:00:01 GMT Session-Id: 53655765 Trigger-URI: http://www.example.com/app_start Expires: 3000 Content-Type: application/sdp Signature: 6ec1ba40e2adf2d783de530ae254acd4f3477ac4 Content-Length: 131 (SDP not shown) ========================= ]]></artwork></figure><t> If the"Trigger-URI"Trigger-URI header field is present, the clientSHOULD<bcp14>SHOULD</bcp14> send an HTTP request to this URI.</t> <t> The second case, with violated networkconstraintsconstraints, is explained in7.5.4.</t><xref target="sec-7.5.4" format="default"/>.</t> </section> <sectiontitle="Qualityanchor="sec-7.5.3" numbered="true" toc="default"> <name>Quality Constraints NotReached" anchor="section-7.5.3"><t>Reached</name> <t> After finishing Stage 1 of the Negotiation phase, the client and the server have eachotherother's measured parameter values as these have been exchanged in the"Measurements" headersMeasurements header fields of the PING and BWIDTH messages. If there is one or more parameters that do not comply with the uplink or downlink application constraintsrequiredrequired, both the server and the client are aware of it.</t> <t> If there is any quality parameter that does not meet the uplink or downlink quality constraints specified in the SDP message, two scenarios are possible depending on the specifiedalerting-modealerting mode (if not present, the default value is"Reactive"Reactive alerting mode):<list style="format (%c)"></t> <ol spacing="normal" type="(%c)"> <li> <t> Q4S-aware-network alertingmode:">mode: the serverMUST<bcp14>MUST</bcp14> send a Q4S-ALERT message to the client including the digitalsignature header,Signature header field, and the clientMUST<bcp14>MUST</bcp14> answer with the same Q4S-ALERT message. The Signature header field contains the signed hash value of the SDP body in order to protect all the SDPthe datadata, and therefore itMUST<bcp14>MUST</bcp14> contain themeasurement"measurement" parameters in the body.<figure><artwork><![CDATA[</t> <artwork name="" type="" align="left" alt=""><![CDATA[ Server request ========================= Q4S-ALERT q4s://www.example.com Q4S/1.0 Host: www.example.com User-Agent: q4s-ua-experimental-1.0 Session-Id: 53655765 Content-Type: application/sdp Content-Length: 142 v=0 o=q4s-UA 53655765 2353687637 IN IP4 192.0.2.33 s=Q4S i=Q4S parameters t=0 0 a=qos-level:1/2 a=alerting-mode: Q4S-aware-network a=alert-pause:5000 a=public-address:client IP4 198.51.100.51 a=public-address:server IP4 198.51.100.58 a=latency:40 a=jitter:10/10 a=bandwidth:20/6000 a=packetloss:0.50/0.50 a=flow:app downlink TCP/10000-20000 a=flow:app uplink TCP/56000 a=flow:q4s downlink UDP/55000 a=flow:q4s downlink TCP/55001 a=flow:q4s uplink UDP/56000 a=flow:q4s uplink TCP/56001 a=measurement:procedure default(50/50,50/50,5000,256/256,256/256) a=measurement:latency 30 a=measurement:jitter 6/4 a=measurement:bandwidth 200/4000 a=measurement:packetloss 0.20/0.33 ========================= ]]></artwork></figure><t> At this point, both the client and server keep on measuring but without sending newQ4S ALERTQ4S-ALERT messages during the "alert-pause" milliseconds. </t><t></li> <li> Reactive alertingmode:">mode: the server stackMUST<bcp14>MUST</bcp14> send an alert notification to the Actuator, and the ActuatorMUST<bcp14>MUST</bcp14> answer with an acknowledgement to the received alert notification. The alert notification sent to the Actuator by the server stack doesn't follow Q4S message style but should have all the information the Actuator will need for the actions to be taken, which will be implementation dependent.</t> </list> </t></li> </ol> <t> At thispoint,point during Negotiation phase, both the client and server keep on measuring without sending new alert notifications to the Actuator during the "alert-pause" milliseconds specified in the SDP. This way, both client and server will detect any improvement in network conditions as soon as the network reacts. The application can start as soon as the number of measurements indicated in themeasurement procedure"measurement:procedure" attribute indicates that the quality parameters are met.</t> <t>SameThe same applies to Continuity phase: the measurement dialog between client and server must not be interrupted by any possible ALERT message.</t> <sectiontitle="Actuator Role" anchor="section-7.5.3.1"><t> Actuatoranchor="sec-7.5.3.1" numbered="true" toc="default"> <name>Actuator Role</name> <t> The actuator receives notifications of unmet requirements from the Q4S serverstack,stack andactacts upon the application or the network policy server, according to logic out of scope of this protocol.</t> <t> The Actuator logic activates mechanisms at the application levelor/andand/or the network level based on a quality level dictionary, in which the meaning of each levelmeaningis implementationdependentdependent, and each levelinvolveinvolves different actions based on rules to keep a certain user experience quality.</t> <t> The type of actions that an Actuator can take at the application level are application dependent andMAY<bcp14>MAY</bcp14> involve:</t><t><list style="symbols"><t>Reduction<ul spacing="normal"> <li>Reduction of application functionalities, such as limitation of application speed or applicationoptions.</t> <t>Reductionoptions.</li> <li>Reduction of application resources usage, such as reduction of frames per second in a videoappapplication or any other parameter modification in order to adapt to networkconditions.</t> </list> </t>conditions.</li> </ul> <t> Apart from actions at the application level, the ActuatorMAY<bcp14>MAY</bcp14> act at the network level if a network policy server is available.</t> </section> <sectiontitle="Policyanchor="sec-7.5.3.2" numbered="true" toc="default"> <name>Policy ServerRole" anchor="section-7.5.3.2"><t>Role</name> <t> A network policy server may be part of thereactive scenarioReactive scenario, and it is in charge of managing network quality provision.NetworkA network policy server may implement all or some of these features (but implementation is not exclusive to):</t><t><list style="symbols"><t>Server<ul spacing="normal"> <li>Server validation in terms of qualityconstraints.</t> <t>Authenticationconstraints</li> <li>Authentication (Signature validation) and security(block(blocking of maliciousclients)</t>clients)</li> <li> <t>Policy rules(following(the following rules are onlyexamples):<list style="symbols"><t>Maximumexamples):</t> <ul spacing="normal"> <li>Maximum quality level allowed for theACP</t> <t>TimeACP</li> <li>Time bands allowed for providing qualitysessions</t> <t>Numbersessions</li> <li>Number of simultaneous quality sessionsallowed</t> <t>Maximumallowed</li> <li>Maximum time used by allowed qualitysessions</t> <t>Etc.</t> </list> </t> </list> </t>sessions</li> <li>Etc.</li> </ul> </li> </ul> <t> If any of the policy rules fail, a Q4S-ALERT messageMUST<bcp14>MUST</bcp14> be answered by a6XX error,6xx error indicating the cause.</t> </section> </section> <sectiontitle="QoS Level Changes" anchor="section-7.5.4"><t>anchor="sec-7.5.4" numbered="true" toc="default"> <name>"qos-level" Changes</name> <t> If any constraint was violated, the serverMAY<bcp14>MAY</bcp14> trigger a Q4S-ALERT asking for a higherqos-level"qos-level" attribute. The maximumqos-level"qos-level" allowed is9,9 for both uplink and downlink.</t> <t> If theqos-level"qos-level" has reached the maximum value for the downlink or uplink without matching the constraints, then a CANCEL requestMUST<bcp14>MUST</bcp14> be sent by the client using the TCP port determined in thehandshakeHandshake phase in order to release the session. In reaction to the reception of the CANCEL request, the serverMUST<bcp14>MUST</bcp14> send a CANCELrequestrequest, too. If no CANCEL request is received, the expiration time cancels the sessionaton the server side.</t><figure><artwork><![CDATA[<artwork name="" type="" align="left" alt=""><![CDATA[ Client Request: ========================= CANCEL q4s://www.example.com Q4S/1.0 User-Agent: q4s-ua-experimental-1.0 Session-Id: 53655765 Content-Type: application/sdp Content-Length: 142 (SDP not shown) ========================= Server Request in reaction to Client Request: ========================= CANCEL q4s://www.example.com Q4S/1.0 Session-Id: 53655765 Expires: 0 Content-Type: application/sdp Signature: 6ec1ba40e2adf2d783de530ae254acd4f3477ac4 Content-Length: 131 (SDP not shown) ========================= ]]></artwork></figure></section> </section> <sectiontitle="Continuity Phase" anchor="section-7.6"><t>anchor="sec-7.6" numbered="true" toc="default"> <name>Continuity Phase</name> <t> During thenegotiationNegotiation phase, latency, jitter,bandwidthbandwidth, and packet loss have been measured. During thecontinuity phaseContinuity phase, bandwidth will not be measured again because bandwidth measurements may disturb application performance.</t> <t> This phase is supposed to be executed at the same time as the real-time application is being used.</t> <t> This document only covers thedefault"default" procedure. The continuity operation withdefaultthe "default" procedure is based on a sliding window of samples. The number of samples involved in the sliding window may be different for jitter and latency than forpacket-losspacket loss calculations according to the fifth and sixth parameters of themeasurement procedure"measurement:procedure" attribute. In the example, shown inFigure 11,<xref target="ref-sliding-samples-window" format="default"/>, the jitter and latency sliding window comprises 40samplessamples, whereas the size of thepacket-losspacket loss sliding window is 100 samples:</t><figure><artwork><![CDATA[<sourcecode type="sdp"><![CDATA[ a=measurement:procedure default(50/50,75/75,5000,40/40,100/100)]]></artwork> </figure>]]></sourcecode> <t> In addition, the sizes of these windows are configurable per direction: uplink and downlink values may differ.</t> <t> PING requests are sent continuously (in bothdirections)directions), and when the Sequence-Number header field reaches the maximum value, the client continues sending PING messages with the Sequence-Number header field starting again at zero. When the server PING Sequence-Number header field reaches the maximum value, it does the same, starting again from zero.</t> <t> On the client side, the measured values of downlink jitter, downlink packetlossloss, and latency are calculated using the last samples, discarding older ones, in a sliding window schema.</t> <figuretitle="Sliding samples window" anchor="ref-sliding-samples-window"><artwork><![CDATA[anchor="ref-sliding-samples-window"> <name>Sliding Samples Window</name> <artwork name="" type="" align="left" alt=""><![CDATA[ +--------------------------------------------------+ | | | 55 56 57 . . . 253 254 255 0 1 2 . . . 55 56 | | A A | | | | | | +-----------------------------------+ | | | +--------------------------------------------------+ ]]></artwork> </figure> <t> Only if the server detects that the measured values (downlink or uplink jitter, packetlossloss, or latency) are not reaching the quality constraints, aQ4S ALERTQ4S-ALERT is triggered and sent either to the client or to the Actuator, depending on the alerting mode, and thealert-pause"alert-pause" timer is started.</t> <t> In the Q4S-aware-network alerting mode shown inFigure 12,<xref target="ref-continuity-in-q4s-aware-network-alerting-mode" format="default"/>, if the client receives aQ4S ALERTQ4S-ALERT message, itMUST<bcp14>MUST</bcp14> answer by sending theQ4S ALERTQ4S-ALERT request messageback to the serverincluding the SDP (with its corresponding digitalsignature).</t>signature) back to the server.</t> <t> Both client and server will keep performingmeasurementsmeasurements, butno other Q4S ALERT message MUSTQ4S-ALERT messages <bcp14>MUST NOT</bcp14> be sent during "alert-pause" milliseconds. The operations needed to act on the network and the agents in charge of them are out of scope of thisdraft.</t>document.</t> <figuretitle="Continuityanchor="ref-continuity-in-q4s-aware-network-alerting-mode"> <name>Continuity inQ4S-aware-network alerting mode" anchor="ref-continuity-in-q4s-aware-network-alerting-mode"><artwork><![CDATA[Q4S-Aware-Network Alerting Mode</name> <artwork name="" type="" align="left" alt=""><![CDATA[ +------------------------------------------------+ | | | Client Server | | | | ... | | ----------- PING ----------> | | <--------- 200 OK ---------- | | <------- Q4S-ALERT --------- | | -------- Q4S-ALERT --------> | | <---------- PING ----------- | | ---------- 200 OK ---------> | | ----------- PING ----------> | | <--------- 200 OK ---------- | | <---------- PING ----------- | | ---------- 200 OK ---------> | | ... | | | +------------------------------------------------+ ]]></artwork> </figure> <t> In the Reactive scenario shown inFigure 13,<xref target="ref-continuity-in-reactive-alerting-mode" format="default"/>, if the server detects that the measured values (downlink or uplink jitter, packetlossloss, or latency) are not reaching the quality constraints, an alert notification is triggered and sent to the Actuator. The ActuatorMUST<bcp14>MUST</bcp14> then answer to the server stack with an alertacknowledgement</t>acknowledgement.</t> <t> The measurement dialog between the client and the serverMUST NOT<bcp14>MUST NOT</bcp14> be interrupted by any possible ALERT message.</t> <figuretitle="Continuityanchor="ref-continuity-in-reactive-alerting-mode"> <name>Continuity in Reactivealerting mode" anchor="ref-continuity-in-reactive-alerting-mode"><artwork><![CDATA[Alerting Mode</name> <artwork name="" type="" align="left" alt=""><![CDATA[ +------------------------------------------------+ | | | Client Server Actuator | | ... | | --- PING ----------> | | <-- 200 OK---------- | | <----- PING -------- | | <--- 200 OK -------- ---- alert | | notification --> | | | | --- PING ----------> <--- alert | | acknowledge --- | | <-- 200 OK---------- | | <----- PING -------- | | --- 200 OK --------> | | ... | | | +------------------------------------------------+ ]]></artwork> </figure> </section> <sectiontitle="Termination Phase" anchor="section-7.7"><t>anchor="sec-7.7" numbered="true" toc="default"> <name>Termination Phase</name> <t> The Termination phase is theend pointendpoint for the established Q4S session that is reached in the following cases:</t><t><list style="symbols"><t>A<ul spacing="normal"> <li>A CANCEL message has been received. The client sends a CANCEL message due to theimpossibility of the networknetwork's inability to meet the required quality constraints. The client and server application will be notified bythetheir respective Q4Sstack.</t> <t>Sessionstacks.</li> <li>Session expires: if after the Expirestimetime, no client or server activity is detected, that end cancels thesession.</t> <t>Asession.</li> <li>A BEGIN message has been received by the server. The pre-existing Q4S quality session iscancelledcanceled, and a new session will beinitiated.</t> </list> </t>initiated.</li> </ul> <t> The meaning of the Termination phase in terms of the release of resources or accounting is application dependent and out of scope of the Q4S protocol.</t> <t> In the Reactive alerting mode, Q4S CANCEL messages received by the Q4S server must cause thesending of cancel notifications sent from theserver stack to send cancel notifications to the Actuator in order to release possible assigned resources for the session.</t> <sectiontitle="Sanityanchor="sec-7.7.1" numbered="true" toc="default"> <name>Sanity Check of QualitySessions" anchor="section-7.7.1"><t>Sessions</name> <t> A session may finish due to several reasons (client shutdown, client CANCEL request, constraints not reached,etc),etc.), and any session finishedMUST<bcp14>MUST</bcp14> release the assigned resources.</t> <t> In order to release the assigned server resources for the session, the"Expires"Expires header field indicates the maximum interval of time without exchanging any Q4S message.</t> </section> </section> <sectiontitle="Dynamicanchor="sec-7.8" numbered="true" toc="default"> <name>Dynamic ConstraintsAnd Flows" anchor="section-7.8"><t>and Flows</name> <t> Depending on the nature of the application, the quality constraints to be reached may evolve, changing some or all quality constraint values in any direction.</t> <t> The clientMUST<bcp14>MUST</bcp14> be able to deal with this possibility. When the server sends an SDP document attached to a response (200OK,OK or Q4S-ALERT,etc),etc.), the clientMUST<bcp14>MUST</bcp14> take all the new received values, overriding any previous value in use.</t> <t> The dynamic changes on the quality constraints can beasa result of two possibilities:</t><t><list style="symbols"><t>The<ul spacing="normal"> <li>The application communicates to the Q4S server a change in the constraints. In thiscasecase, the application requirements canevolveevolve, and the Q4S server will be aware ofthem.</t> <t>Thethem.</li> <li>The application uses TCP flows. In that case, in order to guarantee a constant throughput, the nature of TCP behavior forces the use of a composite constraint function, which depends on RTT, packetlossloss, and a window control mechanism implemented in each TCPstack.</t> </list> </t>stack.</li> </ul> <t> TCP throughput can be less than actual bandwidth if the Bandwidth-Delay Product (BDP) islargelarge, or if the network suffers from a high packet loss rate. In both cases, TCP congestion control algorithms may result in a suboptimal performance.</t> <t> Different TCP congestion control implementations like Reno <xreftarget="ref-23" />,target="RENO" format="default"/>, High Speed TCP(RFC 3649<xreftarget="ref-24"/>),target="RFC3649" format="default"/>, CUBIC <xreftarget="ref-25"/>,target="I-D.rhee-tcpm-cubic" format="default"/>, Compound TCP(CTCP(CTCP) <xreftarget="ref-26"/>), etc.target="I-D.sridharan-tcpm-ctcp" format="default"/>, etc., reach different throughputs under the same network conditions of RTT and packet loss. In all cases, depending on theRTT measuredRTT-measured value, the Q4S server couldchangedynamically change the packetloss constraints (defined in the SDP) in order to make it possible to reach a required throughput or vice versa(use packetloss measurement(using "measurement:packetloss" to change dynamically the latency constraints).</t> <t> A general guidelineto calculatefor calculating thepacketlosspacket loss constraint and the RTT constraint consistsinof approximating the throughput by using a simplified formula, which should take into account the TCP stack implementation of the receiver, in addition to the RTT and packet loss:</t><figure><artwork><![CDATA[<sourcecode type="pseudocode"><![CDATA[ Th= Function( RTT, packet loss, ...)]]></artwork> </figure>]]></sourcecode> <t> Then, depending onRTT measuredRTT-measured values, set dynamically thepacketlosspacket loss constraint.</t> <t> It is possible to easily calculate a worst-case boundary for the Reno algorithm, which should ensure for all algorithms that the target throughput is actuallyachieved. Except that,achieved, except that high-speed algorithms will then have evenalargerthroughput,throughput if more bandwidth is available.</t> <t> For the Reno algorithm, theMathis'Mathis formula may be used <xreftarget="ref-23"/>target="RENO" format="default"/> for the upper bound on the throughput:</t><t><list hangIndent="9" style="hanging"><t><sourcecode type="pseudocode"><![CDATA[ Th<=<= (MSS/RTT)*(1 /sqrt{p})</t> </list> </t>sqrt{p}) ]]></sourcecode> <t> In the absence of packet loss, a practical limit for the TCP throughput is the receiver_window_size divided by theround-trip time.RTT. However, if the TCP implementation uses a window scale option, this limit can reach the available bandwidth value.</t> </section> <sectiontitle="Qos-levelanchor="sec-7.9" numbered="true" toc="default"> <name>"qos-level" UpgradeAndand DowngradeOperation" anchor="section-7.9"><t>Operation</name> <t> Each time the server detects a violation of constraints, the alert mechanism is triggered, thealert-pause"alert-pause" timer is started, and theqos-level"qos-level" is increased. When this happens repeatedly, and theqos-level"qos-level" reaches its maximum value (value 9), the session iscancelled.canceled. But when the violation of constraints stops before reachingqos-level"qos-level" maximum value, the recovery mechanism allows for theqos-level"qos-level" upgrade gradually.</t> <t>Following, thisThis downgrade and upgrade ofqos-level"qos-level" is explained withanthe following example:</t><t><list style="numbers"><t>A<ol spacing="normal" type="1"> <li>A Q4S session is initiated successfully withqos-level=0.</t> <t>During"qos-level=0".</li> <li>During thecontinuityContinuity phase, violation of constraints is detected;qos-levelthe "qos-level" is increased to 1, a Q4S-ALERT is sent by the server to theclientclient, andalert-pausean "alert-pause" timer isstarted.</t> <t>Alert-pausestarted.</li> <li>The "alert-pause" timerexpiresexpires, and still a violation of constraints is detected;qos-levelthe "qos-level" is increased to 2, a Q4S-ALERT is sent by the server to theclientclient, andalert-pausean "alert-pause" timer isstarted.</t> <t>Alert-pausestarted.</li> <li>The "alert-pause" timerexpiresexpires, but the violation of constraints has stopped;recovery-pausethe "recovery-pause" timer isstarted.</t> <t>Recovery-pausestarted.</li> <li>The "recovery-pause" timer expires, and no violation of constraints has beendetected meanwhile; qos-leveldetected. Meanwhile, the "qos-level" is decreased to 1, a Q4S-RECOVERY is sent by the server to theclientclient, andrecovery-pausethe "recovery-pause" timer is startedagain.</t> <t>Recovery-pauseagain.</li> <li>The "recovery-pause" timer expiresagainagain, and no violation of constraints has beendetected meanwhile; qos-leveldetected. Meanwhile, the "qos-level" is decreased to00, and a Q4S-RECOVERY is sent by the server to theclient; recovery-pauseclient. The "recovery-pause" timer is not started this time asqos-levelthe "qos-level" has reached its initialvalue.</t> </list> </t>value.</li> </ol> <t> When the network configuration allows for the possibility of managing Q4S flows and application flows independently (either is a network-based QoS or aQ4S awareQ4S-aware network), theqos-level"qos-level" downgrade process could be managed more efficiently using a strategy that allows for carrying outqos-level"qos-level" downgrades excludingappapplication flows from SDP dynamically. The Q4S flows would be downgraded to allow for measurements on a lower quality level without interference of the application flows. A Q4S clientMUST<bcp14>MUST</bcp14> allow this kind of SDPmodificationsmodification by the server.</t> <t> Periodically (every several minutes, depending on the implementation) a Q4S-ALERT could be triggered, in which the level is downgraded for Q4S flows, excluding application flows from the embedded SDP of that request.</t> <t> This mechanism allowsto measurethe measurement at lower levels of quality while application flows continue using a higherqos level"qos-level" value.</t><t><list style="symbols"><t>If<ul spacing="normal"> <li>If the measurements in the lower level meet the quality constraints, then a Q4S-RECOVERY message to this lowerqos-level"qos-level" may be triggered, in which the SDP includes the application flows in addition to the Q4Sflows.</t> <t>Ifflows.</li> <li>If the measurements in the lower level do not meet the constraints, then a new Q4S-ALERT to the previousqos-level MUST"qos-level" <bcp14>MUST</bcp14> be triggered, in which the SDP includes only the Q4Sflows.</t> </list> </t>flows.</li> </ul> <figuretitle="Possible evolution of qos-level" anchor="ref-possible-evolution-of-qos-level"><artwork><![CDATA[anchor="ref-possible-evolution-of-qos-level"> <name>Possible Evolution of "qos-level"</name> <artwork name="" type="" align="left" alt=""><![CDATA[ +------------------------------------------------+ | | | qos-level | | A | | | | | 4| | | | | | 3| +------+ | | | | | | | 2| +----+ +----+ +--- | | | | | | | | 1| +----+ +-----+ | | | | | | 0+---+---------------------------------> time | | | +------------------------------------------------+ ]]></artwork> </figure> <t> This mechanism, illustrated inFigure 14,<xref target="ref-possible-evolution-of-qos-level" format="default"/>, avoids the risk of disturbing theapplication,application while the measurements are being run in lower levels. However, this optional optimization of resourcesMUST<bcp14>MUST</bcp14> be used carefully.</t> <t> The chosen period to measure a lowerqos level"qos-level" is implementation dependent. Therefore, it is not included as ameasurement procedure"measurement:procedure" parameter. It isRECOMMENDED<bcp14>RECOMMENDED</bcp14> to use a large value, such as 20 minutes.</t> </section> </section> <sectiontitle="Generalanchor="sec-8" numbered="true" toc="default"> <name>General User AgentBehavior" anchor="section-8"><section title="RolesBehavior</name> <section anchor="sec-8.1" numbered="true" toc="default"> <name>Roles in Peer-to-PeerScenarios" anchor="section-8.1"><t>Scenarios</name> <t> In order to allowpeer to peerpeer-to-peer applications, a Q4S User Agent (UA)MUST<bcp14>MUST</bcp14> be able to assume both the client and server role. The role assumed depends on who sends the first message.</t> <t> In a communication between two UAs, the UA that first sends the Q4S BEGIN requestin the first place, for startingto start thehandshake phase,Handshake phase shall assume the client role.</t> <t> If bothUASsUAs send the BEGIN request at the same time, they will wait for a random time to restart again as shown inFigure 15.</t><xref target="ref-p2p-roles" format="default"/>.</t> <t> Otherwise, an UA may be configured to act only as server (e.g., content provider's side).</t> <figuretitle="P2P roles." anchor="ref-p2p-roles."><artwork><![CDATA[anchor="ref-p2p-roles"> <name>P2P Roles</name> <artwork name="" type="" align="left" alt=""><![CDATA[ +-----------------------------------------------+ | | | UA(Client) UA(Server) | | | | -------- Q4S BEGIN -------------> | | <------- Q4S BEGIN -------------- | | | | ------- Q4S BEGIN --------------> | | <------ Q4S 200 OK -------------- | | | | | +-----------------------------------------------+ ]]></artwork> </figure> </section> <sectiontitle="Multipleanchor="sec-8.2" numbered="true" toc="default"> <name>Multiple Quality Sessions inParallel" anchor="section-8.2"><t>Parallel</name> <t> A Q4S session is intended to be used for an application.ItThis means that for using the application, the clientMUST<bcp14>MUST</bcp14> establish only one Q4S session against the server. Indeed, the relation betweensession-idthe Session-Id and the application is 1 to 1.</t> <t> If a user wants to participate in several independent Q4S sessions simultaneously against different servers (or against the sameserver)server), it can execute different Q4S clients to establish separately different Q4Ssessionssessions, but it isNOT RECOMMENDED,<bcp14>NOT RECOMMENDED</bcp14> because:</t><t><list style="symbols"><t>The<ul spacing="normal"> <li>The establishment of a new Q4S session may affect other running applications over other Q4S sessions during bandwidthmeasurement.</t> <t>Ifmeasurement.</li> <li>If thenegotiationNegotiation phase is executed separately before running any application, the summation of bandwidth requirements could not be met when the applications are running inparallel.</t> </list> </t>parallel.</li> </ul> </section> <sectiontitle="Generalanchor="sec-8.3" numbered="true" toc="default"> <name>General ClientbBhavior" anchor="section-8.3"><t>Behavior</name> <t> A Q4SClientclient has different behaviors. We will use lettersX,Y,ZX, Y, and Z to designate each different behavior (follow theletter bulletsletters infigure 16).</t> <t><list hangIndent="3" style="hanging"> <t hangText="X)">When<xref target="ref-phases-client-behaviors" format="default"/> and their descriptions below).</t> <dl newline="false" spacing="normal" indent="4"> <dt>X)</dt> <dd>When it sends messages over TCP (methods BEGIN, READY, Q4S-ALERT,Q4S-RECOVERYQ4S-RECOVERY, andCANCEL)CANCEL), it behaves strictly like a state machine that sends requests and waits for responses. Depending on the responsetypetype, it entersininto a newstate.</t> </list> </t>state.</dd> </dl> <t> When it sends UDP messages (methods PING and BWIDTH), a Q4S client is not strictly a state machine that sends messages and waits for responsesbecause:</t> <t><list hangIndent="3" style="hanging"> <t hangText="Y)">Atbecause of the following:</t> <dl newline="false" spacing="normal" indent="4"> <dt>Y)</dt> <dd>During the measurement of latency,jitterjitter, and packetloss measurement,loss, the PING requests are sent periodically, not just after receiving the response to the previous request. In addition, the clientMUST<bcp14>MUST</bcp14> answer the PING requests coming from the server, therefore the client assumes temporarily the role of aserver.</t> </list> </t> <t><list hangIndent="3" style="hanging"> <t hangText="Z)">Atserver.</dd> </dl> <dl newline="false" spacing="normal" indent="4"> <dt>Z)</dt> <dd>During the bandwidth and packet loss measurement stage, the client does not expect to receive responses when sending BWIDTH requests to the server. In addition, itMUST<bcp14>MUST</bcp14> receive and process all server messages in order to achieve the downlinkmeasurement.</t> </list> </t>measurement.</dd> </dl> <t> The Q4S-ALERT and CANCEL may have a conventional answer if an error is produced, otherwise the corresponding answer is formatted as a request message.</t> <figuretitle="Phases & client behaviors" anchor="ref-phases-client-behaviors"><artwork><![CDATA[anchor="ref-phases-client-behaviors"> <name>Phases and Client Behaviors</name> <artwork name="" type="" align="left" alt=""><![CDATA[ +-----------+------------------------+-----------+-----------+ | Handshake | Negotiation |Continuity |Termination| | Phase | Phase | Phase | Phase | | | | | | | X ---------> Y --> X --> Z --> X ---> Y --> X ---> X | | | A | A | | A | | | | | | | | | | | | | | | | +-----+ +-----+ | +-----+ | | | | | | | +------------------------------------------------+-----------+ ]]></artwork> </figure> <sectiontitle="Generating Requests" anchor="section-8.3.1"><t>anchor="sec-8.3.1" numbered="true" toc="default"> <name>Generating Requests</name> <t> A valid Q4S request formulated by aClient MUST,client <bcp14>MUST</bcp14>, at a minimum,containscontain the following header fields:</t><t><list style="symbols"><t>If<dl> <dt>If no SDP isincluded: theincluded:</dt><dd>the header fields Session-Id and Sequence-Number aremandatory.</t> <t>Ifmandatory.</dd> <dt>If SDP isincluded:included:</dt><dd>the Session-Id is embedded into the SDP, therefore the inclusion of the Session-Id header field isoptionaloptional, but ifpresentpresent, must have the same value. Measurements are embedded into the SDP only for Q4S-ALERT messages in order to besigned.</t> </list> </t>signed.</dd> </dl> <t> At any time, if the server sendsanew SDP with updated values, the clientMUST<bcp14>MUST</bcp14> take it into account.</t> </section> </section> <sectiontitle="Generalanchor="sec-8.4" numbered="true" toc="default"> <name>General ServerBehavior" anchor="section-8.4"><t>Behavior</name> <t> If a server does not understand a header field in a request (that is, the header field is not defined in this specification or in any supported extension), the serverMUST<bcp14>MUST</bcp14> ignore that header field and continue processing the message.</t> <t> The role of the server is changed atnegotiationNegotiation andcontinuityContinuity phases, in which the serverMUST<bcp14>MUST</bcp14> send packets to measure jitter,latencylatency, and bandwidth. Therefore, the different behaviors of the server are (follow theletter bulletsletters inthe figure 17):</t> <t><list hangIndent="3" style="hanging"> <t hangText="R)"><xref target="ref-phases-server-behaviours" format="default"/> and their descriptions below):</t> <dl newline="false" spacing="normal" indent="4"> <dt>R)</dt> <dd> When the client sends messages over TCP (methods BEGIN, READY Q4S-ALERT,Q4S-RECOVERYQ4S-RECOVERY, andCANCEL)CANCEL), it behaves strictly like a state machine that receives messages and sendsresponses.</t> </list> </t>responses.</dd> </dl> <t> When the client begins to send UDP messages (methods PING and BWIDTH), a Q4S server is not strictly a state machine that receives messages and sends responsesbecause:</t> <t><list hangIndent="3" style="hanging"> <t hangText="S)"> Atbecause of the following:</t> <dl newline="false" spacing="normal" indent="4"> <dt>S)</dt> <dd> During the measurement of latency,jitterjitter, and packetloss measurement,loss, the PING requests are sent periodically by the clientbutand also by the server. In thiscasecase, the server behaves as a server answering client requests but also behaves temporarily as a client, sending PING requests toward the client and receivingresponses.</t> </list> </t> <t><list hangIndent="3" style="hanging"> <t hangText="T)"> Atresponses.</dd> </dl> <dl newline="false" spacing="normal" indent="4"> <dt>T)</dt> <dd> During bandwidth and packet loss measurement, the server sends BWIDTH requests to the client. In addition, itMUST<bcp14>MUST</bcp14> receive and process client messages in order to achieve the uplinkmeasurement.</t> </list> </t>measurement.</dd> </dl> <t> The Q4S-ALERT and CANCEL may have a conventional answer if an error is produced, otherwise the corresponding answer is formatted as a request message.</t> <figuretitle="Phases & server behaviours" anchor="ref-phases-server-behaviours"><artwork><![CDATA[anchor="ref-phases-server-behaviours"> <name>Phases and Server Behaviors</name> <artwork name="" type="" align="left" alt=""><![CDATA[ +-----------+------------------------+-----------+-----------+ | Handshake | Negotiation |Continuity |Termination| | Phase | Phase | Phase | Phase | | | | | | | R ---------> S --> R --> T --> R ---> S --> R ---> R | | | A | A | | A | | | | | | | | | | | | | | | | +-----+ +-----+ | +-----+ | | | | | | | +------------------------------------------------+-----------+ ]]></artwork> </figure> </section> </section> <sectiontitle="Implementation Recommendations" anchor="section-9"><section title="Defaultanchor="sec-9" numbered="true" toc="default"> <name>Implementation Recommendations</name> <section anchor="sec-9.1" numbered="true" toc="default"> <name>Default ClientConstraints" anchor="section-9.1"><t>Constraints</name> <t> To provide a default configuration, it would be goodthatif the client had a configurable set ofQualityquality headers in the implementation settings menu.OtherwiseOtherwise, these quality headers will not be present in the first message.</t> <t> Different business models (out of scope of this proposal) may be achieved: depending on who pays for the quality session, the server can accept certainClientclient parameters sent in the first message, or force billing parameters on the server side.</t> </section> <sectiontitle="Latencyanchor="sec-9.2" numbered="true" toc="default"> <name>Latency and JitterMeasurements" anchor="section-9.2"><t>Measurements</name> <t> Different client and server implementations may send a different number of PING messages for measuring, although at least 255 messages should be considered to perform the latency measurement. The Stage 0 measurementsonlymay be considered ended only when neither the client nor server receive new PING messages after an implementation-dependent guard time. Only after, the client can send a "READY 1" message.</t> <t> In execution systems, where the timers are not accurate, a recommended approach consists of including the optional Timestamp header"Timestamp"field in the PING request with the time in which the message has been sent. This allows an accurate measurement of the jitter even with no identical intervals of time between PINGs.</t> </section> <sectiontitle="Bandwidth Measurements" anchor="section-9.3"><t>anchor="sec-9.3" numbered="true" toc="default"> <name>Bandwidth Measurements</name> <t> In programming languages orOperating Systemsoperating systems with limited timers or clock resolution, it is recommended to use an approach based on several intervals to send messages of 1KB (= 8000bits),bits) in order to reach the required bandwidthconsumptionconsumption, using a rate as close as possible to a constant rate.</t> <t> For example, if the resolution is 1 millisecond, and the bandwidth to reach is11Mbps,11 Mbps, a good approach consists of sending: </t><figure><artwork><![CDATA[<artwork name="" type="" align="left" alt=""><![CDATA[ 1 message of 1KB every 1 millisecond + 1 message of 1KB every 3 milliseconds + 1 message of 1KB every 23 milliseconds]]></artwork></figure>]]></artwork> <t> The number of intervals depends on the required bandwidth and accuracy that the programmer wants to achieve.</t> <t> Considering messages of 1KB (= 8000 bits), a general approach to determine these intervals is<list style="format (%d)"> <t>the following: </t> <ol spacing="normal" type="(%d)"> <li> ComputeTargettarget bandwidth / 8000 bits. In the exampleaboveabove, it is11Mbps/800011 Mbps / 8000 = 1375 messages persecond </t> <t>second. </li> <li> Divide the number of messages per second by 1000 to determine the number of messages permillisecond. 1375/1000millisecond: 1375 / 1000 =1'375.1.375. The integer value is the number of messages per millisecond (in this case, one). The pending bandwidth is now 375 messages persecond </t>second. </li> <li> <t> To achieve the 375 messages per second, use asub-multiplesubmultiple of10001000, which must be less than375 <figure><artwork><![CDATA[ 1000/2375: </t> <artwork name="" type="" align="left" alt=""><![CDATA[ 1000 / 2 = 500 > 3751000/31000 / 3 = 333 < 375]]></artwork></figure>]]></artwork> <t> In thiscasecase, a message every 3 ms is suitable. The new pending target bandwidth is 375-333- 333 = 42 messages persecond</t>second.</t> </li> <li> <t> Repeat the same strategy as point3,3 to reach the pending bandwidth. In this case, 23 ms is suitablebecause: <figure><artwork><![CDATA[ 1000/22because of the following: </t> <artwork name="" type="" align="left" alt=""><![CDATA[ 1000 / 22 = 45>42 1000/23> 42 1000 / 23 = 43>42> 42 1000 / 24 = 41.6 < 42]]></artwork></figure> </t> </list> </t>]]></artwork> </li> </ol> <t> We can choose 24msms, but then we need to cover an additional 0.4 messages per second(42-41.6=0.4)(42 - 41.6 = 0.4), and 43 is a number higher than 42 but very close to it.</t> <t> In execution systems where the timers are not accurate, a recommended approach consists of checking at each interval the number of packets that should have been sent at this timestamp since origin and send the needed number of packets in order to reach the required bandwidth.</t> <t> The shorter the packetsareused, the more constantisthe rate of bandwidth measurement. However, this may stress the execution system in charge of receiving and processing packets. As a consequence, some packets may be lost because of stack overflows. To deal with this potential issue, a larger packet isRECOMMENDED<bcp14>RECOMMENDED</bcp14> (2KB ormore)more), taking into account the overhead produced by thechunkschunks' headers.</t> </section> <sectiontitle="Packetanchor="sec-9.4" numbered="true" toc="default"> <name>Packet Loss MeasurementResolution" anchor="section-9.4"><t>Resolution</name> <t> Depending on the application nature and network conditions, a packet loss resolution less than 1% may be needed. In such cases, there is no limit to the number of samples used for this calculation. Atradeofftrade-off between time and resolution should be reached in each case. For example, in order to have a resolution of 1/10000, the last 10000 samples should be considered in the packet loss measured value.</t> <t> The problem of this approach is the reliability of old samples. If the interval used between PING messages is50ms,50 ms, then to have a resolution of1/10001/1000, it takes 50secondsseconds, and a resolution of 1/10000 takes 500 seconds (more than 8 minutes). The reliability of a packet loss calculation based on a sliding window of 8 minutes depends on how fast network conditions evolve.</t> </section> <sectiontitle="Measurementsanchor="sec-9.5" numbered="true" toc="default"> <name>Measurements andReactions" anchor="section-9.5"><t>Reactions</name> <t> Q4S can be used as a mechanism to measure and trigger network tuning andapplication levelapplication-level actions (i.e. lowering video bit-rate,reducereducing multiplayer interaction speed,etc)etc.) inreal-timereal time in order to reach the application constraints, addressing measured possible network degradation.</t> </section> <sectiontitle="Instability Treatments" anchor="section-9.6"><t>anchor="sec-9.6" numbered="true" toc="default"> <name>Instability Treatments</name> <t> There are two scenarios in which Q4S can be affected by network problems: loss of Q4S packets and outlier samples.</t> <sectiontitle="Lossanchor="sec-9.6.1" numbered="true" toc="default"> <name>Loss of ControlPackets" anchor="section-9.6.1"><t>Packets</name> <t> Lost UDP packets (PING or BWIDTH messages) don't cause any problems for the Q4S state machine, but if TCP packets are delivered too late (which we will consider as "lost"), some undesirable consequences could arise.</t> <t> Q4S does have protection mechanisms to overcome these situations. Examples:</t><t><list style="symbols"><t>If<ul spacing="normal"> <li>If a BEGIN packetis lostor its correspondinganswer,answer is lost, after a certain timeout, the clientSHOULD<bcp14>SHOULD</bcp14> resend another BEGIN packet, resetting thesession</t> <t>Ifsession</li> <li>If a READY packet is lost, after a certain timeout, the clientSHOULD<bcp14>SHOULD</bcp14> resend another READYpacket.</t> <t>Ifpacket.</li> <li>If aQOS ALERTQ4S-ALERT requestis lostor its correspondinganswer,answer is lost, after a certain timeout, the originatorSHOULD<bcp14>SHOULD</bcp14> resend another Q4S-ALERTpacket.</t> <t>Ifpacket.</li> <li>If a CANCEL requestis lostor its correspondinganswer,answer is lost, after a certain timeout, the originatorSHOULD<bcp14>SHOULD</bcp14> resend another CANCELpacket.</t> </list> </t>packet.</li> </ul> </section> <sectiontitle="Outlier Samples" anchor="section-9.6.2"><t>anchor="sec-9.6.2" numbered="true" toc="default"> <name>Outlier Samples</name> <t> Outlier samples are those jitter or latency values far from the general/average values of most samples.</t> <t>HenceHence, the Q4S default measurement method uses the statistical median formula for latency calculation, and the outlier samples are neutralized. This is a very commonfilteringfilter for noise or errors on signal and image processing.</t> </section> </section> <sectiontitle="Scenarios" anchor="section-9.7"><t>anchor="sec-9.7" numbered="true" toc="default"> <name>Scenarios</name> <t> Q4S could be used in two scenarios:</t><t><list style="symbols"><t>client<ul spacing="normal"> <li>client to ACP(Application content provider)</t> <t>client</li> <li>client to client(peer to peer scenario)</t> </list> </t>(peer-to-peer scenario)</li> </ul> <sectiontitle="Clientanchor="sec-9.7.1" numbered="true" toc="default"> <name>Client toACP" anchor="section-9.7.1"><t>ACP</name> <t> One server:</t> <t> It is the common scenario in which the clientcontactcontacts the server to establish a Q4S session.</t> <t> N servers:</t> <t> In Content Delivery Networks and in general applications where delivery of contents can be achieved by different delivery nodes, two working mechanisms can bedefined</t> <t><list style="symbols"><t>Starting mode: End-userdefined:</t> <dl> <dt>Starting mode:</dt><dd>the end user may run Q4S against several delivery nodes and after some seconds choose the best one to start the multimediasession</t> <t>Prevention mode: Duringsession.</dd> <dt>Prevention mode:</dt><dd>during a streaming session, the user keeps several Q4S dialogs against different alternative delivery nodes. In case of congestion,end-user MAYthe end user <bcp14>MAY</bcp14> change to the best alternative deliverynode</t> </list> </t>node.</dd> </dl> </section> <sectiontitle="Clientanchor="sec-9.7.2" numbered="true" toc="default"> <name>Client toClient" anchor="section-9.7.2"><t>Client</name> <t> In order to solve theclient to clientclient-to-client scenario, a Q4S register functionMUST<bcp14>MUST</bcp14> be implemented. This allows clients to contact each other for sending the BEGIN message. In this scenario, the Register server would be used by peers to publish their Q4S-Resource-Server header and their public IP address tomake possibleenable the assumption of the server role.</t> <t> The register function is out of scope of this protocolversion,version because different HTTP mechanisms can beusedused, and Q4SMUST NOT<bcp14>MUST NOT</bcp14> force any.</t> </section> </section> </section> <sectiontitle="Security Considerations" anchor="section-10"><section title="Confidentiality Issues" anchor="section-10.1"><t> Henceanchor="sec-10" numbered="true" toc="default"> <name>Security Considerations</name> <section anchor="sec-10.1" numbered="true" toc="default"> <name>Confidentiality Issues</name> <t> Because Q4S does not transport any application data, Q4S does not jeopardize the security of application data. However, other certain considerations may take place, like identity impersonation and measurements privacy and integrity.</t> </section> <sectiontitle="Integrityanchor="sec-10.2" numbered="true" toc="default"> <name>Integrity of Measurements andAuthentication" anchor="section-10.2"><t>Authentication</name> <t> Identity impersonation could potentially produce anomalous Q4S measurements. If this attack is based on spoofing of the server IP address, it can be avoided using the digital signaturemechanism,mechanism included in the SDP. The network can easily validate this digital signature using the public key of the server certificate.</t> <t> Integrity of Q4S measurements under any malicious manipulation (such as a Man-in-the-Middle (MITM) attack)relayrelies on the same mechanism, the SDP signature.</t> <t> The Signature header field contains the signed hash value of the SDP body in order to protect all the SDP data, including the measurements. This signature not only protects the integrity of data but also authenticates the server.</t> </section> <sectiontitle="Privacyanchor="sec-10.3" numbered="true" toc="default"> <name>Privacy ofMeasurements" anchor="section-10.3"><t>Measurements</name> <t> This protocol could be supported overIPSec.IPsec. Q4Srelaysrelies on UDP and TCP, andIPSecIPsec supports both. If Q4S is used for application-based QoS, then IPsec is operationallyvalid butvalid; however, if Q4S is used to trigger network-based actions, then measurements could bewrong,incorrect unlessIPSecthe IPsec ports can beconsidered at anya target of potential action over the network (such asprioritizationprioritizing IPsec flows to measure the new, upgraded state of certain applicationflows).</t>flows). </t> </section> <sectiontitle="Availability Issues" anchor="section-10.4"><t>anchor="sec-10.4" numbered="true" toc="default"> <name>Availability Issues</name> <t> Any loss of connectivity may interrupt the availability of the Q4Sservice,service andresultsmay result in higherpacket-losspacket loss measurements, which is just the desired behavior in these situations.</t> <t> In order to mitigate availability issues caused by malicious attacks (such as DoS and DDoS), a good practice is to enable the Q4S service only for authenticated users. Q4S can be launched after the user is authenticated by the application. At this moment,histhe user's IP address isknownknown, and the Q4S service may be enabled for this IP address.OtherwiseOtherwise, the Q4S service should appear unreachable.</t> </section> <sectiontitle="Bandwidthanchor="sec-10.5" numbered="true" toc="default"> <name>Bandwidth OccupancyIssues" anchor="section-10.5"><t>Issues</name> <t> Q4S bandwidth measurement is limited to the application needs. It means that all available bandwidth is not measured, but only the fraction required by the application. This allows other applications to usenormallythe rest of availablebandwidth.</t>bandwidth normally.</t> <t> However, a malicious Q4S client couldre-startsrestart Q4S sessions just after finishing thenegotiationNegotiation phase. The consequence would be to waste bandwidth for nothing.</t> <t> In order to mitigate this possible anomalous behavior, it isRECOMMENDED<bcp14>RECOMMENDED</bcp14> to configure the server to reject sessions from the sameend-pointendpoint when this situation is detected.</t> </section> </section> <sectiontitle="Futureanchor="sec-11" numbered="true" toc="default"> <name>Future Code PointRequirements" anchor="section-11"><t>Requirements</name> <t> If the ideas described in this document are pursued to become a protocol specification, then the code points described in this document will need to be assigned by IANA.</t> <sectiontitle="Service Port" anchor="section-11.1"><t> The need for ananchor="sec-11.1" numbered="true" toc="default"> <name>Service Port</name> <t> An assignedPORT is toport would make possible a futureQ4S aware network,Q4S-aware network capable ofreactreacting by itself to Q4S alerts. A specific port would simplify the identification of the protocol by network elements in charge oftakemaking possible reactive decisions. Therefore, the need for a port assignment by IANA may be postponedtountil there is the need for a futureQ4S awareQ4S-aware network.</t> <t> Service Name: Q4S</t> <t> Transport Protocol(s): TCP</t><t><list style="hanging" hangIndent="3"><t hangText="Assignee :"> <vspace blankLines="1"/> Name :<dl newline="true" spacing="normal" indent="3"> <dt>Assignee:</dt> <dd> <t> Name: Jose Javier Garcia Aranda<vspace blankLines="1"/></t> <t> Email: jose_javier.garcia_aranda@nokia.com </t><t hangText="Contact :"> <vspace blankLines="1"/> Name :</dd> <dt>Contact:</dt> <dd> <t> Name: Jose Javier Garcia Aranda<vspace blankLines="1"/></t> <t> Email: jose_javier.garcia_aranda@nokia.com </t></list> </t> <t><list style="hanging" hangIndent="6"> <t hangText="Description:"></dd> </dl> <dl newline="false" spacing="normal" indent="6"> <dt>Description:</dt> <dd> The service associated with this request is in charge of the establishment of new Q4S sessions, and during thesessionsession, manages thepasshandoff to a new protocolstage (handshake, negotiationphase (Handshake, Negotiation andcontinuity)Continuity) as well asinform ofsends alerts when measurements do not meet therequirements.</t> <t hangText="Reference:"> thisrequirements.</dd> <dt>Reference:</dt> <dd> This document. This service does not use IP-layer broadcast, multicast, or anycastcommunication.</t> </list> </t>communication.</dd> </dl> </section> </section> <section anchor="sec-12" numbered="true" toc="default"> <name>IANA Considerations</name> <t>This document has no IANA actions.</t> </section> </middle> <back><references title="Normative References"><displayreference target="I-D.ietf-quic-transport" to="QUIC"/> <displayreference target="I-D.rhee-tcpm-cubic" to="CUBIC"/> <displayreference target="I-D.sridharan-tcpm-ctcp" to="CTCP"/> <references> <name>References</name> <references> <name>Normative References</name> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.7230.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.7231.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.7232.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.7233.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.7234.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.7235.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.2818.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.2119.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.8174.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.3986.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.3629.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.5322.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.5234.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.6234.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.8017.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.3264.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.4566.xml"/> </references> <references> <name>Informative References</name> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.3550.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.0793.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.0792.xml"/> <xi:include href="https://www.rfc-editor.org/refs/bibxml3/reference.I-D.ietf-quic-transport.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.4656.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.5357.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.3261.xml"/> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.0768.xml"/> <referenceanchor="ref-1" target="https://www.rfc-editor.org/info/rfc7230"><front> <title>Hypertext Transfer Protocol (HTTP/1.1): Message Syntax and Routing</title> <author fullname="R. Fielding" initials="R." surname="Fielding" role="editor"> </author> <author fullname="J. Reschke" initials="J." surname="Reschke" role="editor"> </author> <date month="June" year="2014"/> </front> <seriesInfo name="RFC" value="7230"/>anchor="RENO"> <front> <title>The Macroscopic Behavior of the TCP Congestion Avoidance Algorithm</title> <seriesInfo name="DOI"value="10.17487/RFC7230"/> </reference> <reference anchor="ref-2" target="https://www.rfc-editor.org/info/rfc7231"><front> <title>Hypertext Transfer Protocol (HTTP/1.1): Semantics and Content</title>value="10.1145/263932.264023"/> <authorfullname="R. Fielding" initials="R." surname="Fielding" role="editor">fullname="M. Mathis" initials="M." surname="Mathis"> </author> <author fullname="J.Reschke"Semke" initials="J."surname="Reschke" role="editor"> </author> <date month="June" year="2014"/> </front> <seriesInfo name="RFC" value="7231"/> <seriesInfo name="DOI" value="10.17487/RFC7231"/> </reference> <reference anchor="ref-3" target="https://www.rfc-editor.org/info/rfc7232"><front> <title>Hypertext Transfer Protocol (HTTP/1.1): Conditional Requests</title> <author fullname="R. Fielding" initials="R." surname="Fielding" role="editor">surname="Semke"> </author> <author fullname="J.Reschke"Mahdavi" initials="J."surname="Reschke" role="editor"> </author> <date month="June" year="2014"/> </front> <seriesInfo name="RFC" value="7232"/> <seriesInfo name="DOI" value="10.17487/RFC7232"/> </reference> <reference anchor="ref-4" target="https://www.rfc-editor.org/info/rfc7233"><front> <title>Hypertext Transfer Protocol (HTTP/1.1): Range Requests</title> <author fullname="R. Fielding" initials="R." surname="Fielding" role="editor"> </author> <author fullname="Y. Lafon" initials="Y." surname="Lafon" role="editor">surname="Mahdavi"> </author> <authorfullname="J. Reschke" initials="J." surname="Reschke" role="editor">fullname="T. Ott" initials="T." surname="Ott"> </author> <datemonth="June" year="2014"/>month="July" year="1997"/> </front><seriesInfo name="RFC" value="7233"/> <seriesInfo name="DOI" value="10.17487/RFC7233"/> </reference> <reference anchor="ref-5" target="https://www.rfc-editor.org/info/rfc7234"><front> <title>Hypertext Transfer Protocol (HTTP/1.1): Caching</title> <author fullname="R. Fielding" initials="R." surname="Fielding" role="editor"> </author> <author fullname="M. Nottingham" initials="M." surname="Nottingham" role="editor"> </author> <author fullname="J. Reschke" initials="J." surname="Reschke" role="editor"> </author> <date month="June" year="2014"/> </front> <seriesInfo name="RFC" value="7234"/> <seriesInfo name="DOI" value="10.17487/RFC7234"/> </reference> <reference anchor="ref-6" target="https://www.rfc-editor.org/info/rfc7235"><front> <title>Hypertext Transfer Protocol (HTTP/1.1): Authentication</title> <author fullname="R. Fielding" initials="R." surname="Fielding" role="editor"> </author> <author fullname="J. Reschke" initials="J." surname="Reschke" role="editor"> </author> <date month="June" year="2014"/> </front> <seriesInfo name="RFC" value="7235"/> <seriesInfo name="DOI" value="10.17487/RFC7235"/> </reference> <reference anchor="ref-8" target="https://www.rfc-editor.org/info/rfc3550"><front> <title>RTP: A Transport Protocol for Real-Time Applications</title> <author fullname="H. Schulzrinne" initials="H." surname="Schulzrinne"> </author> <author fullname="S. Casner" initials="S." surname="Casner"> </author> <author fullname="R. Frederick" initials="R." surname="Frederick"> </author> <author fullname="V. Jacobson" initials="V." surname="Jacobson"> </author> <date month="July" year="2003"/> </front> <seriesInfo name="STD" value="64"/> <seriesInfo name="RFC" value="3550"/> <seriesInfo name="DOI" value="10.17487/RFC3550"/> </reference> <!-- Version-Independent Properties of QUIC; companion document RFC YYYY --> <reference anchor="ref-9"> <front> <title>Version-Independent Properties of QUIC</title> <author initials='M' surname='Thomson' fullname='Martin Thomson'> <organization /> </author> <date month='July' day='10' year='2019' /> <abstract><t>This document defines the properties of the QUIC transport protocol that are expected to remain unchanged over time as new versions of the protocol are developed. Note to Readers Discussion of this draft takes place on the QUIC working group mailing list (quic@ietf.org), which is archived at https://mailarchive.ietf.org/arch/search/?email_list=quic [1]. Working Group information can be found at https://github.com/quicwg [2]; source code and issues list for this draft can be found at https://github.com/quicwg/base-drafts/labels/-invariants [3].</t></abstract> </front> <seriesInfo name="RFC" value="YYYY"/> <seriesInfo name="DOI" value="10.17487/RFCYYYY"/> </reference> <reference anchor="ref-10" target="https://www.rfc-editor.org/info/rfc4566"><front> <title>SDP: Session Description Protocol</title> <author fullname="M. Handley" initials="M." surname="Handley"> </author> <author fullname="V. Jacobson" initials="V." surname="Jacobson"> </author> <author fullname="C. Perkins" initials="C." surname="Perkins"> </author> <date month="July" year="2006"/> </front> <seriesInfo name="RFC" value="4566"/> <seriesInfo name="DOI" value="10.17487/RFC4566"/> </reference> <reference anchor="ref-11" target="https://www.rfc-editor.org/info/rfc2119"><front> <title>Key words for use in RFCs to Indicate Requirement Levels</title> <author fullname="S. Bradner" initials="S." surname="Bradner"> </author> <date month="March" year="1997"/> </front> <seriesInfo name="BCP" value="14"/> <seriesInfo name="RFC" value="2119"/> <seriesInfo name="DOI" value="10.17487/RFC2119"/> </reference> <reference anchor="ref-12" target="https://www.rfc-editor.org/info/rfc3986"><front> <title>Uniform Resource Identifier (URI): Generic Syntax</title> <author fullname="T. Berners-Lee" initials="T." surname="Berners-Lee"> </author> <author fullname="R. Fielding" initials="R." surname="Fielding"> </author> <author fullname="L. Masinter" initials="L." surname="Masinter"> </author> <date month="January" year="2005"/> </front> <seriesInfo name="STD" value="66"/> <seriesInfo name="RFC" value="3986"/> <seriesInfo name="DOI" value="10.17487/RFC3986"/> </reference> <reference anchor="ref-13" target="https://www.rfc-editor.org/info/rfc3264"><front> <title>An Offer/Answer Model with Session Description Protocol (SDP)</title> <author fullname="J. Rosenberg" initials="J." surname="Rosenberg"> </author> <author fullname="H. Schulzrinne" initials="H." surname="Schulzrinne"> </author> <date month="June" year="2002"/> </front> <seriesInfo name="RFC" value="3264"/> <seriesInfo name="DOI" value="10.17487/RFC3264"/> </reference> <reference anchor="ref-14" target="https://www.rfc-editor.org/info/rfc4634"><front> <title>US Secure Hash Algorithms (SHA and HMAC-SHA)</title> <author fullname="D. Eastlake 3rd" initials="D." surname="Eastlake 3rd"> </author> <author fullname="T. Hansen" initials="T." surname="Hansen"> </author> <date month="July" year="2006"/> </front> <seriesInfo name="RFC" value="4634"/> <seriesInfo name="DOI" value="10.17487/RFC4634"/> </reference> <reference anchor="ref-15" target="https://www.rfc-editor.org/info/rfc8017"><front> <title>PKCS #1: RSA Cryptography Specifications Version 2.2</title> <author fullname="K. Moriarty" initials="K." surname="Moriarty" role="editor"> </author> <author fullname="B. Kaliski" initials="B." surname="Kaliski"> </author> <author fullname="J. Jonsson" initials="J." surname="Jonsson"> </author> <author fullname="A. Rusch" initials="A." surname="Rusch"> </author> <date month="November" year="2016"/> </front> <seriesInfo name="RFC" value="8017"/> <seriesInfo name="DOI" value="10.17487/RFC8017"/> </reference> <reference anchor="ref-16" target="https://www.rfc-editor.org/info/rfc793"><front> <title>Transmission Control Protocol</title> <author fullname="J. Postel" initials="J." surname="Postel"> </author> <date month="September" year="1981"/> </front> <seriesInfo name="STD" value="7"/> <seriesInfo name="RFC" value="793"/> <seriesInfo name="DOI" value="10.17487/RFC0793"/> </reference> <reference anchor="ref-17" target="https://www.rfc-editor.org/info/rfc768"><front> <title>User Datagram Protocol</title> <author fullname="J. Postel" initials="J." surname="Postel"> </author> <date month="August" year="1980"/> </front> <seriesInfo name="STD" value="6"/> <seriesInfo name="RFC" value="768"/> <seriesInfo name="DOI" value="10.17487/RFC0768"/> </reference> <reference anchor="ref-18" target="https://www.rfc-editor.org/info/rfc3550"><front> <title>RTP: A Transport Protocol for Real-Time Applications</title> <author fullname="H. Schulzrinne" initials="H." surname="Schulzrinne"> </author> <author fullname="S. Casner" initials="S." surname="Casner"> </author> <author fullname="R. Frederick" initials="R." surname="Frederick"> </author> <author fullname="V. Jacobson" initials="V." surname="Jacobson"> </author> <date month="July" year="2003"/> </front> <seriesInfo name="STD" value="64"/> <seriesInfo name="RFC" value="3550"/> <seriesInfo name="DOI" value="10.17487/RFC3550"/> </reference> <reference anchor="ref-19" target="https://www.rfc-editor.org/info/rfc3629"><front> <title>UTF-8, a transformation format of ISO 10646</title> <author fullname="F. Yergeau" initials="F." surname="Yergeau"> </author> <date month="November" year="2003"/> </front> <seriesInfo name="STD" value="63"/> <seriesInfo name="RFC" value="3629"/> <seriesInfo name="DOI" value="10.17487/RFC3629"/> </reference> <reference anchor="ref-20" target="https://www.rfc-editor.org/info/rfc5322"><front> <title>Internet Message Format</title> <author fullname="P. Resnick" initials="P." surname="Resnick" role="editor"> </author> <date month="October" year="2008"/> </front> <seriesInfo name="RFC" value="5322"/> <seriesInfo name="DOI" value="10.17487/RFC5322"/> </reference> <reference anchor="ref-21" target="https://www.rfc-editor.org/info/rfc8174"><front> <title>Ambiguity of Uppercase vs Lowercase in RFC 2119 Key Words</title> <author fullname="B. Leiba" initials="B." surname="Leiba"> </author> <date month="May" year="2017"/> </front> <seriesInfo name="BCP" value="14"/> <seriesInfo name="RFC" value="8174"/> <seriesInfo name="DOI" value="10.17487/RFC8174"/> </reference> </references> <references title="Informative References"> <reference anchor="ref-22" target="https://www.rfc-editor.org/info/rfc3261"><front> <title>SIP: Session Initiation Protocol</title> <author fullname="J. Rosenberg" initials="J." surname="Rosenberg"> </author> <author fullname="H. Schulzrinne" initials="H." surname="Schulzrinne"> </author> <author fullname="G. Camarillo" initials="G." surname="Camarillo"> </author> <author fullname="A. Johnston" initials="A." surname="Johnston"> </author> <author fullname="J. Peterson" initials="J." surname="Peterson"> </author> <author fullname="R. Sparks" initials="R." surname="Sparks"> </author> <author fullname="M. Handley" initials="M." surname="Handley"> </author> <author fullname="E. Schooler" initials="E." surname="Schooler"> </author> <date month="June" year="2002"/> </front> <seriesInfo name="RFC" value="3261"/> <seriesInfo name="DOI" value="10.17487/RFC3261"/> </reference> <!-- [rfced] [ref-23] URL: https://cseweb.ucsd.edu/classes/wi01/cse222/papers/mathis-tcpmodel-ccr97.pdf --> <reference anchor="ref-23"><front> <title>The Macroscopic Behavior of the TCP Congestion Avoidance Algorithm</title> <author fullname="M. Mathis" initials="M." surname="Mathis"> </author> <author fullname="J. Semke" initials="J." surname="Semke"> </author> <author fullname="J. Mahdavi" initials="J." surname="Mahdavi"> </author> <author fullname="T. Ott" initials="T." surname="Ott"> </author> <date month="July" year="1997"/> </front> <seriesInfo name="Computer Communications Review," value="27(3)"/> </reference> <reference anchor="ref-24" target="https://www.rfc-editor.org/info/rfc3649"><front> <title>HighSpeed TCP for Large Congestion Windows</title> <author fullname="S. Floyd" initials="S." surname="Floyd"> </author> <date month="December" year="2003"/> </front> <seriesInfo name="RFC" value="3649"/> <seriesInfo name="DOI" value="10.17487/RFC3649"/> </reference> <!-- draft-rhee-tcpm-cubic-02; Expired --> <reference anchor='ref-25'> <front> <title>CUBIC for Fast Long-Distance Networks</title> <author initials='I' surname='Rhee' fullname='Injong Rhee'> <organization /> </author> <author initials='L' surname='Xu' fullname='Lisong Xu'> <organization /> </author> <author initials='S' surname='Ha' fullname='Sangtae Ha'> <organization /> </author> <date month='August' day='26' year='2008' /> <abstract><t>CUBIC is an extension to the current TCP standards. The protocol differs from the current TCP standards only in the congestion window adjustment function in the sender side. In particular, it uses a cubic function instead of a linear window increase of the current TCP standards to improve scalability and stability under fast and long distance networks. BIC-TCP, a predecessor of CUBIC, has been a default TCP adopted by Linux since year 2005 and has already been deployed globally and in use for several years by the Internet community at large. CUBIC is using a similar window growth function as BIC-TCP and is designed to be less aggressive and fairer to TCP in bandwidth usage than BIC-TCP while maintaining the strengths of BIC-TCP such as stability, window scalability and RTT fairness. Through extensive testing in various Internet scenarios, we believe that CUBIC is safe for deployment and testing in the global Internet. The intent of this document is to provide the protocol specification of CUBIC for a third party implementation and solicit the community feedback through experimentation on the performance of CUBIC. We expect this document to be eventually published as an experimental RFC.</t></abstract> </front> <seriesInfo name='Work in Progress,' value='draft-rhee-tcpm-cubic-02' /><refcontent>ACM SIGCOMM Computer Communication Review, pp. 67-82</refcontent> </reference> <xi:include href="https://xml2rfc.tools.ietf.org/public/rfc/bibxml/reference.RFC.3649.xml"/> <!-- draft-rhee-tcpm-cubic-02; Expired since 2009 --> <xi:include href="https://www.rfc-editor.org/refs/bibxml3/reference.I-D.rhee-tcpm-cubic.xml"/> <!-- draft-sridharan-tcpm-ctcp-02; Expired since 2009 --><reference anchor='ref-26'> <front> <title>Compound TCP: A New TCP Congestion Control for High-Speed and Long Distance Networks</title> <author initials='M' surname='Sridharan' fullname='Murali Sridharan'> <organization /> </author> <author initials='K' surname='Tan' fullname='Kun Tan'> <organization /> </author> <author initials='D' surname='Bansal' fullname='Deepak Bansal'> <organization /> </author> <author initials='D' surname='Thaler' fullname='Dave Thaler'> <organization /> </author> <date month='November' day='11' year='2008' /> <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='Work in Progress,' value='draft-sridharan-tcpm-ctcp-02' /> </reference> <reference anchor="ref-27" target="https://www.rfc-editor.org/info/rfc4656"><front> <title>A One-way Active Measurement Protocol (OWAMP)</title> <author fullname="S. Shalunov" initials="S." surname="Shalunov"> </author> <author fullname="B. Teitelbaum" initials="B." surname="Teitelbaum"> </author> <author fullname="A. Karp" initials="A." surname="Karp"> </author> <author fullname="J. Boote" initials="J." surname="Boote"> </author> <author fullname="M. Zekauskas" initials="M." surname="Zekauskas"> </author> <date month="September" year="2006"/> </front> <seriesInfo name="RFC" value="4656"/> <seriesInfo name="DOI" value="10.17487/RFC4656"/> </reference> <reference anchor="ref-28" target="https://www.rfc-editor.org/info/rfc5357"><front> <title>A Two-Way Active Measurement Protocol (TWAMP)</title> <author fullname="K. Hedayat" initials="K." surname="Hedayat"> </author> <author fullname="R. Krzanowski" initials="R." surname="Krzanowski"> </author> <author fullname="A. Morton" initials="A." surname="Morton"> </author> <author fullname="K. Yum" initials="K." surname="Yum"> </author> <author fullname="J. Babiarz" initials="J." surname="Babiarz"> </author> <date month="October" year="2008"/> </front> <seriesInfo name="RFC" value="5357"/> <seriesInfo name="DOI" value="10.17487/RFC5357"/> </reference><xi:include href="https://www.rfc-editor.org/refs/bibxml3/reference.I-D.sridharan-tcpm-ctcp.xml"/> </references> </references> <sectiontitle="Acknowledgments" anchor="section-13"><t>anchor="sec-13" numbered="false" toc="default"> <name>Acknowledgements</name> <t> Many people have made comments and suggestions contributing to this document. In particular, we would like to thank:</t> <t>Victor Villagra, Sonia Herranz, Clara<contact fullname="Victor Villagra"/>, <contact fullname="Sonia Herranz"/>, <contact fullname="Clara CubilloPastor, FranciscoPastor"/>, <contact fullname="Francisco DuranPina, Michael Scharf, JesusPina"/>, <contact fullname="Michael Scharf"/>, <contact fullname="Jesus SotoVisoViso"/>, andFederico Guillen.</t><contact fullname="Federico Guillen"/>.</t> <t> Additionally, we want to thank the Spanish Centre for the Development of Industrial Technology (CDTI) as well as the Spanish Science and TechMinistryMinistry, which funds this initiative through their innovation programs.</t> </section> <sectiontitle="Contributors" anchor="section-14"><figure><artwork><![CDATA[ Jacoboanchor="sec-14" numbered="false" toc="default"> <name>Contributors</name> <contact fullname="Jacobo PerezLajo Nokia Spain Email: jacobo.perez@nokia.com LuisLajo"> <organization>Nokia Spain</organization> <address> <email>jacobo.perez@nokia.com</email> </address> </contact> <contact fullname="Luis Miguel DiazVizcaino Nokia Spain Email: Luismi.Diaz@nokia.com GonzaloVizcaino"> <organization>Nokia Spain</organization> <address> <email>Luismi.Diaz@nokia.com</email> </address> </contact> <contact fullname="Gonzalo MunozFernandez Nokia Spain Email: gonzalo.munoz_fernandez.ext@nokia.com ManuelFernandez"> <organization>Nokia Spain</organization> <address> <email>gonzalo.munoz_fernandez.ext@nokia.com</email> </address> </contact> <contact fullname="Manuel AlarconGranero Nokia Spain Email: manuel.alarcon_granero.ext@nokia.com FranciscoGranero"> <organization>Nokia Spain</organization> <address> <email>manuel.alarcon_granero.ext@nokia.com</email> </address> </contact> <contact fullname="Francisco Josejuan Quintanilla Nokia Spain Email: francisco_jose.juan_quintanilla.ext@nokia.com Carlos Barcenilla UniversidadJuan Quintanilla"> <organization>Nokia Spain</organization> <address> <email>francisco_jose.juan_quintanilla.ext@nokia.com</email> </address> </contact> <contact fullname="Carlos Barcenilla"> <organization>Universidad Politecnica deMadrid Juan Quemada UniversidadMadrid</organization> </contact> <contact fullname="Juan Quemada"> <organization>Universidad Politecnica deMadrid Email: jquemada@dit.upm.es Ignacio Maestro TecnaliaMadrid</organization> <address> <email>jquemada@dit.upm.es</email> </address> </contact> <contact fullname="Ignacio Maestro"> <organization>Tecnalia Research& Innovation Email: ignacio.maestro@tecnalia.com Lara& Innovation</organization> <address> <email>ignacio.maestro@tecnalia.com</email> </address> </contact> <contact fullname="Lara FajardoIbanez Optiva Media Email: lara.fajardo@optivamedia.com Pablo Lopez Zapico Optiva Media Email: Pablo.lopez@optivamedia.com DavidIbañez"> <organization>Optiva Media</organization> <address> <email>lara.fajardo@optivamedia.com</email> </address> </contact> <contact fullname="Pablo López Zapico"> <organization>Optiva Media</organization> <address> <email>Pablo.lopez@optivamedia.com</email> </address> </contact> <contact fullname="David MuelasRecuenco UniversidadRecuenco"> <organization>Universidad Autonoma deMadrid Email: dav.muelas@uam.es JesusMadrid</organization> <address> <email>dav.muelas@uam.es</email> </address> </contact> <contact fullname="Jesus MolinaMerchan UniversidadMerchan"> <organization>Universidad Autonoma deMadrid jesus.molina@uam.es JorgeMadrid</organization> <address> <email>jesus.molina@uam.es</email> </address> </contact> <contact fullname="Jorge E. Lopez de VergaraMendez UniversidadMendez"> <organization>Universidad Autonoma deMadrid Email: jorge.lopez_vergara@uam.es VictorMadrid</organization> <address> <email>jorge.lopez_vergara@uam.es</email> </address> </contact> <contact fullname="Victor Manuel MarotoOrtega Optiva Media Email: victor.maroto@optivamedia.com ]]></artwork> </figure>Ortega"> <organization>Optiva Media</organization> <address> <email>victor.maroto@optivamedia.com</email> </address> </contact> </section> </back> </rfc>