idnits 2.17.1 draft-ietf-tsvwg-rtcweb-qos-06.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (December 18, 2015) is 3024 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) ** Downref: Normative reference to an Informational draft: draft-ietf-dart-dscp-rtp (ref. 'I-D.ietf-dart-dscp-rtp') == Outdated reference: A later version (-26) exists of draft-ietf-rtcweb-rtp-usage-25 == Outdated reference: A later version (-12) exists of draft-ietf-rtcweb-security-08 == Outdated reference: A later version (-17) exists of draft-ietf-rtcweb-transports-10 ** Downref: Normative reference to an Informational RFC: RFC 4594 Summary: 2 errors (**), 0 flaws (~~), 4 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group S. Dhesikan 3 Internet-Draft C. Jennings 4 Intended status: Standards Track Cisco Systems 5 Expires: June 20, 2016 D. Druta, Ed. 6 AT&T 7 P. Jones 8 Cisco Systems 9 December 18, 2015 11 DSCP and other packet markings for WebRTC QoS 12 draft-ietf-tsvwg-rtcweb-qos-06 14 Abstract 16 Many networks, such as service provider and enterprise networks, can 17 provide treatment for individual packets based on Differentiated 18 Services Code Point (DSCP) values on a per-hop basis. This document 19 provides the recommended DSCP values for browsers to use for various 20 classes of traffic. 22 Status of This Memo 24 This Internet-Draft is submitted in full conformance with the 25 provisions of BCP 78 and BCP 79. 27 Internet-Drafts are working documents of the Internet Engineering 28 Task Force (IETF). Note that other groups may also distribute 29 working documents as Internet-Drafts. The list of current Internet- 30 Drafts is at http://datatracker.ietf.org/drafts/current/. 32 Internet-Drafts are draft documents valid for a maximum of six months 33 and may be updated, replaced, or obsoleted by other documents at any 34 time. It is inappropriate to use Internet-Drafts as reference 35 material or to cite them other than as "work in progress." 37 This Internet-Draft will expire on June 20, 2016. 39 Copyright Notice 41 Copyright (c) 2015 IETF Trust and the persons identified as the 42 document authors. All rights reserved. 44 This document is subject to BCP 78 and the IETF Trust's Legal 45 Provisions Relating to IETF Documents 46 (http://trustee.ietf.org/license-info) in effect on the date of 47 publication of this document. Please review these documents 48 carefully, as they describe your rights and restrictions with respect 49 to this document. Code Components extracted from this document must 50 include Simplified BSD License text as described in Section 4.e of 51 the Trust Legal Provisions and are provided without warranty as 52 described in the Simplified BSD License. 54 Table of Contents 56 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 57 2. Relation to Other Standards . . . . . . . . . . . . . . . . . 3 58 3. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 4 59 4. Inputs . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 60 5. DSCP Mappings . . . . . . . . . . . . . . . . . . . . . . . . 5 61 6. Security Considerations . . . . . . . . . . . . . . . . . . . 7 62 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 7 63 8. Downward References . . . . . . . . . . . . . . . . . . . . . 7 64 9. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 7 65 10. Dedication . . . . . . . . . . . . . . . . . . . . . . . . . 8 66 11. Document History . . . . . . . . . . . . . . . . . . . . . . 8 67 12. References . . . . . . . . . . . . . . . . . . . . . . . . . 8 68 12.1. Normative References . . . . . . . . . . . . . . . . . . 8 69 12.2. Informative References . . . . . . . . . . . . . . . . . 9 70 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 9 72 1. Introduction 74 Differentiated Services Code Points (DSCP) [RFC2474] style packet 75 marking can help provide QoS in some environments. There are many 76 use cases where such marking does not help, but it seldom makes 77 things worse if packets are marked appropriately. In other words, if 78 too many packets, say all audio or all audio and video, are marked 79 for a given network condition then it can prevent desirable results. 80 Either too much other traffic will be starved, or there is not enough 81 capacity for the preferentially marked packets (i.e., audio and/or 82 video). 84 This specification proposes how WebRTC applications can mark packets. 85 This specification does not contradict or redefine any advice from 86 previous IETF RFCs, but merely provides a simple set of 87 recommendations for implementers based on the previous RFCs 89 There are some environments where DSCP markings frequently help. 90 These include: 92 1. Private, wide-area networks. 94 2. Residential Networks. If the congested link is the broadband 95 uplink in a cable or DSL scenario, often residential routers/NAT 96 support preferential treatment based on DSCP. 98 3. Wireless Networks. If the congested link is a local wireless 99 network, marking may help. 101 Traditionally DSCP values have been thought of as being site 102 specific, with each site selecting its own code points for 103 controlling per-hop-behavior to influence the QoS for transport-layer 104 flows. However in the WebRTC use cases, the browsers need to set 105 them to something when there is no site specific information. In 106 this document, "browsers" is used synonymously with "Interactive User 107 Agent" as defined in the HTML specification, 108 [W3C.REC-html5-20141028]. This document describes a subset of DSCP 109 code point values drawn from existing RFCs and common usage for use 110 with WebRTC applications. These code points are solely defaults. 112 This specification defines some inputs that the browser in a WebRTC 113 application can consider to aid in determining how to set the various 114 packet markings and defines the mapping from abstract QoS policies 115 (data type, priority level) to those packet markings. 117 2. Relation to Other Standards 119 This document exists as a complement to [I-D.ietf-dart-dscp-rtp], 120 which describes the interaction between DSCP and real-time 121 communications. It covers the implications of using various DSCP 122 values, particularly focusing on Real-time Transport Protocol (RTP) 123 [RFC3550] streams that are multiplexed onto a single transport-layer 124 flow. 126 There are a number of guidelines specified in 127 [I-D.ietf-dart-dscp-rtp] that should be followed when marking traffic 128 sent by WebRTC applications, as it is common for multiple RTP streams 129 to be multiplexed on the same transport-layer flow. Generally, the 130 RTP streams would be marked with a value as appropriate from Table 1. 131 A WebRTC application might also multiplex data channel 132 [I-D.ietf-rtcweb-data-channel] traffic over the same 5-tuple as RTP 133 streams, which would also be marked as per that table. The guidance 134 in [I-D.ietf-dart-dscp-rtp] says that all data channel traffic would 135 be marked with a single value that is typically different than the 136 value(s) used for RTP streams multiplexed with the data channel 137 traffic over the same 5-tuple, assuming RTP streams are marked with a 138 value other than default forwarding (DF). This is expanded upon 139 further in the next section. 141 This specification does not change or override the advice in any 142 other standards about setting packet markings. It simply selects a 143 subset of DSCP values that is relevant in the WebRTC context. This 144 document also specifies the inputs that are needed by the browser to 145 provide to the media engine. 147 The DSCP value set by the endpoint is not always trusted by the 148 network. Therefore, the DSCP value may be remarked at any place in 149 the network for a variety of reasons to any other DSCP value, 150 including default forwarding (DF) value to provide basic best effort 151 service. The mitigation for such action is through an authorization 152 mechanism. Such authorization mechanism is outside the scope of this 153 document. There is benefit in marking traffic even if it only 154 benefits the first few hops. 156 3. Terminology 158 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 159 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 160 document are to be interpreted as described in [RFC2119]. 162 4. Inputs 164 The below uses the concept of a media flow, however this is usually 165 not equivalent to a transport-layer flow defined by a 5-tuple (source 166 address, destination address, source port, destination port, and 167 protocol). Instead each media flow, such as an RTP stream 168 [I-D.ietf-rtcweb-rtp-usage] or SCTP association carrying data channel 169 packets [I-D.ietf-rtcweb-data-channel], contains all the packets 170 associated with an independent media entity within one 5-tuple. 171 Specifically, a media flow is the transmitted packets for an RTP 172 session or an SCTP association. There may be multiple media flows 173 within the same 5-tuple. These media flows might consist of 174 different media types and have different levels of importance to the 175 application and, therefore, each potentially marked using different 176 DSCP values than for another media flow multiplexed over the same 177 transport-layer flow. The following are the inputs that the browser 178 provides to the media engine: 180 o Data Type: The browser provides this input as it knows if the flow 181 is audio, interactive video with or without audio, non-interactive 182 video with or without audio, or data. 183 o Application Priority: Another input is the relative importance of 184 the flow within that data type. Many applications have multiple 185 media flows of the same data type and often some flows are more 186 important than others. For example, in a video conference where 187 there are usually audio and video flows, the audio flow may be 188 more important than the video flow. JavaScript applications can 189 tell the browser whether a particular media flow is high, medium, 190 low or very low importance to the application. 192 [I-D.ietf-rtcweb-transports] defines in more detail what an 193 individual media flow is within the WebRTC context. 195 As an example of different media flows that might be multiplexed over 196 the same transport-layer flow, packets related to one RTP stream 197 (e.g., an audio flow) carried over UDP might be one media flow, 198 packets related to a second RTP stream (e.g., presentation video) 199 carried over UDP might be a second media flow, and finally data 200 channel packets carried via SCTP over DTLS might be third media flow. 202 5. DSCP Mappings 204 Below is a table of DSCP markings for each data type of interest to 205 WebRTC. These DSCP values for each data type listed are a reasonable 206 subset of code point values taken from [RFC4594]. A web browser 207 SHOULD use these values to mark the appropriate media packets. More 208 information on EF can be found in [RFC3246]. More information on AF 209 can be found in [RFC2597]. DF is default forwarding which provides 210 the basic best effort service. 212 +------------------------+-------+------+-------------+-------------+ 213 | Data Type | Very | Low | Medium | High | 214 | | Low | | | | 215 +------------------------+-------+------+-------------+-------------+ 216 | Audio | CS1 | DF | EF (46) | EF (46) | 217 | | (8) | (0) | | | 218 | | | | | | 219 | Interactive Video with | CS1 | DF | AF42, AF43 | AF41, AF42 | 220 | or without audio | (8) | (0) | (36, 38) | (34, 36) | 221 | | | | | | 222 | Non-Interactive Video | CS1 | DF | AF32, AF33 | AF31, AF32 | 223 | with or without audio | (8) | (0) | (28, 30) | (26, 28) | 224 | | | | | | 225 | Data | CS1 | DF | AF11 | AF21 | 226 | | (8) | (0) | | | 227 +------------------------+-------+------+-------------+-------------+ 229 Table 1: Recommended DSCP Values for WebRTC Applications 231 The columns "very low", "low", "Medium" and "high" signify the 232 relative importance of the media flow within the application and is 233 an input that the browser receives to assist it in selecting the DSCP 234 value. These are referred to as application priority in this 235 document. Application priority does not refer to priority in the 236 network transport. 238 The above table assumes that packets marked with CS1 are treated as 239 "less than best effort". However, the treatment of CS1 is 240 implementation dependent. If an implementation treats CS1 as other 241 than "less than best effort", then the actual priority (or, more 242 precisely, the per-hop-behavior) of the packets may be changed from 243 what is intended. It is common for CS1 to be treated the same as DF 244 so anyone using CS1 cannot assume that CS1 will be treated 245 differently than DF. Implementers should also note that the excess 246 EF traffic is dropped. This could mean that a packet marked as EF 247 may not get through as opposed to a packet marked with a different 248 DSCP value. 250 The browser SHOULD first select the data type of the media flow. 251 Within the data type, the relative importance of the media flow 252 SHOULD be used to select the appropriate DSCP value. 254 The combination of data type and application priority provides 255 specificity and helps in selecting the right DSCP value for the media 256 flow. In some cases, the different drop precedence values provides 257 additional granularity in classifying packets within a media flow. 258 For example, in a video conference, the video media flow may have 259 medium application priority. If so, either AF42 or AF43 may be 260 selected. If the I-frames in the stream are more important than the 261 P-frames, then the I-frames can be marked with AF42 and the P-frames 262 marked with AF43. 264 All packets within a media flow SHOULD have the same application 265 priority. In some cases, the selected cell may have multiple DSCP 266 values, such as AF41 and AF42. These offer different drop 267 precedences. With the exception of data channel traffic, one may 268 select different drop precedences for the different packets in the 269 same media flow. Therefore, all packets in the media flow SHOULD be 270 marked with the same application priority, but can have different 271 drop precedences. 273 For reasons discussed in Section 6 of [I-D.ietf-dart-dscp-rtp], if 274 multiple media flows are multiplexed using a reliable transport 275 (e.g., TCP) then all of the packets for all media flows multiplexed 276 over that transport-layer flow MUST be marked using the same DSCP 277 value. Likewise, all WebRTC data channel packets transmitted over an 278 SCTP association MUST be marked using the same DSCP value, regardless 279 of how many data channels (streams) exist or what kind of traffic is 280 carried over the various SCTP streams. In the event that the browser 281 wishes to change the DSCP value in use for an SCTP association, it 282 MUST reset the SCTP congestion controller after changing values. 283 Frequent changes in the DSCP value used for an SCTP association are 284 discouraged, though, as this would defeat any attempts at effectively 285 managing congestion. It should also be noted that any change in DSCP 286 value that results in a reset of the congestion controller puts the 287 SCTP association back into slow start, which may have undesirable 288 effects on application performance. 290 For the data channel traffic multiplexed over an SCTP association, it 291 is RECOMMENDED that the DSCP value selected be the one associated 292 with the highest priority requested for all data channels multiplexed 293 over the SCTP association. Likewise, when multiplexing multiple 294 media flows over a TCP connection, the DCSP value selected should be 295 the one associated with the highest priority requested for all 296 multiplexed flows. 298 If a packet enters a QoS domain that has no support for the above 299 defined data types/application priority (service class), then the 300 network node at the edge will remark the DSCP value based on 301 policies. This could result in the media flow not getting the 302 network treatment it expects based on the original DSCP value in the 303 packet. Subsequently, if the packet enters a QoS domain that 304 supports a larger number of service classes, there may not be 305 sufficient information in the packet to restore the original 306 markings. Mechanisms for restoring such original DSCP is outside the 307 scope of this document. 309 In summary, there are no guarantees or promised level of service with 310 the use of DSCP. The service provided to a packet is dependent upon 311 the network design along the path, as well as the congestion levels 312 at every hop. 314 6. Security Considerations 316 This specification does not add any additional security implication 317 other than the normal application use of DSCP. For security 318 implications on use of DSCP, please refer to Section 6 of RFC 4594. 319 Please also see [I-D.ietf-rtcweb-security] as an additional 320 reference. 322 7. IANA Considerations 324 This specification does not require any actions from IANA. 326 8. Downward References 328 This specification contains a downwards reference to [RFC4594]. 329 However, the parts of that RFC used by this specification are 330 sufficiently stable for this downward reference. 332 9. Acknowledgements 334 Thanks To David Black, Magnus Westerland, Paolo Severini, Jim 335 Hasselbrook, Joe Marcus, Erik Nordmark, and Michael Tuexen for their 336 help. 338 10. Dedication 340 This document is dedicated to the memory of James Polk, a long-time 341 friend and colleague. James made important contributions to this 342 specification, including being one of its primary authors. The IETF 343 global community mourns his loss and he will be missed dearly. 345 11. Document History 347 Note to RFC Editor: Please remove this section. 349 This document was originally an individual submission in RTCWeb WG. 350 The RTCWeb working group selected it to be become a WG document. 351 Later the transport ADs requested that this be moved to the TSVWG WG 352 as that seemed to be a better match. This document is now being 353 submitted as individual submission to the TSVWG with the hope that WG 354 will select it as a WG draft and move it forward to an RFC. 356 12. References 358 12.1. Normative References 360 [I-D.ietf-dart-dscp-rtp] 361 Black, D. and P. Jones, "Differentiated Services 362 (DiffServ) and Real-time Communication", draft-ietf-dart- 363 dscp-rtp-10 (work in progress), November 2014. 365 [I-D.ietf-rtcweb-data-channel] 366 Jesup, R., Loreto, S., and M. Tuexen, "WebRTC Data 367 Channels", draft-ietf-rtcweb-data-channel-13 (work in 368 progress), January 2015. 370 [I-D.ietf-rtcweb-rtp-usage] 371 Perkins, C., Westerlund, M., and J. Ott, "Web Real-Time 372 Communication (WebRTC): Media Transport and Use of RTP", 373 draft-ietf-rtcweb-rtp-usage-25 (work in progress), June 374 2015. 376 [I-D.ietf-rtcweb-security] 377 Rescorla, E., "Security Considerations for WebRTC", draft- 378 ietf-rtcweb-security-08 (work in progress), February 2015. 380 [I-D.ietf-rtcweb-transports] 381 Alvestrand, H., "Transports for WebRTC", draft-ietf- 382 rtcweb-transports-10 (work in progress), October 2015. 384 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 385 Requirement Levels", BCP 14, RFC 2119, March 1997. 387 [RFC4594] Babiarz, J., Chan, K., and F. Baker, "Configuration 388 Guidelines for DiffServ Service Classes", RFC 4594, August 389 2006. 391 12.2. Informative References 393 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 394 "Definition of the Differentiated Services Field (DS 395 Field) in the IPv4 and IPv6 Headers", RFC 2474, December 396 1998. 398 [RFC2597] Heinanen, J., Baker, F., Weiss, W., and J. Wroclawski, 399 "Assured Forwarding PHB Group", RFC 2597, June 1999. 401 [RFC3246] Davie, B., Charny, A., Bennet, J., Benson, K., Le Boudec, 402 J., Courtney, W., Davari, S., Firoiu, V., and D. 403 Stiliadis, "An Expedited Forwarding PHB (Per-Hop 404 Behavior)", RFC 3246, March 2002. 406 [RFC3550] Schulzrinne, H., Casner, S., Frederick, R., and V. 407 Jacobson, "RTP: A Transport Protocol for Real-Time 408 Applications", STD 64, RFC 3550, DOI 10.17487/RFC3550, 409 July 2003, . 411 [W3C.REC-html5-20141028] 412 Hickson, I., Berjon, R., Faulkner, S., Leithead, T., 413 Navara, E., O'Connor, E., and S. Pfeiffer, "HTML5", 414 World Wide Web Consortium Recommendation REC- 415 html5-20141028, October 2014, 416 . 418 Authors' Addresses 420 Subha Dhesikan 421 Cisco Systems 423 Email: sdhesika@cisco.com 425 Cullen Jennings 426 Cisco Systems 428 Email: fluffy@cisco.com 429 Dan Druta (editor) 430 AT&T 432 Email: dd5826@att.com 434 Paul E. Jones 435 Cisco Systems 437 Email: paulej@packetizer.com