idnits 2.17.1 draft-lennox-avtcore-rtp-multi-stream-02.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 (Using the creation date from RFC3550, updated by this document, for RFC5378 checks: 1998-04-07) -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (February 25, 2013) is 4071 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) == Missing Reference: 'TBA' is mentioned on line 457, but not defined == Missing Reference: 'RFCXXXX' is mentioned on line 629, but not defined ** Obsolete normative reference: RFC 6222 (Obsoleted by RFC 7022) == Outdated reference: A later version (-13) exists of draft-ietf-avtcore-multi-media-rtp-session-01 == Outdated reference: A later version (-25) exists of draft-ietf-clue-framework-09 == Outdated reference: A later version (-54) exists of draft-ietf-mmusic-sdp-bundle-negotiation-03 == Outdated reference: A later version (-02) exists of draft-westerlund-avtcore-rtp-topologies-update-01 Summary: 1 error (**), 0 flaws (~~), 7 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 AVTCORE J. Lennox 3 Internet-Draft Vidyo 4 Updates: 3550 (if approved) M. Westerlund 5 Intended status: Standards Track Ericsson 6 Expires: August 29, 2013 Q. Wu 7 Huawei 8 C. Perkins 9 University of Glasgow 10 February 25, 2013 12 RTP Considerations for Endpoints Sending Multiple Media Streams 13 draft-lennox-avtcore-rtp-multi-stream-02 15 Abstract 17 This document expands and clarifies the behavior of the Real-Time 18 Transport Protocol (RTP) endpoints when they are sending multiple 19 media streams in a single RTP session. In particular, issues 20 involving Real-Time Transport Control Protocol (RTCP) messages are 21 described. 23 This document updates RFC 3550 in regards to handling of multiple 24 SSRCs per endpoint in RTP sessions. 26 Status of This Memo 28 This Internet-Draft is submitted in full conformance with the 29 provisions of BCP 78 and BCP 79. 31 Internet-Drafts are working documents of the Internet Engineering 32 Task Force (IETF). Note that other groups may also distribute 33 working documents as Internet-Drafts. The list of current Internet- 34 Drafts is at http://datatracker.ietf.org/drafts/current/. 36 Internet-Drafts are draft documents valid for a maximum of six months 37 and may be updated, replaced, or obsoleted by other documents at any 38 time. It is inappropriate to use Internet-Drafts as reference 39 material or to cite them other than as "work in progress." 41 This Internet-Draft will expire on August 29, 2013. 43 Copyright Notice 45 Copyright (c) 2013 IETF Trust and the persons identified as the 46 document authors. All rights reserved. 48 This document is subject to BCP 78 and the IETF Trust's Legal 49 Provisions Relating to IETF Documents 50 (http://trustee.ietf.org/license-info) in effect on the date of 51 publication of this document. Please review these documents 52 carefully, as they describe your rights and restrictions with respect 53 to this document. Code Components extracted from this document must 54 include Simplified BSD License text as described in Section 4.e of 55 the Trust Legal Provisions and are provided without warranty as 56 described in the Simplified BSD License. 58 Table of Contents 60 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 61 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 3 62 3. Use Cases For Multi-Stream Endpoints . . . . . . . . . . . . 3 63 3.1. Multiple-Capturer Endpoints . . . . . . . . . . . . . . . 3 64 3.2. Multi-Media Sessions . . . . . . . . . . . . . . . . . . 4 65 3.3. Multi-Stream Mixers . . . . . . . . . . . . . . . . . . . 4 66 4. Issue Cases . . . . . . . . . . . . . . . . . . . . . . . . . 4 67 4.1. Cascaded Multi-party Conference with Source Projecting 68 Mixers . . . . . . . . . . . . . . . . . . . . . . . . . 5 69 5. Multi-Stream Endpoint RTP Media Recommendations . . . . . . . 5 70 6. Multi-Stream Endpoint RTCP Recommendations . . . . . . . . . 5 71 6.1. RTCP Reporting Requirement . . . . . . . . . . . . . . . 6 72 6.2. Initial Reporting Interval . . . . . . . . . . . . . . . 6 73 6.3. Compound RTCP Packets . . . . . . . . . . . . . . . . . . 6 74 7. RTCP Bandwidth Considerations for Sources with Disparate 75 Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 76 8. Grouping of RTCP Reception Statistics and Other Feedback . . 7 77 8.1. Semantics and Behavior of Reporting Groups . . . . . . . 8 78 8.2. Determine the Report Group . . . . . . . . . . . . . . . 9 79 8.3. RTCP Packet Reporting Group's Reporting Sources . . . . . 9 80 8.4. RTCP Source Description (SDES) item for Reporting Groups 11 81 8.5. Middlebox Considerations . . . . . . . . . . . . . . . . 11 82 8.6. SDP signaling for Reporting Groups . . . . . . . . . . . 11 83 8.7. Bandwidth Benefits of RTCP Reporting Groups . . . . . . . 11 84 8.8. Consequences of RTCP Reporting Groups . . . . . . . . . . 12 85 9. Security Considerations . . . . . . . . . . . . . . . . . . . 13 86 10. Open Issues . . . . . . . . . . . . . . . . . . . . . . . . . 13 87 11. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 13 88 11.1. RTCP SDES Item . . . . . . . . . . . . . . . . . . . . . 13 89 11.2. RTCP Packet Type . . . . . . . . . . . . . . . . . . . . 14 90 12. References . . . . . . . . . . . . . . . . . . . . . . . . . 14 91 12.1. Normative References . . . . . . . . . . . . . . . . . . 14 92 12.2. Informative References . . . . . . . . . . . . . . . . . 14 93 Appendix A. Changes From Earlier Versions . . . . . . . . . . . 15 94 A.1. Changes From Draft -01 . . . . . . . . . . . . . . . . . 15 95 A.2. Changes From Draft -00 . . . . . . . . . . . . . . . . . 16 97 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 16 99 1. Introduction 101 At the time The Real-Time Transport Protocol (RTP) [RFC3550] was 102 originally written, and for quite some time after, endpoints in RTP 103 sessions typically only transmitted a single media stream per RTP 104 session, where separate RTP sessions were typically used for each 105 distinct media type. 107 Recently, however, a number of scenarios have emerged (discussed 108 further in Section 3) in which endpoints wish to send multiple RTP 109 media streams, distinguished by distinct RTP synchronization source 110 (SSRC) identifiers, in a single RTP session. Although RTP's initial 111 design did consider such scenarios, the specification was not 112 consistently written with such use cases in mind. The specifications 113 are thus somewhat unclear. 115 The purpose of this document is to expand and clarify [RFC3550]'s 116 language for these use cases. The authors believe this does not 117 result in any major normative changes to the RTP specification, 118 however this document defines how the RTP specification is to be 119 interpreted. In these cases, this document updates RFC3550. 121 The document starts with terminology and some use cases where 122 multiple sources will occur. This is followed by some case studies 123 to try to identify issues that exist and need considerations. This 124 is followed by RTP and RTCP recommendations to resolve issues. Next 125 are security considerations and remaining open issues. 127 2. Terminology 129 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 130 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 131 "OPTIONAL" in this document are to be interpreted as described in RFC 132 2119 [RFC2119] and indicate requirement levels for compliant 133 implementations. 135 3. Use Cases For Multi-Stream Endpoints 137 This section discusses several use cases that have motivated the 138 development of endpoints that send multiple streams in a single RTP 139 session. 141 3.1. Multiple-Capturer Endpoints 143 The most straightforward motivation for an endpoint to send multiple 144 media streams in a session is the scenario where an endpoint has 145 multiple capture devices of the same media type and characteristics. 146 For example, telepresence endpoints, of the type described by the 147 CLUE Telepresence Framework [I-D.ietf-clue-framework] is designed, 148 often have multiple cameras or microphones covering various areas of 149 a room. 151 3.2. Multi-Media Sessions 153 Recent work has been done in RTP 154 [I-D.ietf-avtcore-multi-media-rtp-session] and SDP 155 [I-D.ietf-mmusic-sdp-bundle-negotiation] to update RTP's historical 156 assumption that media streams of different media types would always 157 be sent on different RTP sessions. In this work, a single endpoint's 158 audio and video media streams (for example) are instead sent in a 159 single RTP session. 161 3.3. Multi-Stream Mixers 163 There are several RTP topologies which can involve a central device 164 that itself generates multiple media streams in a session. 166 One example is a mixer providing centralized compositing for a multi- 167 capture scenario like that described in Section 3.1. In this case, 168 the centralized node is behaving much like a multi-capturer endpoint, 169 generating several similar and related sources. 171 More complicated is the Source Projecting Mixer, see Section 3.6 of 172 [I-D.westerlund-avtcore-rtp-topologies-update]. This is a central 173 box that receives media streams from several endpoints, and then 174 selectively forwards modified versions of some of the streams toward 175 the other endpoints it is connected to. Toward one destination, a 176 separate media source appears in the session for every other source 177 connected to the mixer, "projected" from the original streams, but at 178 any given time many of them can appear to be inactive (and thus are 179 receivers, not senders, in RTP). This sort of device is closer to 180 being an RTP mixer than an RTP translator, in that it terminates RTCP 181 reporting about the mixed streams, and it can re-write SSRCs, 182 timestamps, and sequence numbers, as well as the contents of the RTP 183 payloads, and can turn sources on and off at will without appearing 184 to be generating packet loss. Each projected stream will typically 185 preserve its original RTCP source description (SDES) information. 187 4. Issue Cases 189 This section illustrates some scenarios that have shown areas where 190 the RTP specification is unclear. 192 4.1. Cascaded Multi-party Conference with Source Projecting Mixers 194 This issue case tries to illustrate the effect of having multiple 195 SSRCs sent by an endpoint, by considering the traffic between two 196 source-projecting mixers in a large multi-party conference. 198 For concreteness, consider a 200-person conference, where 16 sources 199 are viewed at any given time. Assuming participants are distributed 200 evenly among the mixers, each mixer would have 100 sources "behind" 201 (projected through) it, of which at any given time eight are active 202 senders. Thus, the RTP session between the mixers consists of two 203 endpoints, but 200 sources. 205 The RTCP bandwidth implications of this scenario are discussed 206 further in Section 8.7. 208 (TBD: Other examples? Can this section be removed?) 210 5. Multi-Stream Endpoint RTP Media Recommendations 212 While an endpoint MUST (of course) stay within its share of the 213 available session bandwidth, as determined by signalling and 214 congestion control, this need not be applied independently or 215 uniformly to each media stream. In particular, session bandwidth MAY 216 be reallocated among an endpoint's media streams, for example by 217 varying the bandwidth use of a variable-rate codec, or changing the 218 codec used by the media stream, up to the constraints of the 219 session's negotiated (or declared) codecs. This includes enabling or 220 disabling media streams as more or less bandwidth becomes available. 222 6. Multi-Stream Endpoint RTCP Recommendations 224 This section contains a number of different RTCP clarifications or 225 recommendations that enables more efficient and simpler behavior 226 without loss of functionality. 228 The RTP Control Protocol (RTCP) is defined in Section 6 of [RFC3550], 229 but it is largely documented in terms of "participants". In many 230 cases, the specification's recommendations for "participants" are to 231 be interpreted as applying to individual media streams, rather than 232 to endpoints. This section describes several concrete cases where 233 this applies. 235 (tbd: rather than think in terms of media streams, it might be 236 clearer to refer to SSRC values, where a participant with multiple 237 active SSRC values counts as multiple participants, once per SSRC) 239 6.1. RTCP Reporting Requirement 241 For each of an endpoint's media streams, whether or not it is 242 currently sending media, SR/RR and SDES packets MUST be sent at least 243 once per RTCP report interval. (For discussion of the content of SR 244 or RR packets' reception statistic reports, see Section 8.) 246 6.2. Initial Reporting Interval 248 When a new media stream is added to a unicast session, the sentence 249 in [RFC3550]'s Section 6.2 applies: "For unicast sessions ... the 250 delay before sending the initial compound RTCP packet MAY be zero." 251 This applies to individual media sources as well. Thus, endpoints 252 MAY send an initial RTCP packet for an SSRC immediately upon adding 253 it to a unicast session. 255 This allowance also applies, as written, when initially joining a 256 unicast session. However, in this case some caution needs to be 257 exercised if the end-point or mixer has a large number of sources 258 (SSRCs) as this can create a significant burst. How big an issue 259 this depends on the number of source to send initial SR or RR and 260 Session Description CNAME items for in relation to the RTCP 261 bandwidth. 263 (tbd: Maybe some recommendation here? The aim in restricting this to 264 unicast sessions was to avoid this burst of traffic, which the usual 265 RTCP timing and reconsideration rules will prevent) 267 6.3. Compound RTCP Packets 269 Section 6.1 gives the following advice to RTP translators and mixers: 271 It is RECOMMENDED that translators and mixers combine individual 272 RTCP packets from the multiple sources they are forwarding into 273 one compound packet whenever feasible in order to amortize the 274 packet overhead (see Section 7). An example RTCP compound packet 275 as might be produced by a mixer is shown in Fig. 1. If the 276 overall length of a compound packet would exceed the MTU of the 277 network path, it SHOULD be segmented into multiple shorter 278 compound packets to be transmitted in separate packets of the 279 underlying protocol. This does not impair the RTCP bandwidth 280 estimation because each compound packet represents at least one 281 distinct participant. Note that each of the compound packets MUST 282 begin with an SR or RR packet. 284 Note: To avoid confusion, an RTCP packet is an individual item, such 285 as a Sender Report (SR), Receiver Report (RR), Source Description 286 (SDES), Goodbye (BYE), Application Defined (APP), Feedback [RFC4585] 287 or Extended Report (XR) [RFC3611] packet. A compound packet is the 288 combination of two or more such RTCP packets where the first packet 289 has to be an SR or an RR packet, and which contains a SDES packet 290 containing an CNAME item. Thus the above results in compound RTCP 291 packets that contain multiple SR or RR packets from different sources 292 as well as any of the other packet types. There are no restrictions 293 on the order in which the packets can occur within the compound 294 packet, except the regular compound rule, i.e., starting with an SR 295 or RR. 297 This advice applies to multi-media-stream endpoints as well, with the 298 same restrictions and considerations. (Note, however, that the last 299 sentence does not apply to AVPF [RFC4585] or SAVPF [RFC5124] feedback 300 packets if Reduced-Size RTCP [RFC5506] is in use.) 302 Due to RTCP's randomization of reporting times, there is a fair bit 303 of tolerance in precisely when an endpoint schedules RTCP to be sent. 304 Thus, one potential way of implementing this recommendation would be 305 to randomize all of an endpoint's sources together, with a single 306 randomization schedule, so an MTU's worth of RTCP all comes out 307 simultaneously. 309 (tbd: Multiplexing RTCP packets from multiple different sources might 310 require some adjustment to the calculation of RTCP's avg_rtcp_size, 311 as the RTCP group interval is proportional to avg_rtcp_size times the 312 group size). 314 7. RTCP Bandwidth Considerations for Sources with Disparate Rates 316 It is possible for an RTP session to carry sources of greatly 317 differing bandwidths. One example is the scenario of 318 [I-D.ietf-avtcore-multi-media-rtp-session], when audio and video are 319 sent in the same session. However, this can occur even within a 320 single media type, for example a video session carrying both 5 fps 321 QCIF and 60 fps 1080p HD video, or an audio session carrying both 322 G.729 and L24/48000/6 audio. 324 (tbd: recommend how RTCP bandwidths are to be chosen in these 325 scenarios. Likely, these recommendations will be different for 326 sessions using AVPF-based profiles (where the trr-int parameter is 327 available) than for those using AVP.) 329 8. Grouping of RTCP Reception Statistics and Other Feedback 331 As specified by [RFC3550], an endpoint MUST send reception reports 332 about every active media stream it is receiving, from at least one 333 local source. 335 However, a naive application of the RTP specification's rules could 336 be quite inefficient. In particular, if a session has N SSRCs 337 (active and inactive, i.e., participant SSRCs), and the session has S 338 active senders in each reporting interval, there will either be N*S 339 report blocks per reporting interval, or (per the round-robin 340 recommendations of [RFC3550] Section 6.1) reception sources would be 341 unnecessarily round-robinned. In a session where most media sources 342 become senders reasonably frequently, this results in quadratically 343 many reception report blocks in the conference, or reporting delays 344 proportional to the number of session members. 346 Since traffic is received by endpoints, however, rather than by media 347 sources, there is not actually any need for this quadratic expansion. 348 All that is needed is for each endpoint to report all the remote 349 sources it is receiving. 351 Thus, this document defines a new RTCP mechanism, Reporting Groups, 352 to indicate sources which originate from the same endpoint, and which 353 therefore would have identical recption reports. 355 8.1. Semantics and Behavior of Reporting Groups 357 An RTCP Reporting Group indicates that a set of sources (SSRCs) that 358 originate from a single entity (endpoint or middlebox) in an RTP 359 session, and therefore all the sources in the group's view of the 360 network is identical. If reporting groups are in use, two sources 361 SHOULD be put into the same reporting group if their view of the 362 network is identical; i.e., if they report on traffic received at the 363 same interface of an RTP endpoint. Sources with different views of 364 the network MUST NOT be put into the same reporting group. 366 If reporting groups are in use, an endpoint MUST NOT send reception 367 reports from one source in a reporting group about another one in the 368 same group ("self-reports"). Similarly, an endpoint MUST NOT send 369 reception reports about a remote media source from more than one 370 source in a reporting group ("cross-reports"). Instead, it MUST pick 371 one of its local media sources as the "reporting" source for each 372 remote media source, and use it to send reception reports about that 373 remote source; all the other media sources in the reporting group 374 MUST NOT send any reception reports about that remote media source. 376 This reporting source MUST also be the source for any RTP/AVPF 377 Feedback [RFC4585] or Extended Report (XR) [RFC3611] packets about 378 the corresponding remote sources as well. If a reporting source 379 leaves the session (i.e., if it sends a BYE, or leaves the group 380 without sending BYE under the rules of [RFC3550] section 6.3.7), 381 another reporting source MUST be chosen if any members of the group 382 still exist. 384 An endpoint or middlebox MAY use multiple sources as reporting 385 sources; however, each reporting source MUST have non-overlapping 386 sets of remote SSRCs it reports on. This is primarily to be done 387 when the reporting source's number of reception report blocks is so 388 large that it would be forced to round robin around the sources. 389 Thus, by splitting the reports among several reporting SSRCs more 390 consistent reporting can be achieved. 392 If RTP/AVPF feedback is in use, a reporting source MAY send immediate 393 or early feedback at any point when any member of the reporting group 394 could validly do so. 396 An endpoint SHOULD NOT create single-source reporting groups, unless 397 it is anticipated that the group might have additional sources added 398 to it in the future. 400 8.2. Determine the Report Group 402 A remote RTP entity, such as an endpoint or a middlebox needs to be 403 able to determine the report group used by another RTP entity. To 404 achieve this goal two RTCP extensions has been defined. For the 405 SSRCs that are reporting on behalf of the reporting group an SDES 406 item RGRP has been defined for providing the report group with an 407 identifier. For SSRCs that aren't reporting on any peer SSRC a new 408 RTCP packet type is defined. This RTCP packet type "Reporting 409 Sources", lists the SSRC that are reporting on this SSRC's behalf. 411 This divided approach has been selected for the following reasons: 413 1. Enable an explicit indication of who reports on this SSRC's 414 behalf. Being explicit prevents the remote entity from detecting 415 that is missing the reports if there issues with the reporting 416 SSRC's RTCP packets. 418 2. Enable explicit identification of the SSRCs that are actively 419 reporting as one entity. 421 8.3. RTCP Packet Reporting Group's Reporting Sources 423 This section defines a new RTCP packet type called "Reporting Group's 424 Reporting Sources" (RGRS). It identifies the SSRC(s) that report on 425 behalf of the SSRC that is the sender of the RGRS packet. 427 This packet consists of the fixed RTCP packet header which indicates 428 the packet type, the number of reporting sources included and the 429 SSRC which behalf the reporting SSRCs report on. This is followed by 430 the list of reporting SSRCs. 432 0 1 2 3 433 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 434 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 435 |V=2|P| SC | PT=RGRS(TBA) | length | 436 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 437 | SSRC of packet sender | 438 +=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+ 439 : SSRC for Reporting Source : 440 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 442 The RTCP Packets field has the following definition. 444 version (V): This field identifies the RTP version. The current 445 version is 2. 447 padding (P): 1 bit If set, the padding bit indicates that the packet 448 contains additional padding octets at the end that are not part of 449 the control information but are included in the length field. See 450 [RFC3550]. 452 Source Count (SC): 5 bits Indicating the number of reporting SSRCs 453 (1-31) that are include in this RTCP packet type. 455 Payload type (PT): 8 bits This is the RTCP packet type that 456 identifies the packet as being an RTCP FB message. The RGRS RTCP 457 packet has the value [TBA]. 459 Length: 16 bits The length of this packet in 32-bit words minus one, 460 including the header and any padding. This is in line with the 461 definition of the length field used in RTCP sender and receiver 462 reports [RFC3550]. 464 SSRC of packet sender: 32 bits. The SSRC of the sender of this 465 packet which indicates which SSRCs that reports on its behalf, 466 instead of reporting itself. 468 SSRC for Reporting Source: A variable number (as indicated by Source 469 Count) of 32-bit SSRC values. Each SSRC is an reporting SSRC 470 belonging to the same Report Group. 472 Each RGRS packet MUST contain at least one reporting SSRC. In case 473 the reporting SSRC field is insufficient to list all the SSRCs that 474 is reporting in this report group, the SSRC SHALL round robin around 475 the reporting sources. 477 Any RTP mixer or translator which forwards SR or RR packets from 478 members of a reporting group MUST forward the corresponding RGRS RTCP 479 packet as well. 481 8.4. RTCP Source Description (SDES) item for Reporting Groups 483 A new RTCP Source Description (SDES) item is defined for the purpose 484 of identifying reporting groups. 486 The Source Description (SDES) item "RGRP" is sent by a reporting 487 group's reporting SSRC. Syntactically, its format is the same as the 488 RTCP SDES CNAME item [RFC6222], and MUST be chosen with the same 489 global-uniqueness and privacy considerations as CNAME. This name 490 MUST be stable across the lifetime of the reporting group, even if 491 the SSRC of a reporting source changes. 493 Every source which belongs to a reporting group MUST either include 494 an RGRP SDES item in an SDES packet (if it is a reporting source), or 495 an RGRS packet (if it is not), in every compound RTCP packet in which 496 it sends an RR or SR packet (i.e., in every RTCP packet it sends, 497 unless Reduced-Size RTCP [RFC5506] is in use). 499 Any RTP mixer or translator which forwards SR or RR packets from 500 members of a reporting group MUST forward the corresponding RGRP SDES 501 items as well, even if it otherwise strips SDES items other than 502 CNAME. 504 8.5. Middlebox Considerations 506 This section discusses middlebox considerations for Reporting groups. 508 To be expanded. 510 8.6. SDP signaling for Reporting Groups 512 TBD 514 8.7. Bandwidth Benefits of RTCP Reporting Groups 516 To understand the benefits of RTCP reporting groups, consider the 517 scenario described in Section 4.1. This scenario describes an 518 environment in which the two endpoints in a session each have a 519 hundred sources, of which eight each are sending within any given 520 reporting interval. 522 For ease of analysis, we can make the simplifying approximation that 523 the duration of the RTCP reporting interval is equal to the total 524 size of the RTCP packets sent during an RTCP interval, divided by the 525 RTCP bandwidth. (This will be approximately true in scenarios where 526 the bandwidth is not so high that the minimum RTCP interval is 527 reached.) For further simplification, we can assume RTCP senders are 528 following the recommendations of Section 6.3; thus, the per-packet 529 transport-layer overhead will be small relative to the RTCP data. 530 Thus, only the actual RTCP data itself need be considered. 532 In a report interval in this scenario, there will, as a baseline, be 533 200 SDES packets, 184 RR packets, and 16 SR packets. This amounts to 534 approximately 6.5 kB of RTCP per report interval, assuming 16-byte 535 CNAMEs and no other SDES information. 537 Using the original [RFC3550] everyone-reports-on-every-sender 538 feedback rules, each of the 184 receivers will send 16 report blocks, 539 and each of the 16 senders will send 15. This amounts to 540 approximately 76 kB of report block traffic per interval; 92% of RTCP 541 traffic consists of report blocks. 543 If reporting groups are used, however, there is only 0.4 kB of 544 reports per interval, with no loss of useful information. 545 Additionally, there will be (assuming 16-byte RGRPs, and a single 546 reporting source per reporting group) an additional 2.4 kB per cycle 547 of RGRP SDES items and RGRS packets. Put another way, the unmodified 548 [RFC3550] reporting interval is approximately 8.9 times longer than 549 if reporting groups are in use. 551 8.8. Consequences of RTCP Reporting Groups 553 The RTCP traffic generated by receivers using RTCP Reporting Groups 554 might appear, to observers unaware of these semantics, to be 555 generated by receivers who are experiencing a network disconnection, 556 as the non-reporting sources appear not to be receiving a given 557 sender at all. 559 This could be a potentially critical problem for such a sender using 560 RTCP for congestion control, as such a sender might think that it is 561 sending so much traffic that it is causing complete congestion 562 collapse. 564 However, such an interpretation of the session statistics would 565 require a fairly sophisticated RTCP analysis. Any receiver of RTCP 566 statistics which is just interested in information about itself needs 567 to be prepared that any given reception report might not contain 568 information about a specific media source, because reception reports 569 in large conferences can be round-robined. 571 Thus, it is unclear to what extent such backward compatibility issues 572 would actually cause trouble in practice. 574 9. Security Considerations 576 In the secure RTP protocol (SRTP) [RFC3711], the cryptographic 577 context of a compound SRTCP packet is the SSRC of the sender of the 578 first RTCP (sub-)packet. This could matter in some cases, especially 579 for keying mechanisms such as Mikey [RFC3830] which use per-SSRC 580 keying. 582 Other than that, the standard security considerations of RTP apply; 583 sending multiple media streams from a single endpoint does not appear 584 to have different security consequences than sending the same number 585 of streams. 587 10. Open Issues 589 At this stage this document contains a number of open issues. The 590 below list tries to summarize the issues: 592 1. Further clarifications on how to handle the RTCP scheduler when 593 sending multiple sources in one compound packet. 595 2. How is the use of reporting groups be signaled in SDP? 597 3. How is the RTCP avg_rtcp_size be calculated when RTCP packets are 598 routinely multiplexed among multiple RTCP senders? 600 4. Do we need to provide a recommendation for unicast session 601 joiners with many sources to not use 0 initial minimal interval 602 from bit-rate burst perspective? 604 11. IANA Considerations 606 This document make several requests to IANA for registering new RTP/ 607 RTCP identifiers. 609 (Note to the RFC-Editor: please replace "TBA" with the IANA-assigned 610 value, and "XXXX" with the number of this document, prior to 611 publication as an RFC.) 613 11.1. RTCP SDES Item 615 This document adds an additional SDES types to the IANA "RTCP SDES 616 Item Types" Registry, as follows: 618 Value Abbrev Name Reference 619 TBA RGRP Reporting Group [RFCXXXX] 621 Figure 1: Item for the IANA Source Attribute Registry 623 11.2. RTCP Packet Type 625 This document defines one new RTCP Control Packet types (PT) to be 626 registered as follows: 628 Value Abbrev Name Reference 629 TBA RGRR Reporting Group Reporting Sources [RFCXXXX] 631 Figure 2: Item for the IANA RTCP Control Packet Types (PT) Registry 633 12. References 635 12.1. Normative References 637 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 638 Requirement Levels", BCP 14, RFC 2119, March 1997. 640 [RFC3550] Schulzrinne, H., Casner, S., Frederick, R., and V. 641 Jacobson, "RTP: A Transport Protocol for Real-Time 642 Applications", STD 64, RFC 3550, July 2003. 644 [RFC3711] Baugher, M., McGrew, D., Naslund, M., Carrara, E., and K. 645 Norrman, "The Secure Real-time Transport Protocol (SRTP)", 646 RFC 3711, March 2004. 648 [RFC4585] Ott, J., Wenger, S., Sato, N., Burmeister, C., and J. Rey, 649 "Extended RTP Profile for Real-time Transport Control 650 Protocol (RTCP)-Based Feedback (RTP/AVPF)", RFC 4585, July 651 2006. 653 [RFC5124] Ott, J. and E. Carrara, "Extended Secure RTP Profile for 654 Real-time Transport Control Protocol (RTCP)-Based Feedback 655 (RTP/SAVPF)", RFC 5124, February 2008. 657 [RFC5506] Johansson, I. and M. Westerlund, "Support for Reduced-Size 658 Real-Time Transport Control Protocol (RTCP): Opportunities 659 and Consequences", RFC 5506, April 2009. 661 [RFC6222] Begen, A., Perkins, C., and D. Wing, "Guidelines for 662 Choosing RTP Control Protocol (RTCP) Canonical Names 663 (CNAMEs)", RFC 6222, April 2011. 665 12.2. Informative References 667 [I-D.ietf-avtcore-multi-media-rtp-session] 668 Westerlund, M., Perkins, C., and J. Lennox, "Multiple 669 Media Types in an RTP Session", draft-ietf-avtcore-multi- 670 media-rtp-session-01 (work in progress), October 2012. 672 [I-D.ietf-clue-framework] 673 Duckworth, M., Pepperell, A., and S. Wenger, "Framework 674 for Telepresence Multi-Streams", draft-ietf-clue- 675 framework-09 (work in progress), February 2013. 677 [I-D.ietf-mmusic-sdp-bundle-negotiation] 678 Holmberg, C., Alvestrand, H., and C. Jennings, 679 "Multiplexing Negotiation Using Session Description 680 Protocol (SDP) Port Numbers", draft-ietf-mmusic-sdp- 681 bundle-negotiation-03 (work in progress), February 2013. 683 [I-D.westerlund-avtcore-rtp-topologies-update] 684 Westerlund, M. and S. Wenger, "RTP Topologies", draft- 685 westerlund-avtcore-rtp-topologies-update-01 (work in 686 progress), October 2012. 688 [RFC3611] Friedman, T., Caceres, R., and A. Clark, "RTP Control 689 Protocol Extended Reports (RTCP XR)", RFC 3611, November 690 2003. 692 [RFC3830] Arkko, J., Carrara, E., Lindholm, F., Naslund, M., and K. 693 Norrman, "MIKEY: Multimedia Internet KEYing", RFC 3830, 694 August 2004. 696 Appendix A. Changes From Earlier Versions 698 Note to the RFC-Editor: please remove this section prior to 699 publication as an RFC. 701 A.1. Changes From Draft -01 703 o Merged with draft-wu-avtcore-multisrc-endpoint-adver. 705 o Changed how Reporting Groups are indicated in RTCP, to make it 706 clear which source(s) is the group's reporting sources. 708 o Clarified the rules for when sources can be placed in the same 709 reporting group. 711 o Clarified that mixers and translators need to pass reporting group 712 SDES information if they are forwarding RR and SR traffic from 713 members of a reporting group. 715 A.2. Changes From Draft -00 717 o Added the Reporting Group semantic to explicitly indicate which 718 sources come from a single endpoint, rather than leaving it 719 implicit. 721 o Specified that Reporting Group semantics (as they now are) apply 722 to AVPF and XR, as well as to RR/SR report blocks. 724 o Added a description of the cascaded source-projecting mixer, along 725 with a calculation of its RTCP overhead if reporting groups are 726 not in use. 728 o Gave some guidance on how the flexibility of RTCP randomization 729 allows some freedom in RTCP multiplexing. 731 o Clarified the language of several of the recommendations. 733 o Added an open issue discussing how avg_rtcp_size ought to be 734 calculated for multiplexed RTCP. 736 o Added an open issue discussing how RTCP bandwidths are to be 737 chosen for sessions where source bandwidths greatly differ. 739 Authors' Addresses 741 Jonathan Lennox 742 Vidyo, Inc. 743 433 Hackensack Avenue 744 Seventh Floor 745 Hackensack, NJ 07601 746 US 748 Email: jonathan@vidyo.com 750 Magnus Westerlund 751 Ericsson 752 Farogatan 6 753 SE-164 80 Kista 754 Sweden 756 Phone: +46 10 714 82 87 757 Email: magnus.westerlund@ericsson.com 758 Qin Wu 759 Huawei 760 101 Software Avenue, Yuhua District 761 Nanjing, Jiangsu 210012 762 China 764 Email: sunseawq@huawei.com 766 Colin Perkins 767 University of Glasgow 768 School of Computing Science 769 Glasgow G12 8QQ 770 United Kingdom 772 Email: csp@csperkins.org