idnits 2.17.1 draft-ietf-xrblock-rtcweb-rtcp-xr-metrics-04.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 : ---------------------------------------------------------------------------- ** The document seems to lack an IANA Considerations section. (See Section 2.2 of https://www.ietf.org/id-info/checklist for how to handle the case when there are no actions for IANA.) Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (September 22, 2016) is 2773 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) == Unused Reference: 'RFC4588' is defined on line 624, but no explicit reference was found in the text == Unused Reference: 'RFC6390' is defined on line 719, but no explicit reference was found in the text == Outdated reference: A later version (-03) exists of draft-ietf-xrblock-independent-burst-gap-discard-00 == Outdated reference: A later version (-12) exists of draft-ietf-rtcweb-security-08 Summary: 1 error (**), 0 flaws (~~), 5 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 XR Block Working Group V. Singh 3 Internet-Draft callstats.io 4 Intended status: Standards Track R. Huang 5 Expires: March 26, 2017 R. Even 6 Huawei 7 D. Romascanu 8 Avaya 9 L. Deng 10 China Mobile 11 September 22, 2016 13 Considerations for Selecting RTCP Extended Report (XR) Metrics for the 14 WebRTC Statistics API 15 draft-ietf-xrblock-rtcweb-rtcp-xr-metrics-04 17 Abstract 19 This document describes monitoring features related to media streams 20 in Web real-time communication (WebRTC). It provides a list of RTCP 21 Sender Report, Receiver Report and Extended Report metrics, which may 22 need to be supported by RTP implementations in some diverse 23 environments. It lists a set of identifiers for the WebRTC's 24 statistics API. These identifiers are a set of RTCP SR, RR, and XR 25 metrics related to the transport of multimedia flows. 27 Status of This Memo 29 This Internet-Draft is submitted in full conformance with the 30 provisions of BCP 78 and BCP 79. 32 Internet-Drafts are working documents of the Internet Engineering 33 Task Force (IETF). Note that other groups may also distribute 34 working documents as Internet-Drafts. The list of current Internet- 35 Drafts is at http://datatracker.ietf.org/drafts/current/. 37 Internet-Drafts are draft documents valid for a maximum of six months 38 and may be updated, replaced, or obsoleted by other documents at any 39 time. It is inappropriate to use Internet-Drafts as reference 40 material or to cite them other than as "work in progress." 42 This Internet-Draft will expire on March 26, 2017. 44 Copyright Notice 46 Copyright (c) 2016 IETF Trust and the persons identified as the 47 document authors. All rights reserved. 49 This document is subject to BCP 78 and the IETF Trust's Legal 50 Provisions Relating to IETF Documents 51 (http://trustee.ietf.org/license-info) in effect on the date of 52 publication of this document. Please review these documents 53 carefully, as they describe your rights and restrictions with respect 54 to this document. Code Components extracted from this document must 55 include Simplified BSD License text as described in Section 4.e of 56 the Trust Legal Provisions and are provided without warranty as 57 described in the Simplified BSD License. 59 Table of Contents 61 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 62 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 3 63 3. RTP Statistics in WebRTC Implementations . . . . . . . . . . 3 64 4. Considerations for Impact of Measurement Interval . . . . . . 4 65 5. Candidate Metrics . . . . . . . . . . . . . . . . . . . . . . 5 66 5.1. Network Impact Metrics . . . . . . . . . . . . . . . . . 5 67 5.1.1. Loss and Discard Packet Count Metric . . . . . . . . 5 68 5.1.2. Burst/Gap Pattern Metrics for Loss and Discard . . . 6 69 5.1.3. Run Length Encoded Metrics for Loss, Discard . . . . 7 70 5.2. Application Impact Metrics . . . . . . . . . . . . . . . 7 71 5.2.1. Discard Octets Metric . . . . . . . . . . . . . . . . 7 72 5.2.2. Frame Impairment Summary Metrics . . . . . . . . . . 8 73 5.2.3. Jitter Buffer Metrics . . . . . . . . . . . . . . . . 8 74 5.3. Recovery metrics . . . . . . . . . . . . . . . . . . . . 9 75 5.3.1. Post-repair Packet Count Metrics . . . . . . . . . . 9 76 5.3.2. Run Length Encoded Metric for Post-repair . . . . . . 9 77 6. Identifiers from Sender, Receiver, and Extended Report Blocks 10 78 6.1. Cumulative Number of Packets and Octets Sent . . . . . . 10 79 6.2. Cumulative Number of Packets and Octets Received . . . . 10 80 6.3. Cumulative Number of Packets Lost . . . . . . . . . . . . 11 81 6.4. Interval Packet Loss and Jitter . . . . . . . . . . . . . 11 82 6.5. Cumulative Number of Packets and Octets Discarded . . . . 11 83 6.6. Cumulative Number of Packets Repaired . . . . . . . . . . 11 84 6.7. Burst Packet Loss and Burst Discards . . . . . . . . . . 11 85 6.8. Burst/Gap Rates . . . . . . . . . . . . . . . . . . . . . 12 86 6.9. Frame Impairment Metrics . . . . . . . . . . . . . . . . 12 87 7. Adding new metrics to WebRTC Statistics API . . . . . . . . . 13 88 8. Security Considerations . . . . . . . . . . . . . . . . . . . 13 89 9. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 13 90 10. References . . . . . . . . . . . . . . . . . . . . . . . . . 13 91 10.1. Normative References . . . . . . . . . . . . . . . . . . 13 92 10.2. Informative References . . . . . . . . . . . . . . . . . 15 93 Appendix A. Change Log . . . . . . . . . . . . . . . . . . . . . 16 94 A.1. changes in draft-ietf-xrblock-rtcweb-rtcp-xr-metrics-04 . 16 95 A.2. changes in draft-ietf-xrblock-rtcweb-rtcp-xr-metrics-02, 96 -03 . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 98 A.3. changes in draft-ietf-xrblock-rtcweb-rtcp-xr-metrics-01 . 16 99 A.4. changes in draft-ietf-xrblock-rtcweb-rtcp-xr-metrics-00 . 16 100 A.5. changes in draft-huang-xrblock-rtcweb-rtcp-xr-metrics-04 16 101 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 17 103 1. Introduction 105 Web real-time communication (WebRTC) deployments are emerging and 106 applications need to be able to estimate the service quality. If 107 sufficient information (metrics or statistics) are provided to the 108 applications, it can attempt to improve the media quality. [RFC7478] 109 specifies a requirement for statistics: 111 F38 The browser must be able to collect statistics, related to the 112 transport of audio and video between peers, needed to estimate 113 quality of experience. 115 The WebRTC Stats API [W3C.WD-webrtc-stats-20150206] currently lists 116 metrics reported in the RTCP Sender and Receiver Report (SR/RR) 117 [RFC3550] to fulfill this requirement. However, the basic metrics 118 from RTCP SR/RR are not sufficient for precise quality monitoring, or 119 diagnosing potential issues. 121 In this document, we provide rationale for choosing additional RTP 122 metrics for the WebRTC getStats() API [W3C.WD-webrtc-20150210]. The 123 document also creates a registry containing identifiers from the 124 metrics reported in the RTCP Sender, Receiver, and Extended Reports. 125 All identifiers proposed in this document are RECOMMENDED to be 126 implemented by an endpoint. An endpoint MAY choose not to expose an 127 identifier if it does not implement the corresponding RTCP Report. 129 2. Terminology 131 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 132 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 133 document are to be interpreted as described in [RFC2119]. 135 ReportGroup: It is a set of metrics identified by a common 136 Synchronization source (SSRC). 138 3. RTP Statistics in WebRTC Implementations 140 The RTCP Sender Reports (SRs) and Receiver Reports (RRs) [RFC3550] 141 exposes the basic metrics for the local and remote media streams. 142 However, these metrics provides only partial or limited information, 143 which may not be sufficient for diagnosing problems or quality 144 monitoring. For example, it may be useful to distinguish between 145 packets lost and packets discarded due to late arrival, even though 146 they have the same impact on the multimedia quality, it helps in 147 identifying and diagnosing issues. 149 RTP Control Protocol Extended Reports (XRs) [RFC3611] and other 150 extensions discussed in the XRBLOCK working group provide more 151 detailed statistics, which complement the basic metrics reported in 152 the RTCP SR and RRs. Section Section 5 discusses the use of XR 153 metrics that may be useful for monitoring the performance of WebRTC 154 applications. Sections Section 6 proposes a set of candidate 155 metrics. 157 The WebRTC application extracts the statistic from the browser by 158 querying the getStats() API [W3C.WD-webrtc-20150210], but the browser 159 currently only reports the local variables i.e., the statistics 160 related to the outgoing RTP media streams and the incoming RTP media 161 streams. Without the support of RTCP XRs or some other signaling 162 mechianism, the WebRTC application cannot expose the remote 163 endpoints' statistics. At the moment [I-D.ietf-rtcweb-rtp-usage] 164 does not mandate the use of any RTCP XRs and since their usage is 165 optional. If the use of RTCP XRs is successfully negotiated between 166 endpoints (via SDP), thereafter the application has access to both 167 local and remote statistics. Alternatively, once the WebRTC 168 application gets the local information, they can report it to an 169 application server or a third-party monitoring system, which provides 170 quality estimations or diagnosis services for application developers. 171 The exchange of statistics between endpoints or between a monitoring 172 server and an endpoint is outside the scope of this document. 174 4. Considerations for Impact of Measurement Interval 176 RTCP extensions like RTCP XR usually share the same timing interval 177 with the RTCP SR/RR, i.e., they are sent as compound packets, 178 together with the RTCP SR/RR. Alternatively, if the RTCP XR uses a 179 different measurement interval, all XRs using the same measurement 180 interval are compounded together and the measurement interval is 181 indicated in a specific measurement information block defined in 182 [RFC6776]. 184 When using WebRTC getStats() APIs (see section 7 of 185 [W3C.WD-webrtc-20150210]), the applications can query this 186 information at arbitrary intervals. For the statistics reported by 187 the remote endpoint, e.g., those conveyed in an RTCP SR/RR/XR, these 188 will not change until the next RTCP report is received. However, 189 statistics generated by the local endpoint have no such restrictions 190 as long as the endpoint is sending and receiving media. For example, 191 an application may choose to poll the stack for statistics every 1 192 second, in this case the underlying stack local will return the 193 current snapshot of the local statistics (for incoming and outgoing 194 media streams). However it may return the same remote statistics as 195 before for the remote statistics, as no new RTCP reports may have 196 been received in the past 1 second. This can occur when the polling 197 interval is shorter than the average RTCP reporting interval. 199 5. Candidate Metrics 201 Since following metrics are all defined in RTCP XR which is not 202 mandated in WebRTC, all of them are local. However, if RTCP XR is 203 supported by negotiation between two browsers, following metrics can 204 also be generated remotely and be sent to local by RTCP XR packets. 206 Following metrics are classified into 3 categories: network impact 207 metrics, application impact metrics and recovery metrics. Network 208 impact metrics are the statistics recording the information only for 209 network transmission. They are useful for network problem diagnosis. 210 Application impact metrics mainly collect the information in the 211 viewpoint of application, e.g., bit rate, frames rate or jitter 212 buffers. Recovery metrics reflect how well the repair mechanisms 213 perform, e.g. loss concealment, retransmission or FEC. All of the 3 214 types of metrics are useful for quality estimations of services in 215 WebRTC implementations. WebRTC application can use these metrics to 216 better calculate MoS values or Media Delivery Index (MDI) for their 217 services. 219 5.1. Network Impact Metrics 221 5.1.1. Loss and Discard Packet Count Metric 223 In multimedia transport, packets which are received abnormally are 224 classified into 3 types: lost, discarded and duplicate packets. 225 Packet loss may be caused by network device breakdown, bit-error 226 corruption or network congestion (packets dropped by an intermediate 227 router queue). Duplicate packets may be a result of network delays, 228 which causes the sender to retransmit the original packets. 229 Discarded packets are packets that have been delayed long enough 230 (perhaps they missed the playout time) and are considered useless by 231 the receiver. Lost and discarded packets cause problems for 232 multimedia services, as missing data and long delays can cause 233 degradation in service quality, e.g., missing large blocks of 234 contiguous packets (lost or discarded) may cause choppy audio, and 235 long network transmission delay time may cause audio or video 236 buffering. The RTCP SR/RR defines a metric for counting the total 237 number of RTP data packets that have been lost since the beginning of 238 reception. But this statistic does not distinguish lost packets from 239 discarded and duplicate packets. Packets that arrive late will be 240 discarded and are not reported as lost, and duplicate packets will be 241 regarded as a normally received packet. Hence, the loss metric can 242 be misleading if many duplicate packets are received or packets are 243 discarded, which causes the quality of the media transport to appear 244 okay from the statistic point of view, but meanwhile the users may 245 actually be experiencing bad service quality. So in such cases, it 246 is better to use more accurate metrics in addition to those defined 247 in RTCP SR/RR. 249 The lost packets and duplicated packets metrics defined in Statistics 250 Summary Report Block of [RFC3611] extend the information of loss 251 carried in standard RTCP SR/RR. They explicitly give an account of 252 lost and duplicated packets. Lost packets counts are useful for 253 network problem diagnosis. It is better to use the loss packets 254 metrics of [RFC3611] to indicate the packet lost count instead of the 255 cumulative number of packets lost metric of [RFC3550]. Duplicated 256 packets are usually rare and have little effect on QoS evaluation. 257 So it may not be suitable for use in WebRTC. 259 Using loss metrics without considering discard metrics may result in 260 inaccurate quality evaluation, as packet discard due to jitter is 261 often more prevalent than packet loss in modern IP networks. The 262 discarded metric specified in [RFC7002] counts the number of packets 263 discarded due to the jitter. It augments the loss statistics metrics 264 specified in standard RTCP SR/RR. For those RTCWEB services with 265 jitter buffer requiring precise quality evaluation and accurate 266 troubleshooting, this metric is useful as a complement to the metrics 267 of RTCP SR/RR. 269 5.1.2. Burst/Gap Pattern Metrics for Loss and Discard 271 RTCP SR/RR defines coarse metrics regarding loss statistics, the 272 metrics are all about per call statistics and are not detailed enough 273 to capture some transitory nature of the impairments like bursty 274 packet loss. Even if the average packet loss rate is low, the lost 275 packets may occur during short dense periods, resulting in short 276 periods of degraded quality. Distributed burst provides a higher 277 subjective quality than a non-burst distribution for low packet loss 278 rates whereas for high packet loss rates the converse is true. So 279 capturing burst gap information is very helpful for quality 280 evaluation and locating impairments. If the WebRTC application needs 281 to evaluate the services quality, burst gap metrics provides more 282 accurate information than RTCP SR/RR. 284 [RFC3611] introduces burst gap metrics in VoIP report block. These 285 metrics record the density and duration of burst and gap periods, 286 which are helpful in isolating network problems since bursts 287 correspond to periods of time during which the packet loss/discard 288 rate is high enough to produce noticeable degradation in audio or 289 video quality. Burst gap related metrics are also introduced in 291 [RFC7003] and [RFC6958] which define two new report blocks for usage 292 in a range of RTP applications beyond those described in [RFC3611]. 293 These metrics distinguish discarded packets from loss packets that 294 occur in the bursts period and provides more information for 295 diagnosing network problems. Additionally, the block reports the 296 frequency of burst events which is useful information for evaluating 297 the quality of experience. Hence, if WebRTC application need to do 298 quality evaluation and observe when and why quality degrades, these 299 metrics should be considered. 301 5.1.3. Run Length Encoded Metrics for Loss, Discard 303 Run-length encoding uses a bit vector to encode information about the 304 packet. Each bit in the vector represents a packet and depending on 305 the signaled metric it defines if the packet was lost, duplicated, 306 discarded, or repaired. An endpoint typically uses the run length 307 encoding to accurately communicate the status of each packet in the 308 interval to the other endpoint. [RFC3611], [RFC7097] define run- 309 length encoding for lost and duplicate packets, and discarded 310 packets, respectively. 312 The WebRTC application could benefit from the additional information. 313 If losses occur after discards, an endpoint may be able to correlate 314 the two run length vectors to identify congestion-related losses, 315 i.e., a router queue became overloaded causing delays and then 316 overflowed. If the losses are independent, it may indicate bit-error 317 corruption. For the WebRTC Stats API [W3C.WD-webrtc-stats-20150206], 318 these types of metrics are not recommended for use due to the large 319 amount of data and the computation involved. 321 5.2. Application Impact Metrics 323 5.2.1. Discard Octets Metric 325 The metric reports the cumulative size of the packets discarded in 326 the interval, it is complementary to number of discarded packets. An 327 application measures sent octets and received octets to calculate 328 sending rate and receiving rate, respectively. The application can 329 calculate the actual bit rate in a particular interval by subtracting 330 the discarded octets from the received octets. 332 For WebRTC, discarded octets supplements the sent and received octets 333 and provides an accurate method for calculating the actual bit rate 334 which is an important parameter to reflect the quality of the media. 335 The discarded bytes metric is defined in [RFC7243]. 337 5.2.2. Frame Impairment Summary Metrics 339 RTP has different framing mechanisms for different payload types. 340 For audio streams, a single RTP packet may contain one or multiple 341 audio frames, each of which has a fixed length. On the other hand, 342 in video streams, a single video frame may be transmitted in multiple 343 RTP packets. The size of each packet is limited by the Maximum 344 Transmission Unit (MTU) of the underlying network. However, 345 statistics from standard SR/RR only collect information from 346 transport layer, which may not fully reflect the quality observed by 347 the application. Video is typically encoded using two frame types 348 i.e., key frames and derived frames. Key frames are normally just 349 spatially compressed, i.e., without prediction from other pictures. 350 The derived frames are temporally compressed, i.e., depend on the key 351 frame for decoding. Hence, key frames are much larger in size than 352 derived frames. The loss of these key frames results in a 353 substantial reduction in video quality. Thus it is reasonable to 354 consider this application layer information in WebRTC 355 implementations, which influence sender strategies to mitigate the 356 problem or require the accurate assessment of users' quality of 357 experience. 359 The following metrics can also be considered for WebRTC's Statistics 360 API: number of discarded key frames, number of lost key frames, 361 number of discarded derived frames, number of lost derived frames. 362 These metrics can be used to calculate Media Loss Rate (MLR) of MDI. 363 Details of the definition of these metrics are described in 364 [RFC7003]. Additionally, the metric provides the rendered frame 365 rate, an important parameter for quality estimation. 367 5.2.3. Jitter Buffer Metrics 369 The size of the jitter buffer affects the end-to-end delay on the 370 network and also the packet discard rate. When the buffer size is 371 too small, slower packets are not played out and dropped, while when 372 the buffer size is too large, packets are held longer than necessary 373 and consequently reduce conversational quality. Measurement of 374 jitter buffer should not be ignored in the evaluation of end user 375 perception of conversational quality. Jitter buffer related metrics, 376 such as maximum and nominal jitter buffer, could be used to show how 377 the jitter buffer behaves at the receiving endpoint. They are useful 378 for providing better end-user quality of experience (QoE) when jitter 379 buffer factors are used as inputs to calculate MoS values. Thus for 380 those cases, jitter buffer metrics should be considered. The 381 definition of these metrics is provided in [RFC7005]. 383 5.3. Recovery metrics 385 This document does not consider concealment metrics as part of 386 recovery metrics. 388 5.3.1. Post-repair Packet Count Metrics 390 Error-resilience mechanisms, like RTP retransmission or FEC, are 391 optional in RTCWEB because the overhead of the repair bits adding to 392 the original streams. But they do help to greatly reduce the impact 393 of packet loss and enhance the quality of transmission. Web 394 applications could support certain repair mechanism after negotiation 395 between both sides of browsers when needed. For these web 396 applications using repair mechanisms, providing some statistic 397 information for the performance of their repair mechanisms could help 398 to have a more accurate quality evaluation. 400 The un-repaired packets count and repaired loss count defined in 401 [RFC7509] provide the recovery information of the error-resilience 402 mechanisms to the monitoring application or the sending endpoint. 403 The endpoint can use these metrics to ascertain the ratio of repaired 404 packets to lost packets. Including this kind of metrics helps the 405 application evaluate the effectiveness of the applied repair 406 mechanisms. 408 5.3.2. Run Length Encoded Metric for Post-repair 410 [RFC5725] defines run-length encoding for post-repair packets. When 411 using error-resilience mechanisms, the endpoint can correlate the 412 loss run length with this metric to ascertain where the losses and 413 repairs occurred in the interval. This provides more accurate 414 information for recovery mechanisms evaluation than those in 415 Section 5.3.1. However, it is not suggested to use due to their 416 enormous amount of data when RTCP XR are supported. 418 For WebRTC, the application may benefit from the additional 419 information. If losses occur after discards, an endpoint may be able 420 to correlate the two run length vectors to identify congestion- 421 related losses, i.e., a router queue became overloaded causing delays 422 and then overflowed. If the losses are independent, it may indicate 423 bit-error corruption. Lastly, when using error-resilience 424 mechanisms, the endpoint can correlate the loss and post-repair run 425 lengths to ascertain where the losses and repairs occurred in the 426 interval. For example, consecutive losses are likely not to be 427 repaired by a simple FEC scheme. 429 6. Identifiers from Sender, Receiver, and Extended Report Blocks 431 This document describes a list of metrics and corresponding 432 identifiers relevant to RTP media in WebRTC. These group of 433 identifiers are defined on a ReportGroup corresponding to an 434 Synchronization source (SSRC). In practice the application MUST be 435 able to query the statistic identifiers on both an incoming (remote) 436 and outgoing (local) media stream. Since sending and receiving SR 437 and RR are mandatory, the metrics defined in the SR and RR report 438 blocks are always available. For XR metrics, it depends on two 439 factors: 1) if it measured at the endpoint, 2) if it reported by the 440 endpoint in an XR report. If a metric is only measured by the 441 endpoint and not reported, the metrics will only be available for the 442 incoming (remote) media stream. Alternatively, if the corresponding 443 metric is also reported in an XR report, it will be available for 444 both the incoming (remote) and outgoing (local) media stream. 446 For a remote statistic, the timestamp represents the timestamp from 447 an incoming SR/RR/XR packet. Conversely, for a local statistic, it 448 refers to the current timestamp generated by the local clock 449 (typically the POSIX timestamp, i.e., milliseconds since Jan 1, 450 1970). 452 As per [RFC3550], the octets metrics represent the payload size 453 (i.e., not including header or padding). 455 6.1. Cumulative Number of Packets and Octets Sent 457 Name: packetsSent 459 Definition: section 6.4.1 in [RFC3550]. 461 Name: bytesSent 463 Definition: section 6.4.1 in [RFC3550]. 465 6.2. Cumulative Number of Packets and Octets Received 467 Name: packetsReceived 469 Definition: section 6.4.1 in [RFC3550]. 471 Name: bytesReceived 473 Definition: section 6.4.1 in [RFC3550]. 475 6.3. Cumulative Number of Packets Lost 477 Name: packetsLost 479 Definition: section 6.4.1 in [RFC3550]. 481 6.4. Interval Packet Loss and Jitter 483 Name: jitter 485 Definition: section 6.4.1 in [RFC3550]. 487 Name: fractionLost 489 Definition: section 6.4.1 in [RFC3550]. 491 6.5. Cumulative Number of Packets and Octets Discarded 493 Name: packetsDiscarded 495 Definition: The cumulative number of RTP packets discarded due to 496 late or early-arrival, Appendix A (a) of [RFC7002]. 498 Name: bytesDiscarded 500 Definition: The cumulative number of octets discarded due to late or 501 early-arrival, Appendix A of [RFC7243]. 503 6.6. Cumulative Number of Packets Repaired 505 Name: packetsRepaired 507 Definition: The cumulative number of lost RTP packets repaired after 508 applying a error-resilience mechanism, Appendix A (b) of [RFC7509]. 509 To clarify, the value is upper bound to the cumulative number of lost 510 packets. 512 6.7. Burst Packet Loss and Burst Discards 514 Name: burstPacketsLost 516 Definition: The cumulative number of RTP packets lost during loss 517 bursts, Appendix A (c) of [RFC6958]. 519 Name: burstLossCount 521 Definition: The cumulative number of bursts of lost RTP packets, 522 Appendix A (e) of [RFC6958]. 524 Name: burstPacketsDiscarded 526 Definition: The cumulative number of RTP packets discarded during 527 discard bursts, Appendix A (b) of [RFC7003]. 529 Name: burstDiscardCount 531 Definition: The cumulative number of bursts of discarded RTP packets, 532 Appendix A (e) of [I-D.ietf-xrblock-independent-burst-gap-discard]. 534 [RFC3611] recommends a Gmin (threshold) value of 16 for classifying 535 packet loss or discard burst. 537 6.8. Burst/Gap Rates 539 Name: burstLossRate 541 Definition: The fraction of RTP packets lost during bursts, 542 Appendix A (a) of [RFC7004]. 544 Name: gapLossRate 546 Definition: The fraction of RTP packets lost during gaps, Appendix A 547 (b) of [RFC7004]. 549 Name: burstDiscardRate 551 Definition: The fraction of RTP packets discarded during bursts, 552 Appendix A (e) of [RFC7004]. 554 Name: gapDiscardRate 556 Definition: The fraction of RTP packets discarded during gaps, 557 Appendix A (f) of [RFC7004]. 559 6.9. Frame Impairment Metrics 561 Name: framesLost 563 Definition: The cumulative number of full frames lost, Appendix A (i) 564 of [RFC7004]. 566 Name: framesCorrupted 568 Definition: The cumulative number of frames partially lost, 569 Appendix A (j) of [RFC7004]. 571 Name: framesDropped 572 Definition: The cumulative number of full frames discarded, 573 Appendix A (g) of [RFC7004]. 575 Name: framesSent 577 Definition: The cumulative number of frames sent. 579 Name: framesReceived 581 Definition: The cumulative number of partial or full frames received. 583 7. Adding new metrics to WebRTC Statistics API 585 The metrics defined in this draft have already been added to the W3C 586 WebRTC specification. The current working process to add new metrics 587 is, create an issue or pull request on the repository of the W3C 588 WebRTC specification (https://github.com/w3c/webrtc-stats). 590 8. Security Considerations 592 The monitoring activities are implemented between two browsers or 593 between a browser and a server. Therefore encryption procedures, 594 such as the ones suggested for a Secure RTCP (SRTCP), need to be 595 used. Currently, the monitoring in RTCWEB introduces no new security 596 considerations beyond those described in [I-D.ietf-rtcweb-rtp-usage], 597 [I-D.ietf-rtcweb-security]. 599 9. Acknowledgements 601 The authors would like to thank Bernard Aboba, Harald Alvestrand, Al 602 Morton, Colin Perkins, and Shida Schubert for their valuable comments 603 and suggestions on earlier version of this document. 605 10. References 607 10.1. Normative References 609 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 610 Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/ 611 RFC2119, March 1997, 612 . 614 [RFC3550] Schulzrinne, H., Casner, S., Frederick, R., and V. 615 Jacobson, "RTP: A Transport Protocol for Real-Time 616 Applications", STD 64, RFC 3550, DOI 10.17487/RFC3550, 617 July 2003, . 619 [RFC3611] Friedman, T., Ed., Caceres, R., Ed., and A. Clark, Ed., 620 "RTP Control Protocol Extended Reports (RTCP XR)", RFC 621 3611, DOI 10.17487/RFC3611, November 2003, 622 . 624 [RFC4588] Rey, J., Leon, D., Miyazaki, A., Varsa, V., and R. 625 Hakenberg, "RTP Retransmission Payload Format", RFC 4588, 626 DOI 10.17487/RFC4588, July 2006, 627 . 629 [RFC5725] Begen, A., Hsu, D., and M. Lague, "Post-Repair Loss RLE 630 Report Block Type for RTP Control Protocol (RTCP) Extended 631 Reports (XRs)", RFC 5725, DOI 10.17487/RFC5725, February 632 2010, . 634 [RFC6776] Clark, A. and Q. Wu, "Measurement Identity and Information 635 Reporting Using a Source Description (SDES) Item and an 636 RTCP Extended Report (XR) Block", RFC 6776, DOI 10.17487/ 637 RFC6776, October 2012, 638 . 640 [RFC6958] Clark, A., Zhang, S., Zhao, J., and Q. Wu, Ed., "RTP 641 Control Protocol (RTCP) Extended Report (XR) Block for 642 Burst/Gap Loss Metric Reporting", RFC 6958, DOI 10.17487/ 643 RFC6958, May 2013, 644 . 646 [RFC7002] Clark, A., Zorn, G., and Q. Wu, "RTP Control Protocol 647 (RTCP) Extended Report (XR) Block for Discard Count Metric 648 Reporting", RFC 7002, DOI 10.17487/RFC7002, September 649 2013, . 651 [RFC7003] Clark, A., Huang, R., and Q. Wu, Ed., "RTP Control 652 Protocol (RTCP) Extended Report (XR) Block for Burst/Gap 653 Discard Metric Reporting", RFC 7003, DOI 10.17487/RFC7003, 654 September 2013, . 656 [RFC7004] Zorn, G., Schott, R., Wu, Q., Ed., and R. Huang, "RTP 657 Control Protocol (RTCP) Extended Report (XR) Blocks for 658 Summary Statistics Metrics Reporting", RFC 7004, DOI 659 10.17487/RFC7004, September 2013, 660 . 662 [RFC7005] Clark, A., Singh, V., and Q. Wu, "RTP Control Protocol 663 (RTCP) Extended Report (XR) Block for De-Jitter Buffer 664 Metric Reporting", RFC 7005, DOI 10.17487/RFC7005, 665 September 2013, . 667 [RFC7097] Ott, J., Singh, V., Ed., and I. Curcio, "RTP Control 668 Protocol (RTCP) Extended Report (XR) for RLE of Discarded 669 Packets", RFC 7097, DOI 10.17487/RFC7097, January 2014, 670 . 672 [RFC7243] Singh, V., Ed., Ott, J., and I. Curcio, "RTP Control 673 Protocol (RTCP) Extended Report (XR) Block for the Bytes 674 Discarded Metric", RFC 7243, DOI 10.17487/RFC7243, May 675 2014, . 677 [RFC7509] Huang, R. and V. Singh, "RTP Control Protocol (RTCP) 678 Extended Report (XR) for Post-Repair Loss Count Metrics", 679 RFC 7509, DOI 10.17487/RFC7509, May 2015, 680 . 682 [I-D.ietf-xrblock-independent-burst-gap-discard] 683 Singh, V., Perkins, C., Clark, A., and R. Huang, "RTP 684 Control Protocol (RTCP) Extended Report (XR) Block for 685 Independent Reporting of Burst/Gap Discard Metric", draft- 686 ietf-xrblock-independent-burst-gap-discard-00 (work in 687 progress), December 2015. 689 10.2. Informative References 691 [RFC7478] Holmberg, C., Hakansson, S., and G. Eriksson, "Web Real- 692 Time Communication Use Cases and Requirements", RFC 7478, 693 DOI 10.17487/RFC7478, March 2015, 694 . 696 [W3C.WD-webrtc-20150210] 697 Bergkvist, A., Burnett, D., Jennings, C., and A. 698 Narayanan, "WebRTC 1.0: Real-time Communication Between 699 Browsers", World Wide Web Consortium WD WD-webrtc- 700 20150210, February 2015, 701 . 703 [I-D.ietf-rtcweb-rtp-usage] 704 Perkins, C., Westerlund, M., and J. Ott, "Web Real-Time 705 Communication (WebRTC): Media Transport and Use of RTP", 706 draft-ietf-rtcweb-rtp-usage-26 (work in progress), March 707 2016. 709 [I-D.ietf-rtcweb-security] 710 Rescorla, E., "Security Considerations for WebRTC", draft- 711 ietf-rtcweb-security-08 (work in progress), February 2015. 713 [W3C.WD-webrtc-stats-20150206] 714 Alvestrand, H. and V. Singh, "Identifiers for 715 WebRTC's Statistics API", World Wide Web Consortium 716 WD WD-webrtc-stats-20150206, February 2015, 717 . 719 [RFC6390] Clark, A. and B. Claise, "Guidelines for Considering New 720 Performance Metric Development", BCP 170, RFC 6390, DOI 721 10.17487/RFC6390, October 2011, 722 . 724 Appendix A. Change Log 726 Note to the RFC-Editor: please remove this section prior to 727 publication as an RFC. 729 A.1. changes in draft-ietf-xrblock-rtcweb-rtcp-xr-metrics-04 731 o Removed IANA registry. 733 A.2. changes in draft-ietf-xrblock-rtcweb-rtcp-xr-metrics-02, -03 735 o Keep-alive versions, updates to references. 737 A.3. changes in draft-ietf-xrblock-rtcweb-rtcp-xr-metrics-01 739 o Create new registry for WebRTC media metrics. 741 o Using camelCase instead of TitleCase for identifier names. 743 o Imported RTCP SR and RR metrics from the registry in alvestrand- 744 rtcweb-stats-registry. 746 o Added Burst/Gap rate metrics. 748 o Added Frames sent and received metrics. 750 A.4. changes in draft-ietf-xrblock-rtcweb-rtcp-xr-metrics-00 752 o Submitted as WG Draft. 754 A.5. changes in draft-huang-xrblock-rtcweb-rtcp-xr-metrics-04 756 o Addressed comments from the London IETF meeting: 758 o Removed ECN metrics. 760 o Merged draft-singh-xrblock-webrtc-additional-stats-01 762 Authors' Addresses 764 Varun Singh 765 CALLSTATS I/O Oy 766 Runeberginkatu 4c A 4 767 Helsinki 00100 768 Finland 770 Email: varun@callstats.io 771 URI: https://www.callstats.io/about 773 Rachel Huang 774 Huawei 775 101 Software Avenue, Yuhua District 776 Nanjing, CN 210012 777 China 779 Email: rachel.huang@huawei.com 781 Roni Even 782 Huawei 783 14 David Hamelech 784 Tel Aviv 64953 785 Israel 787 Email: roni.even@mail01.huawei.com 789 Dan Romascanu 790 Avaya 791 Park Atidim, Bldg. #3 792 Tel Aviv 61581 793 Israel 795 Email: dromasca@avaya.com 797 Lingli Deng 798 China Mobile 800 Email: denglingli@chinamobile.com