idnits 2.17.1 draft-ietf-rmcat-eval-criteria-05.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 (March 21, 2016) is 2956 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Unused Reference: 'RFC5681' is defined on line 525, but no explicit reference was found in the text == Unused Reference: 'SA4-EVAL' is defined on line 539, but no explicit reference was found in the text == Unused Reference: 'SA4-LR' is defined on line 543, but no explicit reference was found in the text == Unused Reference: 'TCP-eval-suite' is defined on line 546, but no explicit reference was found in the text == Outdated reference: A later version (-18) exists of draft-ietf-avtcore-rtp-circuit-breakers-14 == Outdated reference: A later version (-11) exists of draft-ietf-rmcat-wireless-tests-01 == Outdated reference: A later version (-10) exists of draft-ietf-rmcat-eval-test-03 == Outdated reference: A later version (-07) exists of draft-ietf-rmcat-video-traffic-model-00 Summary: 0 errors (**), 0 flaws (~~), 9 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 RMCAT WG V. Singh 3 Internet-Draft callstats.io 4 Intended status: Informational J. Ott 5 Expires: September 22, 2016 Technical University of Munich 6 S. Holmer 7 Google 8 March 21, 2016 10 Evaluating Congestion Control for Interactive Real-time Media 11 draft-ietf-rmcat-eval-criteria-05 13 Abstract 15 The Real-time Transport Protocol (RTP) is used to transmit media in 16 telephony and video conferencing applications. This document 17 describes the guidelines to evaluate new congestion control 18 algorithms for interactive point-to-point real-time media. 20 Status of This Memo 22 This Internet-Draft is submitted in full conformance with the 23 provisions of BCP 78 and BCP 79. 25 Internet-Drafts are working documents of the Internet Engineering 26 Task Force (IETF). Note that other groups may also distribute 27 working documents as Internet-Drafts. The list of current Internet- 28 Drafts is at http://datatracker.ietf.org/drafts/current/. 30 Internet-Drafts are draft documents valid for a maximum of six months 31 and may be updated, replaced, or obsoleted by other documents at any 32 time. It is inappropriate to use Internet-Drafts as reference 33 material or to cite them other than as "work in progress." 35 This Internet-Draft will expire on September 22, 2016. 37 Copyright Notice 39 Copyright (c) 2016 IETF Trust and the persons identified as the 40 document authors. All rights reserved. 42 This document is subject to BCP 78 and the IETF Trust's Legal 43 Provisions Relating to IETF Documents 44 (http://trustee.ietf.org/license-info) in effect on the date of 45 publication of this document. Please review these documents 46 carefully, as they describe your rights and restrictions with respect 47 to this document. Code Components extracted from this document must 48 include Simplified BSD License text as described in Section 4.e of 49 the Trust Legal Provisions and are provided without warranty as 50 described in the Simplified BSD License. 52 Table of Contents 54 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 55 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 3 56 3. Metrics . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 57 3.1. RTP Log Format . . . . . . . . . . . . . . . . . . . . . 5 58 4. List of Network Parameters . . . . . . . . . . . . . . . . . 5 59 4.1. One-way Propagation Delay . . . . . . . . . . . . . . . . 5 60 4.2. End-to-end Loss . . . . . . . . . . . . . . . . . . . . . 5 61 4.3. DropTail Router Queue Length . . . . . . . . . . . . . . 6 62 4.4. Loss generation model . . . . . . . . . . . . . . . . . . 6 63 4.5. Jitter models . . . . . . . . . . . . . . . . . . . . . . 6 64 4.5.1. Random Bounded PDV (RBPDV) . . . . . . . . . . . . . 7 65 4.5.2. Approximately Random Subject to No-Reordering Bounded 66 PDV (NR-RPVD) . . . . . . . . . . . . . . . . 8 67 5. WiFi or Cellular Links . . . . . . . . . . . . . . . . . . . 9 68 6. Traffic Models . . . . . . . . . . . . . . . . . . . . . . . 9 69 6.1. TCP taffic model . . . . . . . . . . . . . . . . . . . . 9 70 6.2. RTP Video model . . . . . . . . . . . . . . . . . . . . . 9 71 6.3. Background UDP . . . . . . . . . . . . . . . . . . . . . 10 72 7. Security Considerations . . . . . . . . . . . . . . . . . . . 10 73 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 10 74 9. Contributors . . . . . . . . . . . . . . . . . . . . . . . . 10 75 10. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 10 76 11. References . . . . . . . . . . . . . . . . . . . . . . . . . 10 77 11.1. Normative References . . . . . . . . . . . . . . . . . . 10 78 11.2. Informative References . . . . . . . . . . . . . . . . . 11 79 Appendix A. Application Trade-off . . . . . . . . . . . . . . . 12 80 A.1. Measuring Quality . . . . . . . . . . . . . . . . . . . . 12 81 Appendix B. Change Log . . . . . . . . . . . . . . . . . . . . . 13 82 B.1. Changes in draft-ietf-rmcat-eval-criteria-05 . . . . . . 13 83 B.2. Changes in draft-ietf-rmcat-eval-criteria-04 . . . . . . 13 84 B.3. Changes in draft-ietf-rmcat-eval-criteria-03 . . . . . . 13 85 B.4. Changes in draft-ietf-rmcat-eval-criteria-02 . . . . . . 13 86 B.5. Changes in draft-ietf-rmcat-eval-criteria-01 . . . . . . 13 87 B.6. Changes in draft-ietf-rmcat-eval-criteria-00 . . . . . . 13 88 B.7. Changes in draft-singh-rmcat-cc-eval-04 . . . . . . . . . 13 89 B.8. Changes in draft-singh-rmcat-cc-eval-03 . . . . . . . . . 14 90 B.9. Changes in draft-singh-rmcat-cc-eval-02 . . . . . . . . . 14 91 B.10. Changes in draft-singh-rmcat-cc-eval-01 . . . . . . . . . 14 92 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 14 94 1. Introduction 96 This memo describes the guidelines to help with evaluating new 97 congestion control algorithms for interactive point-to-point real 98 time media. The requirements for the congestion control algorithm 99 are outlined in [I-D.ietf-rmcat-cc-requirements]). This document 100 builds upon previous work at the IETF: Specifying New Congestion 101 Control Algorithms [RFC5033] and Metrics for the Evaluation of 102 Congestion Control Algorithms [RFC5166]. 104 The guidelines proposed in the document are intended to help prevent 105 a congestion collapse, promote fair capacity usage and optimize the 106 media flow's throughput. Furthermore, the proposed algorithms are 107 expected to operate within the envelope of the circuit breakers 108 defined in [I-D.ietf-avtcore-rtp-circuit-breakers]. 110 This document only provides broad-level criteria for evaluating a new 111 congestion control algorithm. The minimal requirement for RMCAT 112 proposals is to produce or present results for the test scenarios 113 described in [I-D.ietf-rmcat-eval-test] (Basic Test Cases). 114 Additionally, proponents may produce evaluation results for the 115 wireless test scenarios [I-D.ietf-rmcat-wireless-tests]. 117 2. Terminology 119 The terminology defined in RTP [RFC3550], RTP Profile for Audio and 120 Video Conferences with Minimal Control [RFC3551], RTCP Extended 121 Report (XR) [RFC3611], Extended RTP Profile for RTCP-based Feedback 122 (RTP/AVPF) [RFC4585] and Support for Reduced-Size RTCP [RFC5506] 123 apply. 125 3. Metrics 127 Each experiment is expected to log every incoming and outgoing packet 128 (the RTP logging format is described in Section 3.1). The logging 129 can be done inside the application or at the endpoints using PCAP 130 (packet capture, e.g., tcpdump, wireshark). The following are 131 calculated based on the information in the packet logs: 133 1. Sending rate, Receiver rate, Goodput (measured at 200ms 134 intervals) 136 2. Packets sent, Packets received 138 3. Bytes sent, bytes received 140 4. Packet delay 141 5. Packets lost, Packets discarded (from the playout or de-jitter 142 buffer) 144 6. If using, retransmission or FEC: post-repair loss 146 7. Fairness or Unfairness: Experiments testing the performance of 147 an RMCAT proposal against any cross-traffic must define its 148 expected criteria for fairness. The "unfairness" test guideline 149 (measured at 1s intervals) is: 150 1. Does not trigger the circuit breaker. 151 2. No RMCAT stream achieves more than 3 times the average 152 throughput of the RMCAT stream with the lowest average 153 throughput, for a case when the competing streams have similar 154 RTTs. 155 3. RTT should not grow by a factor of 3 for the existing flows 156 when a new flow is added. 157 For example, see the test scenarios described in 158 [I-D.ietf-rmcat-eval-test]. 160 8. Convergence time: The time taken to reach a stable rate at 161 startup, after the available link capacity changes, or when new 162 flows get added to the bottleneck link. 164 9. Instability or oscillation in the sending rate: The frequency or 165 number of instances when the sending rate oscillates between an 166 high watermark level and a low watermark level, or vice-versa in 167 a defined time window. For example, the watermarks can be set 168 at 4x interval: 500 Kbps, 2 Mbps, and a time window of 500ms. 170 10. Bandwidth Utilization, defined as ratio of the instantaneous 171 sending rate to the instantaneous bottleneck capacity. This 172 metric is useful only when an RMCAT flow is by itself or 173 competing with similar cross-traffic. 175 From the logs the statistical measures (min, max, mean, standard 176 deviation and variance) for the whole duration or any specific part 177 of the session can be calculated. Also the metrics (sending rate, 178 receiver rate, goodput, latency) can be visualized in graphs as 179 variation over time, the measurements in the plot are at 1 second 180 intervals. Additionally, from the logs it is possible to plot the 181 histogram or CDF of packet delay. 183 [Open issue (1): Using Jain-fairness index (JFI) for measuring self- 184 fairness between RTP flows? measured at what intervals? visualized as 185 a CDF or a timeseries? Additionally: Use JFI for comparing fairness 186 between RTP and long TCP flows? ] 188 3.1. RTP Log Format 190 The log file is tab or comma separated containing the following 191 details: 193 Send or receive timestamp (unix) 194 RTP payload type 195 SSRC 196 RTP sequence no 197 RTP timestamp 198 marker bit 199 payload size 201 If the congestion control implements, retransmissions or FEC, the 202 evaluation should report both packet loss (before applying error- 203 resilience) and residual packet loss (after applying error- 204 resilience). 206 4. List of Network Parameters 208 The implementors initially are encouraged to choose evaluation 209 settings from the following values: 211 4.1. One-way Propagation Delay 213 Experiments are expected to verify that the congestion control is 214 able to work in challenging situations, for example over trans- 215 continental and/or satellite links. Typical values are: 217 1. Very low latency: 0-1ms 219 2. Low latency: 50ms 221 3. High latency: 150ms 223 4. Extreme latency: 300ms 225 4.2. End-to-end Loss 227 To model lossy links, the experiments can choose one of the following 228 loss rates, the fractional loss is the ratio of packets lost and 229 packets sent. 231 1. no loss: 0% 233 2. 1% 235 3. 5% 236 4. 10% 238 5. 20% 240 4.3. DropTail Router Queue Length 242 The router queue length is measured as the time taken to drain the 243 FIFO queue. It has been noted in various discussions that the queue 244 length in the current deployed Internet varies significantly. While 245 the core backbone network has very short queue length, the home 246 gateways usually have larger queue length. Those various queue 247 lengths can be categorized in the following way: 249 1. QoS-aware (or short): 70ms 251 2. Nominal: 300-500ms 253 3. Buffer-bloated: 1000-2000ms 255 Here the size of the queue is measured in bytes or packets and to 256 convert the queue length measured in seconds to queue length in 257 bytes: 259 QueueSize (in bytes) = QueueSize (in sec) x Throughput (in bps)/8 261 4.4. Loss generation model 263 [Open Issue: Describes the model for generating packet losses, for 264 example, losses can be generated using traces, or using the Gilbert- 265 Elliot model, or randomly (uncorrelated loss).] 267 4.5. Jitter models 269 This section defines jitter model for the purposes of this document. 270 When jitter is to be applied to both the RMCAT flow and any competing 271 flow (such as a TCP competing flow), the competing flow will use the 272 jitter definition below that does not allow for re-ordering of 273 packets on the competing flow (see NR-RBPDV definition below). 275 Jitter is an overloaded term in communications. Its meaning is 276 typically associated with the variation of a metric (e.g., delay) 277 with respect to some reference metric (e.g., average delay or minimum 278 delay). For example, RFC 3550 jitter is a smoothed estimate of 279 jitter which is particularly meaningful if the underlying packet 280 delay variation was caused by a Gaussian random process. 282 Because jitter is an overloaded term, we instead use the term Packet 283 Delay Variation (PDV) to describe the variation of delay of 284 individual packets in the same sense as the IETF IPPM WG has defined 285 PDV in their documents (e.g., RFC 3393) and as the ITU-T SG16 has 286 defined IP Packet Delay Variation (IPDV) in their documents (e.g., 287 Y.1540). 289 Most PDV distributions in packet network systems are one-sided 290 distributions (the measurement of which with a finite number of 291 measurement samples result in one-sided histograms). In the usual 292 packet network transport case there is typically one packet that 293 transited the network with the minimum delay, then a majority of 294 packets also transit the system within some variation from this 295 minimum delay, and then a minority of the packets transits the 296 network with delays higher than the median or average transit time 297 (these are outliers). Although infrequent, outliers can cause 298 significant deleterious operation in adaptive systems and should be 299 considered in RMCAT adaptation designs. 301 In this section we define two different bounded PDV characteristics, 302 1) Random Bounded PDV and 2) Approximately Random Subject to No- 303 Reordering Bounded PDV. 305 [Open issue: which one is used in evaluations? Are both used?] 307 4.5.1. Random Bounded PDV (RBPDV) 309 The RBPDV probability distribution function (pdf) is specified to be 310 of some mathematically describable function which includes some 311 practical minimum and maximum discrete values suitable for testing. 312 For example, the minimum value, x_min, might be specified as the 313 minimum transit time packet and the maximum value, x_max, might be 314 idefined to be two standard deviations higher than the mean. 316 Since we are typically interested in the distribution relative to the 317 mean delay packet, we define the zero mean PVD sample, z(n), to be 318 z(n) = x(n) - x_mean, where x(n) is a sample of the RBPDV random 319 variable x and x_mean is the mean of x. 321 We assume here that s(n) is the original source time of packet n and 322 the post-jitter induced emmission time, j(n), for packet n is j(n) = 323 {[z(n) + x_mean] + s(n)}. It follows that the separation in the post- 324 jitter time of packets n and n+1 is {[s(n+1)-s(n)] - [z(n)-z(n+1)]}. 325 Since the first term is always a positive quantity, we note that 326 packet reordering at the receiver is possible whenever the second 327 term is greater than the first. Said another way, whenever the 328 difference in possible zero mean PDV sample delays (i.e., [x_max- 329 x_min]) exceeds the inter-departure time of any two sent packets, we 330 have the possibility of packet re-ordering. 332 There are important use cases in real networks where packets can 333 become re-ordered such as in load balancing topologies and during 334 route changes. However, for the vast majority of cases there is no 335 packet re-ordering because most of the time packets follow the same 336 path. Due to this, if a packet becomes overly delayed, the packets 337 after it on that flow are also delayed. This is especially true for 338 mobile wireless links where there are per-flow queues prior to base 339 station scheduling. Owing to this important use case, we define 340 another PDV profile similar to the above, but one that does not allow 341 for re-ordering within a flow. 343 4.5.2. Approximately Random Subject to No-Reordering Bounded PDV (NR- 344 RPVD) 346 No Reordering RPDV, NR-RPVD, is defined similarly to the above with 347 one important exception. Let serial(n) be defined as the 348 serialization delay of packet n at the lowest bottleneck link rate 349 (or other appropriate rate) in a given test. Then we produce all the 350 post-jitter values for j(n) for n = 1, 2, ... N, where N is the 351 length of the source sequence s to be offset-ed. The exception can 352 be stated as follows: We revisit all j(n) beginning from index n=2, 353 and if j(n) is determined to be less than [j(n-1)+serial(n-1)], we 354 redefine j(n) to be equal to [j(n-1)+serial(n-1)] and continue for 355 all remaining n (i.e., n = 3, 4, .. N). This models the case where 356 the packet n is sent immediately after packet (n-1) at the bottleneck 357 link rate. Although this is generally the theoretical minimum in 358 that it assumes that no other packets from other flows are in-between 359 packet n and n+1 at the bottleneck link, it is a reasonable 360 assumption for per flow queuing. 362 We note that this assumption holds for some important exception 363 cases, such as packets immediately following outliers. There are a 364 multitude of software controlled elements common on end-to-end 365 Internet paths (such as firewalls, ALGs and other middleboxes) which 366 stop processing packets while servicing other functions (e.g., 367 garbage collection). Often these devices do not drop packets, but 368 rather queue them for later processing and cause many of the 369 outliers. Thus NR-RPVD models this particular use case (assuming 370 serial(n+1) is defined appropriately for the device causing the 371 outlier) and thus is believed to be important for adaptation 372 development for RMCAT. 374 [Editor's Note: It may require to define test distributions as well. 375 Example test distribution may include- 377 1 - Two-sided: Uniform PDV Distribution. Two quantities to define: 378 x_min and x_max. 380 2 - Two-sided: Truncated Gaussian PDV Distribution. Four quantities 381 to define: the appropriate x_min and x_max for test (e.g., +/- two 382 sigma values), the standard deviation, and the mean. 384 3 - One Sided: Truncated Gaussian PDV Distribution. Quantities to 385 define: three sigma value, the standard deviation, and the mean] 387 5. WiFi or Cellular Links 389 [I-D.ietf-rmcat-wireless-tests] describes the test cases to simulate 390 networks with wireless links. The document describes mechanism to 391 simulate both cellular and WiFi networks. 393 6. Traffic Models 395 6.1. TCP taffic model 397 Long-lived TCP flows will download data throughout the session and 398 are expected to have infinite amount of data to send or receive. For 399 example, to 401 Each short TCP flow is modeled as a sequence of file downloads 402 interleaved with idle periods. Not all short TCPs start at the same 403 time, i.e., some start in the ON state while others start in the OFF 404 state. 406 The short TCP flows can be modelled as follows: 30 connections start 407 simultaneously fetching small (30-50 KB) amounts of data. This 408 covers the case where the short TCP flows are not fetching a video 409 file. 411 The idle period between bursts of starting a group of TCP flows is 412 typically derived from an exponential distribution with the mean 413 value of 10 seconds. 415 [These values were picked based on the data available at 416 http://httparchive.org/interesting.php as of October 2015]. 418 6.2. RTP Video model 420 [I-D.ietf-rmcat-video-traffic-model] describes two types of video 421 traffic models for evaluating RMCAT candidate algorithms. The first 422 model statistically characterizes the behavior of a video encoder. 423 Whereas the second model uses video traces. 425 For example, test sequences are available at: [xiph-seq] and 426 [HEVC-seq]. 428 [Open issue: Which sequences are used? All? Some subset?] 430 6.3. Background UDP 432 [Open issue: Background UDP flow is modeled as a constant bit rate 433 (CBR) flow. It will download data at a particular CBR rate for the 434 complete session, or will change to particular CBR rate at predefined 435 intervals. They parameters are still TBD. e.g., packet size, packet 436 spacing interval, etc.] 438 7. Security Considerations 440 Security issues have not been discussed in this memo. 442 8. IANA Considerations 444 There are no IANA impacts in this memo. 446 9. Contributors 448 The content and concepts within this document are a product of the 449 discussion carried out in the Design Team. 451 Michael Ramalho provided the text for the Jitter model. 453 10. Acknowledgements 455 Much of this document is derived from previous work on congestion 456 control at the IETF. 458 The authors would like to thank Harald Alvestrand, Anna Brunstrom, 459 Luca De Cicco, Wesley Eddy, Lars Eggert, Kevin Gross, Vinayak Hegde, 460 Stefan Holmer, Randell Jesup, Mirja Kuehlewind, Karen Nielsen, Piers 461 O'Hanlon, Colin Perkins, Michael Ramalho, Zaheduzzaman Sarker, 462 Timothy B. Terriberry, Michael Welzl, and Mo Zanaty for providing 463 valuable feedback on earlier versions of this draft. Additionally, 464 also thank the participants of the design team for their comments and 465 discussion related to the evaluation criteria. 467 11. References 469 11.1. Normative References 471 [RFC3550] Schulzrinne, H., Casner, S., Frederick, R., and V. 472 Jacobson, "RTP: A Transport Protocol for Real-Time 473 Applications", STD 64, RFC 3550, DOI 10.17487/RFC3550, 474 July 2003, . 476 [RFC3551] Schulzrinne, H. and S. Casner, "RTP Profile for Audio and 477 Video Conferences with Minimal Control", STD 65, RFC 3551, 478 DOI 10.17487/RFC3551, July 2003, 479 . 481 [RFC3611] Friedman, T., Ed., Caceres, R., Ed., and A. Clark, Ed., 482 "RTP Control Protocol Extended Reports (RTCP XR)", RFC 483 3611, DOI 10.17487/RFC3611, November 2003, 484 . 486 [RFC4585] Ott, J., Wenger, S., Sato, N., Burmeister, C., and J. Rey, 487 "Extended RTP Profile for Real-time Transport Control 488 Protocol (RTCP)-Based Feedback (RTP/AVPF)", RFC 4585, DOI 489 10.17487/RFC4585, July 2006, 490 . 492 [RFC5506] Johansson, I. and M. Westerlund, "Support for Reduced-Size 493 Real-Time Transport Control Protocol (RTCP): Opportunities 494 and Consequences", RFC 5506, DOI 10.17487/RFC5506, April 495 2009, . 497 [I-D.ietf-rmcat-cc-requirements] 498 Jesup, R. and Z. Sarker, "Congestion Control Requirements 499 for Interactive Real-Time Media", draft-ietf-rmcat-cc- 500 requirements-09 (work in progress), December 2014. 502 [I-D.ietf-avtcore-rtp-circuit-breakers] 503 Perkins, C. and V. Varun, "Multimedia Congestion Control: 504 Circuit Breakers for Unicast RTP Sessions", draft-ietf- 505 avtcore-rtp-circuit-breakers-14 (work in progress), March 506 2016. 508 [I-D.ietf-rmcat-wireless-tests] 509 Sarker, Z., Johansson, I., Zhu, X., Fu, J., Tan, W., and 510 M. Ramalho, "Evaluation Test Cases for Interactive Real- 511 Time Media over Wireless Networks", draft-ietf-rmcat- 512 wireless-tests-01 (work in progress), November 2015. 514 11.2. Informative References 516 [RFC5033] Floyd, S. and M. Allman, "Specifying New Congestion 517 Control Algorithms", BCP 133, RFC 5033, DOI 10.17487/ 518 RFC5033, August 2007, 519 . 521 [RFC5166] Floyd, S., Ed., "Metrics for the Evaluation of Congestion 522 Control Mechanisms", RFC 5166, DOI 10.17487/RFC5166, March 523 2008, . 525 [RFC5681] Allman, M., Paxson, V., and E. Blanton, "TCP Congestion 526 Control", RFC 5681, DOI 10.17487/RFC5681, September 2009, 527 . 529 [I-D.ietf-rmcat-eval-test] 530 Sarker, Z., Varun, V., Zhu, X., and M. Ramalho, "Test 531 Cases for Evaluating RMCAT Proposals", draft-ietf-rmcat- 532 eval-test-03 (work in progress), March 2016. 534 [I-D.ietf-rmcat-video-traffic-model] 535 Zhu, X., Cruz, S., and Z. Sarker, "Modeling Video Traffic 536 Sources for RMCAT Evaluations", draft-ietf-rmcat-video- 537 traffic-model-00 (work in progress), January 2016. 539 [SA4-EVAL] 540 R1-081955, 3GPP., "LTE Link Level Throughput Data for SA4 541 Evaluation Framework", 3GPP R1-081955, 5 2008. 543 [SA4-LR] S4-050560, 3GPP., "Error Patterns for MBMS Streaming over 544 UTRAN and GERAN", 3GPP S4-050560, 5 2008. 546 [TCP-eval-suite] 547 Lachlan, A., Marcondes, C., Floyd, S., Dunn, L., Guillier, 548 R., Gang, W., Eggert, L., Ha, S., and I. Rhee, "Towards a 549 Common TCP Evaluation Suite", Proc. PFLDnet. 2008, August 550 2008. 552 [xiph-seq] 553 Xiph.org, , "Video Test Media", 554 http://media.xiph.org/video/derf/ , . 556 [HEVC-seq] 557 HEVC, , "Test Sequences", 558 http://www.netlab.tkk.fi/~varun/test_sequences/ , . 560 Appendix A. Application Trade-off 562 Application trade-off is yet to be defined. see RMCAT requirements 563 [I-D.ietf-rmcat-cc-requirements] document. Perhaps each experiment 564 should define the application's expectation or trade-off. 566 A.1. Measuring Quality 568 No quality metric is defined for performance evaluation, it is 569 currently an open issue. However, there is consensus that congestion 570 control algorithm should be able to show that it is useful for 571 interactive video by performing analysis using a real codec and video 572 sequences. 574 Appendix B. Change Log 576 Note to the RFC-Editor: please remove this section prior to 577 publication as an RFC. 579 B.1. Changes in draft-ietf-rmcat-eval-criteria-05 581 o Improved text surrounding wireless tests, video sequences, and 582 short-TCP model. 584 B.2. Changes in draft-ietf-rmcat-eval-criteria-04 586 o Removed the guidelines section, as most of the sections are now 587 covered: wireless tests, video model, etc. 589 o Improved Short TCP model based on the suggestion to use 590 httparchive.org. 592 B.3. Changes in draft-ietf-rmcat-eval-criteria-03 594 o Keep-alive version. 596 o Moved link parameters and traffic models from eval-test 598 B.4. Changes in draft-ietf-rmcat-eval-criteria-02 600 o Incorporated fairness test as a working test. 602 o Updated text on mimimum evaluation requirements. 604 B.5. Changes in draft-ietf-rmcat-eval-criteria-01 606 o Removed Appendix B. 608 o Removed Section on Evaluation Parameters. 610 B.6. Changes in draft-ietf-rmcat-eval-criteria-00 612 o Updated references. 614 o Resubmitted as WG draft. 616 B.7. Changes in draft-singh-rmcat-cc-eval-04 618 o Incorporate feedback from IETF 87, Berlin. 620 o Clarified metrics: convergence time, bandwidth utilization. 622 o Changed fairness criteria to fairness test. 624 o Added measuring pre- and post-repair loss. 626 o Added open issue of measuring video quality to appendix. 628 o clarified use of DropTail and AQM. 630 o Updated text in "Minimum Requirements for Evaluation" 632 B.8. Changes in draft-singh-rmcat-cc-eval-03 634 o Incorporate the discussion within the design team. 636 o Added a section on evaluation parameters, it describes the flow 637 and network characteristics. 639 o Added Appendix with self-fairness experiment. 641 o Changed bottleneck parameters from a proposal to an example set. 643 o 645 B.9. Changes in draft-singh-rmcat-cc-eval-02 647 o Added scenario descriptions. 649 B.10. Changes in draft-singh-rmcat-cc-eval-01 651 o Removed QoE metrics. 653 o Changed stability to steady-state. 655 o Added measuring impact against few and many flows. 657 o Added guideline for idle and data-limited periods. 659 o Added reference to TCP evaluation suite in example evaluation 660 scenarios. 662 Authors' Addresses 663 Varun Singh 664 Nemu Dialogue Systems Oy 665 Runeberginkatu 4c A 4 666 Helsinki 00100 667 Finland 669 Email: varun.singh@iki.fi 670 URI: http://www.callstats.io/ 672 Joerg Ott 673 Technical University of Munich 674 Faculty of Informatics 675 Boltzmannstrasse 3 676 Garching bei Muenchen, DE 85748 677 Germany 679 Email: ott@in.tum.de 681 Stefan Holmer 682 Google 683 Kungsbron 2 684 Stockholm 11122 685 Sweden 687 Email: holmer@google.com