idnits 2.17.1 draft-livingood-woundy-p4p-experiences-08.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** The document seems to lack a License Notice according IETF Trust Provisions of 28 Dec 2009, Section 6.b.ii or Provisions of 12 Sep 2009 Section 6.b -- however, there's a paragraph with a matching beginning. Boilerplate error? (You're using the IETF Trust Provisions' Section 6.b License Notice from 12 Feb 2009 rather than one of the newer Notices. See https://trustee.ietf.org/license-info/.) 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 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 (June 11, 2009) is 5432 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- No issues found here. Summary: 1 error (**), 0 flaws (~~), 1 warning (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Internet Engineering Task Force C. Griffiths 3 Internet-Draft J. Livingood, Ed. 4 Intended status: Informational Comcast 5 Expires: December 13, 2009 L. Popkin 6 Pando 7 R. Woundy, Ed. 8 Comcast 9 Y. Yang 10 Yale 11 June 11, 2009 13 Comcast's ISP Experiences In a P4P Technical Trial 14 draft-livingood-woundy-p4p-experiences-08 16 Status of this Memo 18 This Internet-Draft is submitted to IETF in full conformance with the 19 provisions of BCP 78 and BCP 79. 21 Internet-Drafts are working documents of the Internet Engineering 22 Task Force (IETF), its areas, and its working groups. Note that 23 other groups may also distribute working documents as Internet- 24 Drafts. 26 Internet-Drafts are draft documents valid for a maximum of six months 27 and may be updated, replaced, or obsoleted by other documents at any 28 time. It is inappropriate to use Internet-Drafts as reference 29 material or to cite them other than as "work in progress." 31 The list of current Internet-Drafts can be accessed at 32 http://www.ietf.org/ietf/1id-abstracts.txt. 34 The list of Internet-Draft Shadow Directories can be accessed at 35 http://www.ietf.org/shadow.html. 37 This Internet-Draft will expire on December 13, 2009. 39 Copyright Notice 41 Copyright (c) 2009 IETF Trust and the persons identified as the 42 document authors. All rights reserved. 44 This document is subject to BCP 78 and the IETF Trust's Legal 45 Provisions Relating to IETF Documents in effect on the date of 46 publication of this document (http://trustee.ietf.org/license-info). 47 Please review these documents carefully, as they describe your rights 48 and restrictions with respect to this document. 50 Abstract 52 This document describes the experiences of Comcast, a large cable 53 broadband Internet Service Provider (ISP) in the U.S., in a Proactive 54 Network Provider Participation for P2P (P4P) technical trial in July 55 2008. This trial used P4P iTracker technology being considered by 56 the IETF, as part of the Application Layer Transport Optimization 57 (ALTO) working group. 59 Table of Contents 61 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 62 2. High-Level Details . . . . . . . . . . . . . . . . . . . . . . 3 63 3. Differences Between the P4P iTrackers Used . . . . . . . . . . 4 64 3.1. P4P Fine Grain . . . . . . . . . . . . . . . . . . . . . . 5 65 3.2. P4P Coarse Grain . . . . . . . . . . . . . . . . . . . . . 5 66 3.3. P4P Generic Weighted . . . . . . . . . . . . . . . . . . . 6 67 4. High-Level Trial Results . . . . . . . . . . . . . . . . . . . 6 68 4.1. Swarm Size . . . . . . . . . . . . . . . . . . . . . . . . 6 69 4.2. Impact on Download Speed . . . . . . . . . . . . . . . . . 7 70 4.3. General Impacts on Upstream and Downstream Traffic and 71 Other Interesting Data . . . . . . . . . . . . . . . . . . 8 72 5. Important Notes on Data Collected . . . . . . . . . . . . . . 8 73 6. Next Steps . . . . . . . . . . . . . . . . . . . . . . . . . . 9 74 7. Security Considerations . . . . . . . . . . . . . . . . . . . 10 75 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 10 76 9. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 10 77 10. Informative References . . . . . . . . . . . . . . . . . . . . 11 78 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 11 80 1. Introduction 82 Comcast is a large broadband Internet Service Provider (ISP), based 83 in the U.S., serving the majority of its customers via cable modem 84 technology. A trial was conducted in July 2008 with Pando Networks, 85 Yale, and several ISP members of the P4P Working Group, which is part 86 of the Distributed Computing Industry Association (DCIA). Comcast is 87 a member of the DCIA's P4P Working Group, whose mission is to work 88 with Internet service providers (ISPs), peer-to-peer (P2P) companies, 89 and technology researchers to develop "P4P" mechanisms, such as so- 90 called "iTrackers" (hereafter P4P iTrackers), that accelerate 91 distribution of content and optimize utilization of ISP network 92 resources. P4P iTrackers theoretically allow P2P networks to 93 optimize traffic within each ISP, reducing the volume of data 94 traversing the ISP's infrastructure and creating a more manageable 95 flow of data. P4P iTrackers can also accelerate P2P downloads for 96 end users. 98 P4P's so-called "iTracker" technology [SIGCOMM] was conceptually 99 discussed with the IETF at the Peer-to-Peer Infrastructure (P2Pi) 100 Workshop held on May 28, 2008, at the Massachusetts Institute of 101 Technology (MIT), as documented in [I-D.p2pi-cooper-workshop-report]. 102 This work was discussed in greater detail at the 72nd meeting of the 103 IETF, in Dublin, Ireland, in the ALTO BoF on July 29, 2008. Due to 104 interest from the community, Comcast shared P4P iTracker trial data 105 at the 73rd meeting of the IETF, in Minneapolis, Minnesota, in the 106 ALTO BoF on November 18, 2008. Since that time, discussion of P4P 107 iTrackers and alternative technologies has continued among 108 participants of the ALTO working group. 110 The P4P iTracker trial was conducted, in cooperation with Pando, 111 Yale, and three other P4P member ISPs, from July 2 to July 17, 2008. 112 This was the first P4P iTracker trial over a cable broadband network. 113 The trial used a Pando P2P client, and Pando distributed a special 21 114 MB licensed video file in order to measure the effectiveness of P4P 115 iTrackers. A primary objective of the trial was to measure the 116 effects that increasing the localization of P2P swarms would have on 117 P2P uploads, P2P downloads, and ISP networks, in comparison to normal 118 P2P activity. 120 2. High-Level Details 122 There were five different swarms for the content used in the trial. 123 A swarm is defined as the group of peers that are downloading and/or 124 uploading a particular file. The first was a random P2P swarm, as a 125 control group. The second, third, and fourth used different P4P 126 iTrackers: Generic, Coarse Grained, and Fine Grained, all of which 127 are described in Section 3. The fifth was a proprietary Pando 128 mechanism. (The results of the fifth swarm, while satisfactory, are 129 not included here since our focus is on open standards and a 130 mechanism which may be leveraged for the benefit of the entire 131 community of P2P clients.) Comcast deployed a P4P iTracker server in 132 its production network to support this trial, and configured multiple 133 iTracker files to provide varying levels of localization to clients. 135 In the trial itself, a P2P client begins a P2P session by querying a 136 pTracker, which runs and manages the P2P network. The pTracker 137 occasionally queries the P4P iTracker, which in this case was 138 maintained by Comcast, the ISP. Other ISPs either managed their own 139 P4P iTracker or used Pando or Yale to host their P4P iTracker files. 140 The P4P iTracker returns network topology information to the 141 pTracker, which then communicates with P2P clients, in order to 142 enable P2P clients to make network-aware decisions regarding peers. 144 The Pando client was enabled to capture extended logging, when the 145 version of the client included support for it. The extended logging 146 included the source and destination IP address of all P2P transfers, 147 the number of bytes transferred, and the start and end timestamps. 148 This information gives a precise measurement of the data flow in the 149 network, allowing computation of data transfer volumes as well as 150 data flow rates at each point in time. With standard logging, Pando 151 captured the start and completion times of every download, as well as 152 the average transfer rate observed by the client for the download. 154 Pando served the data from an origin server external to Comcast's 155 network. This server served about 10 copies of the file, after which 156 all transfers (about 1 million downloads across all ISPs) were 157 performed purely via P2P. 159 The P2P clients in the trial start with tracker-provided peers, then 160 use peer exchange to discover additional peers. Thus, the initial 161 peers were provided according to P4P iTracker guidance (90% guidance 162 based on P4P iTracker topology, and 10% random guidance), then later 163 peers discover the entire swarm via either additional announces or 164 peer exchange. 166 3. Differences Between the P4P iTrackers Used 168 Given the size of the Comcast network, it was felt that in order to 169 truly evaluate the P4P iTracker application we would need to test 170 various network topologies that reflected its network and would help 171 gauge the level of effort and design requirements necessary to get 172 correct statistical data out of the trial. In all cases, P4P 173 iTrackers were configured with automation in mind, so that any 174 successful P4P iTracker configuration would be automatically 175 updating, rather than manually configured on an on-going basis. All 176 P4P iTrackers were hosted on the same small server, and it appeared 177 to be relatively easy and inexpensive to scale up a P4P iTracker 178 infrastructure should P4P iTracker-like mechanisms become 179 standardized and widely adopted. 181 3.1. P4P Fine Grain 183 The Fine Grain topology was the first and most complex P4P iTracker 184 that we built for this trial. It was a detailed mapping of Comcast 185 backbone-connected network Autonomous System Numbers (ASN) to IP 186 Aggregates which were weighted based on priority and distance from 187 each other. Included in this design was a prioritization of all Peer 188 and Internet transit connected ASNs to the Comcast backbone to ensure 189 that P4P traffic would prefer settlement free and lower cost networks 190 first, and then more expensive transit links. This attempted to 191 optimize and lower transit costs associated with this traffic. We 192 then took the additional step of detailing each ASN and IP aggregate 193 into IP subnets down to Optical Transport Nodes (OTN) where all Cable 194 Modem Termination Systems (CMTS, as briefly defined in Section 2.6 of 195 [RFC3083]) reside . This design gave a highly localized and detailed 196 description of the Comcast network for the iTracker to disseminate. 197 This design defined 1,182 P4P iTracker node identifiers, and resulted 198 in a 107,357 line configuration file. 200 This P4P iTracker was obviously the most time-consuming to create and 201 the most complex to maintain. Trial results indicated that this 202 level of localization was too high, and was less effective compared 203 to lower levels of localization. 205 3.2. P4P Coarse Grain 207 Given the level of detail in the Fine Grain design, it was important 208 that we also enable a high-level design which still used priority and 209 weighting mechanisms for the Comcast backbone and transit links. The 210 Coarse Grain design was a limited or summarized version of the Fine 211 Grain design, which used the ASN to IP Aggregate and weighted data 212 for transit links, but removed all additional localization data. 213 This insured we would get similar data sets from the Fine Grain 214 design, but without the more detailed localization of each of the 215 networks attached to the Comcast backbone. This design defined 22 216 P4P iTracker node identifiers, and resulted in a 998 line 217 configuration file. 219 From an overall cost, complexity, risk, and effectiveness standpoint, 220 this was judged to be the optimal P4P iTracker for Comcast. 221 Importantly, this did not require revealing the complex, internal 222 network topology that the Fine Grain did. Updates to this iTracker 223 were also far simpler to automate, which will better ensure that it 224 is accurate over time, and keeps administrative overhead relatively 225 low. However, the differences, costs, and benefits of Coarse Grain 226 and Generic Weighted (see below) likely merit further study. 228 3.3. P4P Generic Weighted 230 The Generic Weighted design was a copy of the Coarse Grained design 231 but instead of using ISP-designated priority and weights, all weights 232 were defaulted to pre-determined parameters that the Yale team had 233 designed. All other data was replicated from the Coarse Grain 234 design. Gathering and providing the information necessary to support 235 the Generic Weighted iTracker was roughly the same level of effort as 236 for Coarse Grain. 238 4. High-Level Trial Results 240 Trial data was collected by Pando Networks and Yale University, and 241 raw trial results were shared with Comcast and all of the other ISPs 242 involved in the trial. Analysis of the raw results was performed by 243 Pando and Yale, and these organizations delivered an analysis of the 244 P4P iTracker trial. Using the raw data, Comcast also analyzed the 245 trial results. Furthermore, the raw trial results for Comcast were 246 shared with Net Forecast, Inc., which performed an independent 247 analysis of the trial for Comcast. 249 4.1. Swarm Size 251 During the trial, downloads peaked at 24,728 per day, per swarm, or 252 nearly 124,000 per day for all five swarms. The swarm size peaked at 253 11,703 peers per swarm, or nearly 57,000 peers for all five swarms. 254 We observed a comparable number of downloads in each of the five 255 swarms. 257 For each swarm, Table 1 below gives the number of downloaders per 258 swarm from Comcast that finished downloading, and the number of 259 downloaders from Comcast that canceled downloading before finishing. 261 Characteristics of P4P iTracker Swarms: 263 +-----------+-----------+---------------+------------+--------------+ 264 | Swarm | Completed | Cancellations | Total | Cancellation | 265 | | Downloads | | Attempts | Rate | 266 +-----------+-----------+---------------+------------+--------------+ 267 | Random | 2,719 | 89 | 2,808 | 3.17% | 268 | (Control) | | | | | 269 | --------- | --------- | ----------- | ---------- | ----------- | 270 | P4P Fine | 2,846 | 64 | 2,910 | 2.20% | 271 | Grained | | | | | 272 | --------- | --------- | ----------- | ---------- | ----------- | 273 | P4P | 2,775 | 63 | 2,838 | 2.22% | 274 | Generic | | | | | 275 | Weight | | | | | 276 | --------- | --------- | ----------- | ---------- | ----------- | 277 | P4P | 2,886 | 52 | 2,938 | 1.77% | 278 | Coarse | | | | | 279 | Grained | | | | | 280 +-----------+-----------+---------------+------------+--------------+ 282 Table 1: Per-Swarm Size and Cancellation Rates 284 4.2. Impact on Download Speed 286 The results of the trial indicated that P4P iTrackers can improve the 287 speed of downloads to P2P clients. In addition, P4P iTrackers were 288 effective in localizing P2P traffic within the Comcast network. 290 Impact of P4P iTrackers on Downloads: 292 +--------------+------------+------------+-------------+------------+ 293 | Swarm | Global Avg | Change | Comcast Avg | Change | 294 | | bps | | bps | | 295 +--------------+------------+------------+-------------+------------+ 296 | Random | 144,045 | n/a | 254,671 bps | n/a | 297 | (Control) | bps | | | | 298 | ---------- | ---------- | ---------- | ---------- | ---------- | 299 | P4P Fine | 162,344 | +13% | 402,043 bps | +57% | 300 | Grained | bps | | | | 301 | ---------- | ---------- | ---------- | ---------- | ---------- | 302 | P4P Generic | 163,205 | +13% | 463,782 bps | +82% | 303 | Weight | bps | | | | 304 | ---------- | ---------- | ---------- | ---------- | ---------- | 305 | P4P Coarse | 166,273 | +15% | 471,218 bps | +85% | 306 | Grained | bps | | | | 307 +--------------+------------+------------+-------------+------------+ 308 Table 2: Per-Swarm Global and Comcast Download Speeds 310 4.3. General Impacts on Upstream and Downstream Traffic and Other 311 Interesting Data 313 An analysis of the effects of P4P iTracker use on upstream 314 utilization and Internet transit was also interesting. It did not 315 appear that P4P iTrackers significantly increased upstream 316 utilization in the Comcast access network; in essence uploading was 317 already occurring no matter what and a P4P iTracker in and of itself 318 did not appear to materially increase uploading for this specific, 319 licensed content. (A P4P iTracker is not intended as a solution for 320 the potential of network congestion to occur.) Random was 143,236 MB 321 and P4P Generic Weight was 143,143 MB, while P4P Coarse Grained was 322 139,669 MB. We also observed that using a P4P iTracker reduced 323 outgoing Internet traffic by an average of 34% at peering points. 324 Random was 134,219 MB and P4P Generic Weight was 91,979 MB, while P4P 325 Coarse Grained was 86,652 MB. 327 In terms of downstream utilization, we observed that the use of a P4P 328 iTracker reduced incoming Internet traffic by an average of 80% at 329 peering points. Random was 47,013 MB, P4P Generic Weight was 8,610 330 MB, and P4P Coarse Grained was 7,764 MB. However, we did notice that 331 download activity in the Comcast access network increased somewhat, 332 from 56,030 MB for Random, to 59,765 MB for P4P Generic Weight, and 333 60,781 MB for P4P Coarse Grained. Note that for each swarm, the 334 number of downloaded bytes according to logging reports is very close 335 to the number of downloaders multiplied by file size. But they do 336 not exactly match due to log report errors and duplicated chunks. 337 One factor contributing to the differences in access network download 338 activity is that different swarms have different numbers of 339 downloaders, due to random variations during uniform random 340 assignment of downloaders to swarms (see Table 1). One interesting 341 observation is that Random has higher cancellation rate (3.17%) than 342 that of the guided swarms (1.77%-2.22%). Whether guided swarms 343 achieve lower cancellation rate is an interesting issue for future 344 research. 346 5. Important Notes on Data Collected 348 Raw data is presented in this document. We did not normalize traffic 349 volume data (e.g. upload and download) by the number of downloads in 350 order to preserve this underlying raw data. 352 We also recommend that readers not focus too much on the absolute 353 numbers, such as bytes downloaded from internal sources and bytes 354 downloaded from external sources. Instead, we recommend readers 355 focus on ratios such as the percentage of bytes downloaded that came 356 from internal sources in each swarm. As a result, the small random 357 variation between number of downloads of each swarm does not distract 358 readers from important metrics like shifting traffic from external to 359 internal sources, among other things. 361 We also wish to note that the data was collected from a sample of the 362 total swarm. Specifically, there were some peers running older 363 versions of the Pando client that did not implement the extended 364 transfer logging. For those nodes, which participated in the swarms 365 but did not report their data transfers, we have download counts. 366 The result of this is that, for example, the download counts 367 generated from the standard logging are a bit higher than the 368 download counts generated by the extended logging. That being said, 369 over 90% of downloads were by peers running the newer software, which 370 we believe shows that the transfer records are highly representative 371 of the total data flow. 373 In terms of which analysis was performed from the standard logging 374 compared to extended logging, all of the data flow analysis was 375 performed using the extended logging. Pando's download counts and 376 performance numbers were generated via standard logging (i.e. all 377 peers report download complete/cancel, data volumes, and measured 378 download speed on the client). Yale's download counts and 379 performance numbers were derived via extended logging (e.g. by 380 summing the transfer records, counting IP addresses reported, etc.). 382 One benefit of having two data sources is that we can compare the 383 two. In this case, the two approaches both reported comparable 384 impacts. 386 6. Next Steps 388 One objective of this document is to share with the IETF community 389 the results of one P4P iTracker trial in a large broadband network, 390 given skepticism regarding the benefits to P2P users as well as to 391 ISPs. From the perspective of P2P users, P4P iTrackers potentially 392 deliver faster P2P downloads. At the same time, ISPs can increase 393 the localization of swarms, enabling them to reduce bytes flowing 394 over transit points, while also delivering an optimized P2P 395 experience to customers. However, an internal analysis of varying 396 levels of P4P iTracker adoption by ISPs leads us to believe that, 397 while P4P iTracker-type mechanisms are valuable on a single ISP 398 basis, the value of P4P iTrackers increases dramatically as many ISPs 399 choose to deploy it. 401 We believe these results can inform the technical discussion in the 402 IETF over how to use P4P iTracker mechanisms. Should such a 403 mechanism be standardized, the use of ISP-provided P4P iTrackers 404 should probably be an opt-in feature for P2P users, or at least a 405 feature of which they are explicitly aware of and which has been 406 enabled by default in a particular P2P client. In this way, P2P 407 users could choose to opt-in either explicitly or by their choice of 408 P2P client in order to choose to use the P4P iTracker to improve 409 performance, which benefits both the user and the ISP at the same 410 time. Importantly in terms of privacy, the P4P iTracker makes 411 available only network topology information, and would not in its 412 current form enable an ISP, via the P4P iTracker, to determine which 413 P2P clients were downloading any specific content, whether to 414 determine for example if content was a song or a movie or even the 415 title. 417 It is also possible that a P4P iTracker type of mechanism, in 418 combination with a P2P cache, could further improve P2P download 419 performance, which merits further study. In addition, this was a 420 limited trial that, while very promising, indicates a need for 421 additional technical investigation and trial work. Such follow-up 422 study should explore the effects of P4P iTrackers when more P2P 423 client software variants are involved, with larger swarms, and with 424 additional and more technically diverse content (file size, file 425 type, duration of content, etc.). 427 7. Security Considerations 429 There are no security considerations in this document. 431 NOTE TO RFC EDITOR: PLEASE REMOVE THIS NULL SECTION PRIOR TO 432 PUBLICATION. 434 8. IANA Considerations 436 There are no IANA considerations in this document. 438 NOTE TO RFC EDITOR: PLEASE REMOVE THIS NULL SECTION PRIOR TO 439 PUBLICATION. 441 9. Acknowledgements 443 The authors wish to acknowledge the hard work of all of the P4P 444 working group members, and specifically the focused efforts of the 445 teams at both Pando and Yale for the trial itself. Finally, the 446 authors recognize and appreciate Peter Sevcik and John Bartlett, of 447 NetForecast, Inc., for their valued independent analysis of the trial 448 results. 450 10. Informative References 452 [I-D.p2pi-cooper-workshop-report] 453 Peterson, J. and A. Cooper, "Report from the IETF workshop 454 on P2P Infrastructure, May 28, 2008", 455 draft-p2pi-cooper-workshop-report-01 (work in progress), 456 February 2009. 458 [RFC3083] Woundy, R., "Baseline Privacy Interface Management 459 Information Base for DOCSIS Compliant Cable Modems and 460 Cable Modem Termination Systems", RFC 3083, March 2001. 462 [SIGCOMM] Xie, H., Yang, Y., Krishnamurthy, A., Liu, Y., and A. 463 Silberschatz, "ACM SIGCOMM 2008 - P4P: Provider Portal for 464 Applications", Association for Computing Machinery SIGCOMM 465 2008 Proceedings, August 2008, 466 . 468 Authors' Addresses 470 Chris Griffiths 471 Comcast Cable Communications 472 One Comcast Center 473 1701 John F. Kennedy Boulevard 474 Philadelphia, PA 19103 475 US 477 Email: chris_griffiths@cable.comcast.com 478 URI: http://www.comcast.com 480 Jason Livingood (editor) 481 Comcast Cable Communications 482 One Comcast Center 483 1701 John F. Kennedy Boulevard 484 Philadelphia, PA 19103 485 US 487 Email: jason_livingood@cable.comcast.com 488 URI: http://www.comcast.com 489 Laird Popkin 490 Pando Networks 491 520 Broadway Street 492 10th Floor 493 New York, NY 10012 494 US 496 Email: laird@pando.com 497 URI: http://www.pando.com 499 Richard Woundy (editor) 500 Comcast Cable Communications 501 27 Industrial Avenue 502 Chelmsford, MA 01824 503 US 505 Email: richard_woundy@cable.comcast.com 506 URI: http://www.comcast.com 508 Richard Yang 509 Yale University 510 51 Prospect Street 511 New Haven, CT 06520 512 US 514 Email: yry@cs.yale.edu 515 URI: http://www.cs.yale.edu