idnits 2.17.1 draft-ietf-ippm-rate-problem-01.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (December 21, 2012) is 4136 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Unused Reference: 'RFC1305' is defined on line 385, but no explicit reference was found in the text == Unused Reference: 'RFC5618' is defined on line 410, but no explicit reference was found in the text == Unused Reference: 'RFC5938' is defined on line 414, but no explicit reference was found in the text == Unused Reference: 'RFC6038' is defined on line 418, but no explicit reference was found in the text ** Obsolete normative reference: RFC 1305 (Obsoleted by RFC 5905) ** Obsolete normative reference: RFC 2679 (Obsoleted by RFC 7679) ** Obsolete normative reference: RFC 2680 (Obsoleted by RFC 7680) Summary: 3 errors (**), 0 flaws (~~), 5 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group A. Morton 3 Internet-Draft AT&T Labs 4 Intended status: Informational December 21, 2012 5 Expires: June 24, 2013 7 Rate Measurement Problem Statement 8 draft-ietf-ippm-rate-problem-01 10 Abstract 12 There is a rate measurement scenario which has wide-spread attention 13 of Internet access subscribers and seemingly all industry players, 14 including regulators. This memo presents an access rate-measurement 15 problem statement for IP Performance Metrics. Key test protocol 16 aspects require the ability to control packet size on the tested path 17 and enable asymmetrical packet size testing in a controller-responder 18 architecture. 20 Requirements Language 22 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 23 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 24 document are to be interpreted as described in RFC 2119 [RFC2119]. 26 Status of this Memo 28 This Internet-Draft is submitted in full conformance with the 29 provisions of BCP 78 and BCP 79. 31 Internet-Drafts are working documents of the Internet Engineering 32 Task Force (IETF). Note that other groups may also distribute 33 working documents as Internet-Drafts. The list of current Internet- 34 Drafts is at http://datatracker.ietf.org/drafts/current/. 36 Internet-Drafts are draft documents valid for a maximum of six months 37 and may be updated, replaced, or obsoleted by other documents at any 38 time. It is inappropriate to use Internet-Drafts as reference 39 material or to cite them other than as "work in progress." 41 This Internet-Draft will expire on June 24, 2013. 43 Copyright Notice 45 Copyright (c) 2012 IETF Trust and the persons identified as the 46 document authors. All rights reserved. 48 This document is subject to BCP 78 and the IETF Trust's Legal 49 Provisions Relating to IETF Documents 50 (http://trustee.ietf.org/license-info) in effect on the date of 51 publication of this document. Please review these documents 52 carefully, as they describe your rights and restrictions with respect 53 to this document. Code Components extracted from this document must 54 include Simplified BSD License text as described in Section 4.e of 55 the Trust Legal Provisions and are provided without warranty as 56 described in the Simplified BSD License. 58 Table of Contents 60 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 61 2. Purpose and Scope . . . . . . . . . . . . . . . . . . . . . . 3 62 3. Active Rate Measurement . . . . . . . . . . . . . . . . . . . 5 63 4. Measurement Method Categories . . . . . . . . . . . . . . . . 6 64 5. Test Protocol Control & Generation Requirements . . . . . . . 8 65 6. Security Considerations . . . . . . . . . . . . . . . . . . . 8 66 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 9 67 8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 9 68 9. Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 69 10. References . . . . . . . . . . . . . . . . . . . . . . . . . . 9 70 10.1. Normative References . . . . . . . . . . . . . . . . . . 9 71 10.2. Informative References . . . . . . . . . . . . . . . . . 10 72 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . . 10 74 1. Introduction 76 There are many possible rate measurement scenarios. This memo 77 describes one rate measurement problem and presents a rate- 78 measurement problem statement for IP Performance Metrics (IPPM). 80 The access-rate scenario or use case has wide-spread attention of 81 Internet access subscribers and seemingly all Internet industry 82 players, including regulators. This problem is being approached with 83 many different measurement methods. 85 2. Purpose and Scope 87 The scope and purpose of this memo is to define the measurement 88 problem statement for test protocols conducting access rate 89 measurement on production networks. Relevant test protocols include 90 [RFC4656] and [RFC5357]), but the problem is stated in a general way 91 so that it can be addressed by any existing test protocol. This memo 92 discusses possibilities for methods of measurement, but does not 93 specify exact methods which would normally be part of the solution, 94 not the problem. 96 We characterize the access rate measurement scenario as follows: 98 o The Access portion of the network is the focus of this problem 99 statement. The user typically subscribes to a service with bi- 100 directional access partly described by rates in bits per second. 102 o Rates at the edge of the network are several orders of magnitude 103 less than aggregation and core portions. 105 o Asymmetrical ingress and egress rates are prevalent. 107 o Extremely large scale of access services requires low complexity 108 devices participating at the user end of the path. 110 Today, the majority of widely deployed access services achieve rates 111 less than 100 Mbit/s, and this is the order of magnitude for which a 112 solution is sought now. 114 This problem statement assumes that the most-likely bottleneck device 115 or link is adjacent to the remote (user-end) measurement device, or 116 is within one or two router/switch hops of the remote measurement 117 device. 119 Other use cases for rate measurement involve situations where the 120 packet switching and transport facilities are leased by one operator 121 from another and the actual capacity available cannot be directly 122 determined (e.g., from device interface utilization). These 123 scenarios could include mobile backhaul, Ethernet Service access 124 networks, and/or extensions of layer 2 or layer 3 networks. The 125 results of rate measurements in such cases could be employed to 126 select alternate routing, investigate whether capacity meets some 127 previous agreement, and/or adapt the rate of traffic sources if a 128 capacity bottleneck is found via the rate measurement. In the case 129 of aggregated leased networks, available capacity may also be 130 asymmetric. In these cases, the tester is assumed to have a sender 131 and receiver location under their control. We refer to this scenario 132 below as the aggregated leased network case. 134 Only active measurement methods will be addressed here, consistent 135 with the IPPM working group's current charter. Active measurements 136 require synthetic traffic dedicated to testing, and do not use user 137 traffic. 139 The actual path used by traffic may influence the rate measurement 140 results for some forms of access, as it may differ between user and 141 test traffic if the test traffic has different characteristics, 142 primarily in terms of the packets themselves (the Type-P described in 143 [RFC2330]). 145 There are several aspects of Type-P where user traffic may be 146 examined and directed to special treatment that may affect 147 transmission rates. The possibilities include: 149 o Packet length 151 o IP addresses used 153 o Transport protocol used (where TCP packets may be routed 154 differently from UDP) 156 o Transport Protocol port numbers used 158 This issue requires further discussion when specific solutions/ 159 methods of measurement are proposed, but for this problem statement 160 it is sufficient to Identify the problem and indicate that the 161 solution may require an extremely close emulation of user traffic, in 162 terms of the factors above. 164 Although the user may have multiple instances of network access 165 available to them, the primary problem scope is to measure one form 166 of access at a time. It is plausible that a solution for the single 167 access problem will be applicable to simultaneous measurement of 168 multiple access instances, but discussion of this is beyond the 169 current scope. 171 A key consideration is whether active measurements will be conducted 172 with user traffic present (In-Service testing), or not present (Out- 173 of-Service testing), such as during pre-service testing or 174 maintenance that interrupts service temporarily. Out-of-Service 175 testing includes activities described as "service commissioning", 176 "service activation", and "planned maintenance". Both In-Service and 177 Out-of-Service testing are within the scope of this problem. 179 It is a non-goal to solve the measurement protocol specification 180 problem in this memo. 182 It is a non-goal to standardize methods of measurement in this memo. 183 However, the problem statement will mandate that support for one or 184 more categories of rate measurement methods and adequate control 185 features for the methods in the test protocol. 187 3. Active Rate Measurement 189 This section lists features of active measurement methods needed to 190 measure access rates in production networks. 192 Test coordination between source and destination devices through 193 control messages and other basic capabilities described in the 194 methods of IPPM RFCs [RFC2679][RFC2680] are taken as given (these 195 could be listed later, if desired). 197 Most forms of active testing intrude on user performance to some 198 degree. One key tenet of IPPM methods is to minimize test traffic 199 effects on user traffic in the production network. Section 5 of 200 [RFC2680] lists the problems with high measurement traffic rates, and 201 the most relevant for rate measurement is the tendency for 202 measurement traffic to skew the results, followed by the possibility 203 of introducing congestion on the access link. Obviously, categories 204 of rate measurement methods that use less active test traffic than 205 others with similar accuracy SHALL be preferred for In-Service 206 testing. 208 On the other hand, Out-of-Service tests where the test path shares no 209 links with In-Service user traffic have none of the congestion or 210 skew concerns, but these tests must address other practical concerns 211 such as conducting measurements within a reasonable time from the 212 tester's point of view. Out-of-Service tests where some part of the 213 test path is shared with In-Service traffic MUST respect the In- 214 Service constraints. 216 The **intended metrics to be measured** have strong influence over 217 the categories of measurement methods required. For example, using 218 the terminology of [RFC5136], a it may be possible to measure a Path 219 Capacity Metric while In-Service if the level of background (user) 220 traffic can be assessed and included in the reported result. 222 The measurement *architecture* MAY be either of one-way (e.g., 223 [RFC4656]) or two-way (e.g., [RFC5357]), but the scale and complexity 224 aspects of end-user or aggregated access measurement clearly favor 225 two-way (with low-complexity user-end device and round-trip results 226 collection, as found in [RFC5357]). However, the asymmetric rates of 227 many access services mean that the measurement system MUST be able to 228 evaluate performance in each direction of transmission. In the two- 229 way architecture, it is expected that both end devices MUST include 230 the ability to launch test streams and collect the results of 231 measurements in both (one-way) directions of transmission (this 232 requirement is consistent with previous protocol specifications, and 233 it is not a unique problem for rate measurements). 235 The following paragraphs describe features for the roles of test 236 packet SENDER, RECEIVER, and results REPORTER. 238 SENDER: 240 Generate streams of test packets with various characteristics as 241 desired (see Section 4). The SENDER may be located at the user end 242 of the access path, or may be located elsewhere in the production 243 network, such as at one end of an aggregated leased network segment. 245 RECEIVER: 247 Collect streams of test packets with various characteristics (as 248 described above), and make the measurements necessary to support rate 249 measurement at the other end of an end-user access or aggregated 250 leased network segment. 252 REPORTER: 254 Use information from test packets and local processes to measure 255 delivered packet rates. 257 4. Measurement Method Categories 259 The design of rate measurement methods can be divided into two 260 phases: test stream design and measurement (SENDER and RECEIVER), and 261 a follow-up phase for analysis of the measurement to produce results 262 (REPORTER). The measurement protocol that addresses this problem 263 MUST only serve the test stream generation and measurement functions. 265 For the purposes of this problem statement, we categorize the many 266 possibilities for rate measurement stream generation as follows; 268 1. Packet pairs, with fixed intra-pair packet spacing and fixed or 269 random time intervals between pairs in a test stream. 271 2. Multiple streams of packet pairs, with a range of intra-pair 272 spacing and inter-pair intervals. 274 3. One or more packet ensembles in a test stream, using a fixed 275 ensemble size in packets and one or more fixed intra-ensemble 276 packet spacings (including zero spacing). 278 4. One or more packet chirps, where intra-packet spacing typically 279 decreases between adjacent packets in the same chirp and each 280 pair of packets represents a rate for testing purposes. 282 For all categories, the test protocol MUST support: 284 1. Variable payload lengths among packet streams 286 2. Variable length (in packets) among packet streams or ensembles 288 3. Variable IP header markings among packet streams 290 4. Choice of UDP transport and variable port numbers, OR, choice of 291 TCP transport and variable port numbers for two-way architectures 292 only, OR BOTH. 294 5. Variable number of packets-pairs, ensembles, or streams used in a 295 test session 297 The items above are additional variables that the test protocol MUST 298 be able to identify and control. 300 The test protocol SHALL support test packet ensemble generation 301 (category 3), as this appears to minimize the demands on measurement 302 accuracy. Other stream generation categories are OPTIONAL. 304 >>>>>> 306 Note: For measurement systems employing TCP Transport protocol, the 307 ability to generate specific stream characteristics requires a sender 308 with the ability to establish and prime the connection such that the 309 desired stream characteristics are allowed. See Mathis' work in 310 progress for more background [draft-mathis-ippm-model-based-metrics]. 312 The general requirement statements needed to describe an "open-loop" 313 TCP sender require some additional discussion. 315 It may also be useful to specify a control for Bulk Transfer Capacity 316 measurement with fully-specified TCP senders and receivers, as 317 envisioned in [RFC3148], but this would be a brute-force assessment 318 which does not follow the conservative tenets of IPPM measurement. 320 >>>>>> 322 Measurements for each test packet transferred between SENDER and 323 RECEIVER MUST be compliant with the singleton measurement methods 324 described in IPPM RFCs [RFC2679][RFC2680] (these could be listed 325 later, if desired). The time-stamp information or loss/arrival 326 status for each packet MUST be available for communication to the 327 protocol entity that collects results. 329 5. Test Protocol Control & Generation Requirements 331 Essentially, the test protocol MUST support the measurement features 332 described in the sections above. This requires: 334 1. Communicating all test variables to the Sender and Receiver 336 2. Results collection in a one-way architecture 338 3. Remote device control for both one-way and two-way architectures 340 4. Asymmetric and/or pseudo-one-way test capability in a two-way 341 measurement architecture 343 The ability to control packet size on the tested path and enable 344 asymmetrical packet size testing in a two-way architecture are 345 REQUIRED. 347 The test protocol SHOULD enable measurement of the [RFC5136] Capacity 348 metric, either Out-of-Service, In-Service, or both. Other [RFC5136] 349 metrics are OPTIONAL. 351 6. Security Considerations 353 The security considerations that apply to any active measurement of 354 live networks are relevant here as well. See [RFC4656] and 355 [RFC5357]. 357 There may be a serious issue if a proprietary Service Level Agreement 358 involved with the access network segment provider were somehow leaked 359 in the process of rate measurement. To address this, test protocols 360 SHOULD NOT convey this information in a way that could be discovered 361 by unauthorized parties. 363 7. IANA Considerations 365 This memo makes no requests of IANA. 367 8. Acknowledgements 369 Dave McDysan provided comments and text for the aggregated leased use 370 case. Yaakov Stein suggested many considerations to address, 371 including the In-Service vs. Out-of-Service distinction and its 372 implication on test traffic limits and protocols. 374 9. Appendix 376 This Appendix was proposed to briefly summarize previous rate 377 measurement experience. (There is a large body of research on rate 378 measurement, so there is a question of what to include and what to 379 omit. Suggestions are welcome.) 381 10. References 383 10.1. Normative References 385 [RFC1305] Mills, D., "Network Time Protocol (Version 3) 386 Specification, Implementation", RFC 1305, March 1992. 388 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 389 Requirement Levels", BCP 14, RFC 2119, March 1997. 391 [RFC2330] Paxson, V., Almes, G., Mahdavi, J., and M. Mathis, 392 "Framework for IP Performance Metrics", RFC 2330, 393 May 1998. 395 [RFC2679] Almes, G., Kalidindi, S., and M. Zekauskas, "A One-way 396 Delay Metric for IPPM", RFC 2679, September 1999. 398 [RFC2680] Almes, G., Kalidindi, S., and M. Zekauskas, "A One-way 399 Packet Loss Metric for IPPM", RFC 2680, September 1999. 401 [RFC4656] Shalunov, S., Teitelbaum, B., Karp, A., Boote, J., and M. 403 Zekauskas, "A One-way Active Measurement Protocol 404 (OWAMP)", RFC 4656, September 2006. 406 [RFC5357] Hedayat, K., Krzanowski, R., Morton, A., Yum, K., and J. 407 Babiarz, "A Two-Way Active Measurement Protocol (TWAMP)", 408 RFC 5357, October 2008. 410 [RFC5618] Morton, A. and K. Hedayat, "Mixed Security Mode for the 411 Two-Way Active Measurement Protocol (TWAMP)", RFC 5618, 412 August 2009. 414 [RFC5938] Morton, A. and M. Chiba, "Individual Session Control 415 Feature for the Two-Way Active Measurement Protocol 416 (TWAMP)", RFC 5938, August 2010. 418 [RFC6038] Morton, A. and L. Ciavattone, "Two-Way Active Measurement 419 Protocol (TWAMP) Reflect Octets and Symmetrical Size 420 Features", RFC 6038, October 2010. 422 10.2. Informative References 424 [RFC3148] Mathis, M. and M. Allman, "A Framework for Defining 425 Empirical Bulk Transfer Capacity Metrics", RFC 3148, 426 July 2001. 428 [RFC5136] Chimento, P. and J. Ishac, "Defining Network Capacity", 429 RFC 5136, February 2008. 431 Author's Address 433 Al Morton 434 AT&T Labs 435 200 Laurel Avenue South 436 Middletown,, NJ 07748 437 USA 439 Phone: +1 732 420 1571 440 Fax: +1 732 368 1192 441 Email: acmorton@att.com 442 URI: http://home.comcast.net/~acmacm/