idnits 2.17.1 draft-ietf-ippm-active-passive-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 doesn't use any RFC 2119 keywords, yet seems to have RFC 2119 boilerplate text. -- The document date (September 6, 2015) is 3155 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Missing Reference: 'RFC 5644' is mentioned on line 304, but not defined == Outdated reference: A later version (-13) exists of draft-ietf-ippm-6man-pdm-option-00 == Outdated reference: A later version (-06) exists of draft-chen-ippm-coloring-based-ipfpm-framework-04 Summary: 0 errors (**), 0 flaws (~~), 5 warnings (==), 1 comment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 September 6, 2015 5 Expires: March 9, 2016 7 Active and Passive Metrics and Methods (and everything in-between, or 8 Hybrid) 9 draft-ietf-ippm-active-passive-01 11 Abstract 13 This memo provides clear definitions for Active and Passive 14 performance assessment. The construction of Metrics and Methods can 15 be described as Active or Passive. Some methods may use a subset of 16 both active and passive attributes, and we refer to these as Hybrid 17 Methods. 19 Status of This Memo 21 This Internet-Draft is submitted in full conformance with the 22 provisions of BCP 78 and BCP 79. 24 Internet-Drafts are working documents of the Internet Engineering 25 Task Force (IETF). Note that other groups may also distribute 26 working documents as Internet-Drafts. The list of current Internet- 27 Drafts is at http://datatracker.ietf.org/drafts/current/. 29 Internet-Drafts are draft documents valid for a maximum of six months 30 and may be updated, replaced, or obsoleted by other documents at any 31 time. It is inappropriate to use Internet-Drafts as reference 32 material or to cite them other than as "work in progress." 34 This Internet-Draft will expire on March 9, 2016. 36 Copyright Notice 38 Copyright (c) 2015 IETF Trust and the persons identified as the 39 document authors. All rights reserved. 41 This document is subject to BCP 78 and the IETF Trust's Legal 42 Provisions Relating to IETF Documents 43 (http://trustee.ietf.org/license-info) in effect on the date of 44 publication of this document. Please review these documents 45 carefully, as they describe your rights and restrictions with respect 46 to this document. Code Components extracted from this document must 47 include Simplified BSD License text as described in Section 4.e of 48 the Trust Legal Provisions and are provided without warranty as 49 described in the Simplified BSD License. 51 Table of Contents 53 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 54 1.1. Requirements Language . . . . . . . . . . . . . . . . . . 3 55 2. Purpose and Scope . . . . . . . . . . . . . . . . . . . . . . 3 56 3. Terms and Definitions . . . . . . . . . . . . . . . . . . . . 3 57 3.1. Performance Metric . . . . . . . . . . . . . . . . . . . 3 58 3.2. Method of Measurement . . . . . . . . . . . . . . . . . . 4 59 3.3. Observation Point . . . . . . . . . . . . . . . . . . . . 4 60 3.4. Active Methods . . . . . . . . . . . . . . . . . . . . . 4 61 3.5. Active Metric . . . . . . . . . . . . . . . . . . . . . . 4 62 3.6. Passive Methods . . . . . . . . . . . . . . . . . . . . . 5 63 3.7. Passive Metric . . . . . . . . . . . . . . . . . . . . . 5 64 3.8. Hybrid Methods and Metrics . . . . . . . . . . . . . . . 6 65 4. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . 7 66 4.1. Graphical Representation . . . . . . . . . . . . . . . . 7 67 4.2. Discussion of PDM . . . . . . . . . . . . . . . . . . . . 9 68 4.3. Discussion of "Coloring" Method . . . . . . . . . . . . . 10 69 5. Security considerations . . . . . . . . . . . . . . . . . . . 10 70 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 10 71 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 10 72 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 11 73 8.1. Normative References . . . . . . . . . . . . . . . . . . 11 74 8.2. Informative References . . . . . . . . . . . . . . . . . 11 75 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 12 77 1. Introduction 79 The adjectives "active" and "passive" have been used for many years 80 to distinguish two different classes of Internet performance 81 assessment. The first Passive and Active Measurement (PAM) 82 Conference was held in 2000, but the earliest proceedings available 83 on-line are from the second PAM conference in 2001 84 [https://www.ripe.net/ripe/meetings/pam-2001]. 86 The notions of "active" and "passive" are well-established. In 87 general: 89 An Active metric or method depends on a dedicated measurement 90 packet stream and observations of the stream. 92 A Passive metric or method depends *solely* on observation of one 93 or more existing packet streams. The streams only serve 94 measurement when they are observed for that purpose, and are 95 present whether measurements take place or not. 97 As new techniques for assessment emerge it is helpful to have clear 98 definitions of these notions. This memo provides more detailed 99 definitions, defines a new category for combinations of traditional 100 active and passive techniques, and discusses means to evaluate new 101 techniques as they emerge. 103 This memo provides definitions for Active and Passive Metrics and 104 Methods based on long usage in the Internet measurement community, 105 and especially the Internet Engineering Task Force. This memo also 106 describes the comnination of fundamental Active and Passive 107 categories, which are called Hybrid Methods and Metrics. 109 1.1. Requirements Language 111 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 112 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 113 document are to be interpreted as described in RFC 2119 [RFC2119]. 115 2. Purpose and Scope 117 The scope of this memo is to define and describe Active and Passive 118 versions of metrics and methods which are consistent with the long- 119 time usage of these adjectives in the Internet measurement community 120 and especially the Internet Engineering Task Force. Since the 121 science of measurement is expanding, we provide a category for 122 combinations of the traditional extremes, treating Active and Passive 123 as a continuum and designating combinations of their attributes as 124 Hybrid methods. 126 Further, this memo's purpose includes describing multiple dimensions 127 in which to evaluate methods as they emerge. 129 3. Terms and Definitions 131 This section defines the key terms of the memo. Some definitions use 132 the notion of "stream of interest" which is synonymous with 133 "population of interest" defined in clause 6.1.1 of ITU-T 134 Recommendation Y.1540 [Y.1540]. The definitions are consistent with 135 [I-D.zheng-ippm-framework-passive]. 137 3.1. Performance Metric 139 The standard definition of a quantity, produced in an assessment of 140 performance and/or reliability of the network, which has an intended 141 utility and is carefully specified to convey the exact meaning of a 142 measured value. (This definition is consistent with that of 143 Performance Metric in RFC 2330 and RFC 6390). 145 3.2. Method of Measurement 147 The procedure or set of operations having the object of determining a 148 Measured Value or Measurement Result. 150 3.3. Observation Point 152 See section 2 of [RFC7011] for this definition (a location in the 153 network where packets can be observed), and related definitions. The 154 comparable term defined in IETF literature on Active measurement is 155 Measurement Point, see section 4.1 of [RFC5835]. Two terms have come 156 into use describing similar actions at the identified point in the 157 network path. 159 3.4. Active Methods 161 Active measurement methods have the following attributes: 163 1. Commonly, the packet stream of interest is generated as the basis 164 of measurement. Another packet stream may be generated to 165 increase traffic load, but the loading stream itself may not be 166 measured. 168 2. The packets in the stream of interest have fields (or are 169 augmented or modified to include fields) which are dedicated to 170 measurement. Since measurement usually requires determining the 171 corresponding packets at multiple measurement points, a sequence 172 number is the most common information dedicated to measurement. 174 3. The Source and Destination of the packet stream of interest are 175 usually known a' priori. 177 4. The characteristics of the packet stream of interest are known at 178 the Source at least, and may be communicated to Destination as 179 part of the method. 181 When adding traffic to the network for measurement, Active Methods 182 influence the quantities measured to some degree, and those 183 performing tests should take steps to quantify the effect(s) and/or 184 minimize such effects. 186 3.5. Active Metric 188 An Active Metric incorporates one or more of the aspects of Active 189 Methods in the metric definition. 191 For example, IETF metrics for IP performance (developed according to 192 the [RFC2330] framework) include the Source packet stream 193 characteristics as metric input parameters, and also specify the 194 packet characteristics (Type-P) and Source and Destination IP 195 addresses (with their implications on both stream treatment and 196 interfaces associated with measurement points). 198 3.6. Passive Methods 200 Passive measurement methods are 202 o based solely on observations of undisturbed and unmodified packet 203 stream of interest 205 o dependent on the existence of one or more packet streams to supply 206 the stream of interest 208 o dependent on the presence of the packet stream of interest at one 209 or more designated observation points. 211 Some passive methods simply observe and collect information on all 212 packets that pass Observation Point(s), while others filter the 213 packets as a first step and only collect information on packets that 214 match the filter criteria, and thereby narrow the stream of interest. 216 It is common that passive methods are conducted at one or more 217 Observation Points. Passive methods to assess Performance Metrics 218 often require multiple observation points, e.g., to assess latency of 219 packet transfer across a network path between two Observation Points. 220 In this case, the observed packets must include enough information to 221 determine the corresponding packets at different Observation Points. 223 Communication of the observations (in some form) to a collector is an 224 essential aspect of Passive Methods. In some configurations, the 225 traffic load associated with results export to a collector may 226 influence the network performance. However, the collection of 227 results is not unique to Passive Methods, and the load from 228 management and operations of measurement systems must always be 229 considered for potential effects on the measured values. 231 3.7. Passive Metric 233 Passive Metrics apply to observations of packet traffic (traffic 234 flows in [RFC7011]). 236 Passive performance metrics are assessed independent of the packets 237 or traffic flows, and solely through observation. Some refer to such 238 assessments as "out-of-band". 240 One example of passive performance metrics for IP packet transfer can 241 be found in ITU-T Recommendation Y.1540 [Y.1540], where the metrics 242 are defined on the basis of reference events as packet pass reference 243 points, so the metrics are agnostic to the distinction between active 244 and passive when the necessary packet correspondence can be derived 245 from the observed stream of interest as required. 247 3.8. Hybrid Methods and Metrics 249 Hybrid Methods are Methods of Measurement which use a combination of 250 Active Methods and Passive Methods, to assess Active Metrics, Passive 251 Metrics, or new metrics derived from the a' priori knowledge and 252 observations of the stream of interest. ITU-T Recommendation Y.1540 253 [Y.1540] defines metrics are applicable to the hybrid category, since 254 packet correspondence at different observation/reference points could 255 be derived from "fields which are dedicated to measurement", but 256 otherwise the methods are passive. 258 There are several types of Hybrid methods, as categorized below. 260 With respect to a *single* stream of interest, Hybrid Type I methods 261 fit in the continuum as follows, in terms of what happens at the 262 Source (or Observation Point nearby): 264 o If you generate the stream of interest => Active 266 o If you augment of modify a stream of interest => Hybrid Type I 268 o If you solely observe a stream of interest => Passive 270 We define Hybrid Type II as follows: Methods that employ two or more 271 different streams of interest with some degree of mutual coordination 272 (one or more Active streams and one or more undisturbed and 273 unmodified packet streams) to collect both Active and Passive Metrics 274 and enable enhanced characterization from additional joint analysis. 275 [I-D.trammell-ippm-hybrid-ps] presents a problem statement for Hybrid 276 Type II methods and metrics. Note that one or more Hybrid Type I 277 streams could be substituted for the Active streams or undisturbed 278 streams in the mutually coordinated set. It is the Type II Methods 279 where unique Hybrid Metrics are atnticipated to emerge. 281 Methods based on a combination of a single (generated) Active stream 282 and Passive observations applied to the stream of interest at 283 intermediate observation points are also a type of Hybrid Methods. 284 However, [RFC5644] already defines these as Spatial Metrics and 285 Methods. It is possible to replace the Active stream of [RFC5644] 286 with a Hybrid Type I stream and measure Spatial Metrics (but this was 287 un-anticipated when [RFC5644] was developed). 289 The Table below illustrates the categorization of methods (where 290 "Synthesis" refers to a combination of Active and Passive Method 291 attributes). 293 | Single Stream | Multiple Simultaneous 294 | of Interest | Streams of Interest 295 | | from Different Methods 296 ==================================================================== 297 Single Fundamental | Active or Passive | 298 Method | | 300 Synthesis of | Hybrid Type I | 301 Fundamental Methods | | 303 Multiple Methods | Spatial Metrics | Hybrid Type II 304 | [RFC 5644] | 306 4. Discussion 308 This section illustrates the definitions and presents some examples. 310 4.1. Graphical Representation 312 If we compare the Active and Passive Methods, there are at least two 313 dimensions on which methods can be evaluated. This evaluation space 314 may be useful when a method is a combination of the two alternative 315 methods. 317 The two dimensions (initially chosen) are: 319 1. The degree to which the stream of interest effects overall 320 network conditions experienced by that stream and other streams. 321 This is a key dimension for Active measurement error analysis. 322 (Comment: There is also the notion of time averages - a 323 measurement stream may have significant effect while it is 324 present, but the stream is only generated 0.1% of the time. On 325 the other hand, observations alone have no effect on network 326 performance. To keep things simple, we consider the stream 327 effect only when it is present.) 329 2. The degree to which stream characteristics are know a' priori. 330 There are methodological advantages of knowing the source stream 331 characteristics, and having complete control of the stream 332 characteristics. For example, knowing the number of packets in a 333 stream allows more efficient operation of the measurement 334 receiver, and so is an asset for active measurement methods. 335 Passive methods (with no sample filter) have few clues available 336 to anticipate what the protocol first packet observed will use or 337 how many packets will comprise the flow, but once the standard 338 protocol of a flow is known the possibilities narrow (for some 339 compliant flows). Therefore this is a key dimension for Passive 340 measurement error analysis. 342 There are a few examples we can plot on a two-dimensional space. We 343 can anchor the dimensions with reference point descriptions. 345 Effect of the measured stream on network conditions 346 ^ Max 347 |* Active using max capacity stream 348 | 349 | 350 | 351 | 352 |* Active using stream with load of typical user 353 | 354 | 355 | 356 |* Active using extremely sparse, randomized stream 357 | * PDM Passive 358 | Min * 359 +----------------------------------------------------------------| 360 | | 361 Stream No Stream 362 Characteristics Characteristics 363 completely Known 364 known 366 We recognize that method categorization could be based on additional 367 dimensions, but this would require a different graphical approach. 369 For example, "effect of stream of interest on network conditions" 370 could easily be further qualified into: 372 1. effect on the performance of the stream of interest itself: for 373 example, choosing a packet marking or DSCP resulting in domain 374 treatment as a real-time stream (as opposed to default/best- 375 effort marking. 377 2. effect on unmeasured streams that share the path and/or 378 bottlenecks: for example, an extremely sparse measured stream of 379 minimal size packets typically has little effect on other flows 380 (and itself), while a stream designed to characterize path 381 capacity may effect all other flows passing through the capacity 382 bottleneck (including itself). 384 3. effect on network conditions resulting in network adaptation: for 385 example, a network monitoring load and congestion conditions 386 might change routing, placing some flows to alternate paths to 387 mitigate the congestion. 389 At present, we have combined 1 and 2 on one axis, as examination of 390 examples indicates strong correlation of affects on this pair, and 391 network adaptation is not addressed. As suggestions emerge we will 392 re-examine the possibilities. 394 It is apparent that different methods of IP network measurement can 395 produce different results, even when measuring the same path at the 396 same time. The two dimensions of the graph help to understand how 397 the results might change with the method chosen. For example, an 398 Active Method to assess throughput adds some amount of traffic to the 399 network which might result in lower throughput for all streams. 400 However, a Passive Method to assess throughput can also err on the 401 low side due to unknown limitations of the hosts providing traffic, 402 competition for host resources, limitations of the network interface, 403 or private sub-networks that are not an intentional part of the path, 404 etc. And Hybrid Methods could easily suffer from both forms of 405 error. Another example of potential errors stems from the pitfalls 406 of using an Active stream with known bias, such as a periodic stream 407 defined in [RFC3432]. The strength of modelling periodic streams 408 (like VoIP) is a potential weakness when extending the measured 409 results to other application whose streams are non-periodic. The 410 solutions are to model the application streams more exactly with an 411 Active Method, or accept the risks and potential errors with the 412 Passive Method discussed above. 414 4.2. Discussion of PDM 416 In [I-D.ietf-ippm-6man-pdm-option], an IPv6 Option Header for 417 Performance and Diagnostic Measurements (PDM) is described which 418 (when added to the stream of interest at strategic interfaces) 419 supports performance measurements. This method processes a user 420 traffic stream and adds "fields which are dedicated to measurement". 421 Thus: 423 o The method may have a small effect on the measured stream and 424 other streams in the network. 426 o The measured stream has unknown characteristics until it is 427 processed to add the PDM Option header. 429 We conclude that this is a Hybrid Type I method, having at least one 430 characteristic of both active and passive methods. 432 4.3. Discussion of "Coloring" Method 434 Draft [I-D.tempia-opsawg-p3m], proposed to color packets by re- 435 writing a field of the stream at strategic interfaces to support 436 performance measurements. This method processes a user traffic 437 stream and inserts "fields which are dedicated to measurement". 438 Thus: 440 o The method may have a small effect on the measured stream and 441 other streams in the network (smaller than PDM above). 443 o The measured stream has unknown characteristics until it is 444 processed to add the coloring in the header, and the stream could 445 be measured and time-stamped during that process. 447 We note that [I-D.chen-ippm-coloring-based-ipfpm-framework] proposes 448 a method similar to [I-D.tempia-opsawg-p3m], and ippm-list discussion 449 indicates [I-D.chen-ippm-coloring-based-ipfpm-framework] may be 450 covered by the same IPR as [I-D.tempia-opsawg-p3m]. 452 We conclude that this is a Hybrid Type I method, having at least one 453 characteristic of both active and passive methods. 455 5. Security considerations 457 When considering privacy of those involved in measurement or those 458 whose traffic is measured, there is sensitive information 459 communicated and observed at observation and measurement points 460 described above. We refer the reader to the privacy considerations 461 described in the Large Scale Measurement of Broadband Performance 462 (LMAP) Framework [I-D.ietf-lmap-framework], which covers active and 463 passive measurement techniques and supporting material on measurement 464 context. 466 6. IANA Considerations 468 This memo makes no requests for IANA consideration. 470 7. Acknowledgements 472 Thanks to Mike Ackermann for asking the right question, and for 473 several suggestions on terminology. Brian Trammell provided key 474 terms and references for the passive category, and suggested ways to 475 expand the Hybrid description and types. Phil Eardley suggested some 476 hybrid scenaios for categorization as part of his review. Tiziano 477 Ionta reviewed the draft and suggested the classification for the 478 "coloring" method of measurement. Nalini Elkins identified several 479 areas for clarification following her review. Bill Jouris suggested 480 several editorial improvements. 482 8. References 484 8.1. Normative References 486 [RFC2330] Paxson, V., Almes, G., Mahdavi, J., and M. Mathis, 487 "Framework for IP Performance Metrics", RFC 2330, 488 DOI 10.17487/RFC2330, May 1998, 489 . 491 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 492 Requirement Levels", BCP 14, RFC 2119, 493 DOI 10.17487/RFC2119, March 1997, 494 . 496 [RFC3432] Raisanen, V., Grotefeld, G., and A. Morton, "Network 497 performance measurement with periodic streams", RFC 3432, 498 DOI 10.17487/RFC3432, November 2002, 499 . 501 [RFC5644] Stephan, E., Liang, L., and A. Morton, "IP Performance 502 Metrics (IPPM): Spatial and Multicast", RFC 5644, 503 DOI 10.17487/RFC5644, October 2009, 504 . 506 [RFC5835] Morton, A., Ed. and S. Van den Berghe, Ed., "Framework for 507 Metric Composition", RFC 5835, DOI 10.17487/RFC5835, April 508 2010, . 510 [RFC7011] Claise, B., Ed., Trammell, B., Ed., and P. Aitken, 511 "Specification of the IP Flow Information Export (IPFIX) 512 Protocol for the Exchange of Flow Information", STD 77, 513 RFC 7011, DOI 10.17487/RFC7011, September 2013, 514 . 516 8.2. Informative References 518 [I-D.ietf-lmap-framework] 519 Eardley, P., Morton, A., Bagnulo, M., Burbridge, T., 520 Aitken, P., and A. Akhter, "A framework for Large-Scale 521 Measurement of Broadband Performance (LMAP)", draft-ietf- 522 lmap-framework-14 (work in progress), April 2015. 524 [I-D.ietf-ippm-6man-pdm-option] 525 Elkins, N. and M. Ackermann, "IPv6 Performance and 526 Diagnostic Metrics (PDM) Destination Option", draft-ietf- 527 ippm-6man-pdm-option-00 (work in progress), June 2015. 529 [I-D.tempia-opsawg-p3m] 530 Capello, A., Cociglio, M., Castaldelli, L., and A. Bonda, 531 "A packet based method for passive performance 532 monitoring", draft-tempia-opsawg-p3m-04 (work in 533 progress), February 2014. 535 [I-D.chen-ippm-coloring-based-ipfpm-framework] 536 Chen, M., Zheng, L., Mirsky, G., and G. Fioccola, "IP Flow 537 Performance Measurement Framework", draft-chen-ippm- 538 coloring-based-ipfpm-framework-04 (work in progress), July 539 2015. 541 [I-D.zheng-ippm-framework-passive] 542 Zheng, L., Elkins, N., Lingli, D., Ackermann, M., and G. 543 Mirsky, "Framework for IP Passive Performance 544 Measurements", draft-zheng-ippm-framework-passive-03 (work 545 in progress), February 2015. 547 [I-D.trammell-ippm-hybrid-ps] 548 Trammell, B., Zheng, L., Berenguer, S., and M. Bagnulo, 549 "Hybrid Measurement using IPPM Metrics", draft-trammell- 550 ippm-hybrid-ps-01 (work in progress), February 2014. 552 [Y.1540] ITU-T Recommendation Y.1540, , "Internet protocol data 553 communication service - IP packet transfer and 554 availability performance parameters", March 2011. 556 Author's Address 558 Al Morton 559 AT&T Labs 560 200 Laurel Avenue South 561 Middletown, NJ 562 USA 564 Email: acmorton@att.com