idnits 2.17.1 draft-morton-bmwg-b2b-frame-02.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- -- The draft header indicates that this document updates RFC2544, but the abstract doesn't seem to directly say this. It does mention RFC2544 though, so this could be OK. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == The document seems to lack the recommended RFC 2119 boilerplate, even if it appears to use RFC 2119 keywords -- however, there's a paragraph with a matching beginning. Boilerplate error? (The document does seem to have the reference to RFC 2119 which the ID-Checklist requires). (Using the creation date from RFC2544, updated by this document, for RFC5378 checks: 1999-03-01) -- 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 (July 1, 2018) is 2125 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- ** Obsolete normative reference: RFC 1944 (Obsoleted by RFC 2544) Summary: 1 error (**), 0 flaws (~~), 2 warnings (==), 3 comments (--). 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 Updates: 2544 (if approved) July 1, 2018 5 Intended status: Informational 6 Expires: January 2, 2019 8 Updates for the Back-to-back Frame Benchmark in RFC 2544 9 draft-morton-bmwg-b2b-frame-02 11 Abstract 13 Fundamental Benchmarking Methodologies for Network Interconnect 14 Devices of interest to the IETF are defined in RFC 2544. This memo 15 updates the procedures of the test to measure the Back-to-back frames 16 Benchmark of RFC 2544, based on further experience. 18 This memo updates Section 26.4 of RFC 2544. 20 Requirements Language 22 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 23 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 24 "OPTIONAL" in this document are to be interpreted as described in BCP 25 14[RFC2119] [RFC8174] when, and only when, they appear in all 26 capitals, as shown here. 28 Status of This Memo 30 This Internet-Draft is submitted in full conformance with the 31 provisions of BCP 78 and BCP 79. 33 Internet-Drafts are working documents of the Internet Engineering 34 Task Force (IETF). Note that other groups may also distribute 35 working documents as Internet-Drafts. The list of current Internet- 36 Drafts is at https://datatracker.ietf.org/drafts/current/. 38 Internet-Drafts are draft documents valid for a maximum of six months 39 and may be updated, replaced, or obsoleted by other documents at any 40 time. It is inappropriate to use Internet-Drafts as reference 41 material or to cite them other than as "work in progress." 43 This Internet-Draft will expire on January 2, 2019. 45 Copyright Notice 47 Copyright (c) 2018 IETF Trust and the persons identified as the 48 document authors. All rights reserved. 50 This document is subject to BCP 78 and the IETF Trust's Legal 51 Provisions Relating to IETF Documents 52 (https://trustee.ietf.org/license-info) in effect on the date of 53 publication of this document. Please review these documents 54 carefully, as they describe your rights and restrictions with respect 55 to this document. Code Components extracted from this document must 56 include Simplified BSD License text as described in Section 4.e of 57 the Trust Legal Provisions and are provided without warranty as 58 described in the Simplified BSD License. 60 Table of Contents 62 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 63 2. Scope and Goals . . . . . . . . . . . . . . . . . . . . . . . 3 64 3. Motivation . . . . . . . . . . . . . . . . . . . . . . . . . 3 65 4. Prerequisites . . . . . . . . . . . . . . . . . . . . . . . . 5 66 5. Back-to-back Frames . . . . . . . . . . . . . . . . . . . . . 5 67 5.1. Preparing the list of Frame sizes . . . . . . . . . . . . 5 68 5.2. Test for a Single Frame Size . . . . . . . . . . . . . . 6 69 5.3. Test Repetition . . . . . . . . . . . . . . . . . . . . . 6 70 5.4. Benchmark Calculations . . . . . . . . . . . . . . . . . 7 71 6. Reporting . . . . . . . . . . . . . . . . . . . . . . . . . . 7 72 7. Security Considerations . . . . . . . . . . . . . . . . . . . 8 73 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 8 74 9. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 8 75 10. References . . . . . . . . . . . . . . . . . . . . . . . . . 9 76 10.1. Normative References . . . . . . . . . . . . . . . . . . 9 77 10.2. Informative References . . . . . . . . . . . . . . . . . 10 78 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 10 80 1. Introduction 82 The IETF's fundamental Benchmarking Methodologies are defined 83 in[RFC2544], supported by the terms and definitions in [RFC1242], and 84 [RFC2544] actually obsoletes an earlier specification, [RFC1944]. 85 Over time, the benchmarking community has updated [RFC2544] several 86 times, including the Device Reset Benchmark [RFC6201], and the 87 important Applicability Statement [RFC6815] concerning use outside 88 the Isolated Test Environment (ITE) required for accurate 89 benchmarking. Other specifications implicitly update [RFC2544], such 90 as the IPv6 Benchmarking Methodologies in [RFC5180]. 92 Recent testing experience with the Back-to-back Frame test and 93 Benchmark in Section 26.4 of [RFC2544] indicates that an update is 94 warranted [OPNFV-2017] [VSPERF-b2b]. This memo describes the 95 rationale and provides the updated method. 97 [RFC2544] provides its own Requirements Language consistent with 98 [RFC2119], since [RFC1944] predates [RFC2119]. Thus, the 99 requirements presented in this memo are expressed in [RFC2119] terms, 100 and intended for those performing/reporting laboratory tests to 101 improve clarity and repeatability, and for those designing devices 102 that facilitate these tests. 104 2. Scope and Goals 106 The scope of this memo is to define an updated method to 107 unambiguously perform tests, measure the benchmark(s), and report the 108 results for Back-to-back Frames (presently described Section 26.4 of 109 [RFC2544]). 111 The goal is to provide more efficient test procedures where possible, 112 and to expand reporting with additional interpretation of the 113 results. 115 [RFC2544] Benchmarks rely on test conditions with constant frame 116 sizes, with the goal of understanding what network device capability 117 has been tested. Tests with the smallest size stress the header 118 processing capacity, and tests with the largest size stress the 119 overall bit processing capacity. Tests with sizes in-between may 120 determine the transition between these two capacities. However, 121 conditions simultaneously sending multiple frame sizes, such as those 122 described in [RFC6985], MUST NOT be used in Back-to-back Frame 123 testing. 125 Section 3 of [RFC8239] describes buffer size testing for physical 126 networking devices in a Data Center. The [RFC8239] methods measure 127 buffer latency directly with traffic on multiple ingress ports that 128 overload an egress port on the Device Under Test (DUT), and are not 129 subject to the revised calculations presented in this memo. 131 3. Motivation 133 Section 3.1 of [RFC1242] describes the rationale for the Back-to-back 134 Frames Benchmark. To summarize, there are several reasons that 135 devices on a network produce bursts of frames at the minimum allowed 136 spacing, and it is therefore worthwhile to understand the Device 137 Under Test (DUT) limit on the length of such bursts in practice. 138 Also, [RFC1242] states: 140 "Tests of this parameter are intended to determine the extent 141 of data buffering in the device." 143 After this test was defined, there have been occasional discussions 144 of the stability and repeatability of the results, both over time and 145 across labs. Fortunately, the Open Platform for Network Function 146 Virtualization (OPNFV) VSPERF project's Continuous Integration (CI) 147 testing routinely repeats Back-to-back Frame tests to verify that 148 test functionality has been maintained through development of the 149 test control programs. These tests were used as a basis to evaluate 150 stability and repeatability, even across lab set-ups when the test 151 platform was migrated to new DUT hardware at the end of 2016. 153 When the VSPERF CI results were examined [VSPERF-b2b], several 154 aspects of the results were considered notable: 156 1. Back-to-back Frame Benchmark was very consistent for some fixed 157 frame sizes, and somewhat variable for others. 159 2. The Back-to-back Frame length reported for large frame sizes was 160 unexpectedly long, and no explanation or measurement limit 161 condition was indicated. 163 3. Calculation of the extent of buffer time in the DUT helped to 164 explain the results observed with all frame sizes (for example, 165 some frame sizes cannot exceed the frame header processing rate 166 of the DUT and therefore no buffering occurs, therefore the 167 results depended on the test equipment and not the DUT). 169 4. It was found that the actual buffer time in the DUT could be 170 estimated using results from the Throughput tests conducted 171 according to Section 26.1 of [RFC2544], because it appears that 172 the DUT's frame processing rate may tend to increase the 173 estimate. 175 Further, if the Throughput tests of Section 26.1 of [RFC2544] are 176 conducted as a prerequisite test, the number of frame sizes required 177 for Back-to-back Frame Benchmarking can be reduced to one or more of 178 the small frame sizes, or the results for large frame sizes can be 179 noted as invalid in the results if tested anyway (these are the frame 180 sizes for which the back-to-back frame rate cannot exceed the exceed 181 the frame header processing rate of the DUT and no buffering occurs). 183 [VSPERF-b2b] provides the details of the calculation to estimate the 184 actual buffer storage available in the DUT, using results from the 185 Throughput tests for each frame size, and the maximum theoretical 186 frame rate for the DUT links (which constrain the minimum frame 187 spacing). Knowledge of approximate buffer storage size (in time or 188 bytes) may be useful to estimate whether frame losses will occur if 189 DUT forwarding is temporarily suspended in a production deployment, 190 due to an unexpected interruption of frame processing (an 191 interruption of duration greater than the estimated buffer would 192 certainly cause lost frames). 194 4. Prerequisites 196 The Test Setup MUST be consistent with Figure 1 of [RFC2544], or 197 Figure 2 when the tester's sender and recover are different devices. 198 Other mandatory testing aspects described in [RFC2544] MUST be 199 included, unless explicitly modified in the next section. 201 The ingress and egress link speeds and link layer protocols MUST be 202 specified and used to compute the maximum theoretical frame rate when 203 respecting the minimum inter-frame gap. 205 The test results for the Throughput Benchmark conducted according to 206 Section 26.1 of [RFC2544] for all [RFC2544]-RECOMMENDED frame sizes 207 MUST be available to reduce the tested frame size list, or to note 208 invalid results for individual frame sizes (because the burst length 209 may be essentially infinite for large frame sizes). 211 Note that: 213 o the Throughput and the Back-to-back Frame measurement 214 configuration traffic characteristics (unidirectional or bi- 215 directional) MUST match. 217 o the Throughput measurement MUST be under zero-loss conditions, 218 according to Section 26.1 of [RFC2544]. 220 The Back-to-back Benchmark described in Section 3.1 of [RFC1242] MUST 221 be measured directly by the tester. Additional measurement 222 requirements are described below in Section 5. 224 5. Back-to-back Frames 226 Objective: To characterize the ability of a DUT to process back-to- 227 back frames as defined in [RFC1242]. 229 The Procedure follows. 231 5.1. Preparing the list of Frame sizes 233 From the list of RECOMMENDED Frame sizes (Section 9 of [RFC2544]), 234 select the subset of Frame sizes whose measured Throughput was less 235 than the maximum theoretical Frame Rate. These are the only Frame 236 sizes where it is possible to produce a burst of frames that cause 237 the DUT buffers to fill and eventually overflow, producing one or 238 more discarded frames. 240 5.2. Test for a Single Frame Size 242 Each trial in the test requires the tester to send a burst of frames 243 (after idle time) with the minimum inter-frame gap, and to count the 244 corresponding frames forwarded by the DUT. 246 The duration of the trial MUST be at least 2 seconds, to allow DUT 247 buffers to deplete. 249 If all frames have been received, the tester increases the length of 250 the burst according to the search algorithm and performs another 251 trial. 253 If the received frame count is less than the number of frames in the 254 burst, then the limit of DUT processing and buffering may have been 255 exceeded, and the burst length is determined by the search algorithm 256 for the next trial. 258 @@@@ OPNFV VSPERF Test Results presented at IETF-102 will have 259 relevance for answering these questions. 261 @@@@ Should a particular search algorithm be included? @@@@ Yes, 262 probably more than one. There are multiple roles for benchmark tests 263 (Highly repetitive CI and Highly Accurate & Repeatable Experiments). 265 @@@@ Should the search include trial repetition whenever frame loss 266 is observed, to avoid the effects of background loss (un-related to 267 buffer overflow)? @@@@ Yes, OPNFV VSPERF results illustrate the value 268 of this, and ETSI NFV TST009 Annex B describes the theory behind this 269 new algorithm. 271 The Back-to-back Frame value is the longest burst of frames that the 272 DUT can successfully process and buffer without frame loss, as 273 determined from the series of trials. The tester may impose a 274 (configurable) minimum step size for burst length, and the step size 275 MUST be reported with the results (as this influences the accuracy 276 and variation of test results). 278 5.3. Test Repetition 280 The test MUST be repeated N times for each frame size in the subset 281 list, and each Back-to-back Frame value made available for further 282 processing (below). 284 5.4. Benchmark Calculations 286 For each Frame size, calculate the following summary statistics for 287 Back-to-back Frame values over the N tests: 289 o Average (Benchmark) 291 o Minimum 293 o Maximum 295 o Standard Deviation 297 Further, calculate the Implied DUT Buffer Time and the Corrected DUT 298 Buffer Time in seconds, as follows: 300 Implied DUT Buffer Time = 302 Average num of Back-to-back Frames / Max Theoretical Frame Rate 304 The formula above is simply expressing the Burst of Frames in units 305 of time. 307 Corrected DUT Buffer Time = 309 Measured Throughput 310 = Implied DUT Buffer Time * -------------------------- 311 Max Theoretical Frame Rate 313 The term on the far right in the formula for Corrected DUT Buffer 314 Time accounts for all the frames in the Burst that were transmitted 315 by the DUT *while the Burst of frames were sent in*. So, these frames 316 are not in the Buffer and the Buffer size is more accurately 317 estimated by excluding them. 319 6. Reporting 321 The back-to-back results SHOULD be reported in the format of a table 322 with a row for each of the tested frame sizes. There SHOULD be 323 columns for the frame size and for the resultant average frame count 324 for each type of data stream tested. 326 The number of tests Averaged for the Benchmark, N, MUST be reported. 328 The Minimum, Maximum, and Standard Deviation across all complete 329 tests SHOULD also be reported. 331 The Corrected DUT Buffer Time SHOULD also be reported. 333 If the tester operates using a maximum burst length in frames, then 334 this maximum length SHOULD be reported. 336 +--------------+----------------+----------------+------------------+ 337 | Frame Size, | Ave B2B | Min,Max,StdDev | Corrected Buff | 338 | octets | Length, frames | | Time, Sec | 339 +--------------+----------------+----------------+------------------+ 340 | 64 | 26000 | 25500,27000,20 | 0.00004 | 341 +--------------+----------------+----------------+------------------+ 343 Back-to-Back Frame Results 345 Static and configuration parameters: 347 Number of test repetitions, N 349 Minimum Step Size (during searches), in frames. 351 7. Security Considerations 353 Benchmarking activities as described in this memo are limited to 354 technology characterization using controlled stimuli in a laboratory 355 environment, with dedicated address space and the other constraints 356 [RFC2544]. 358 The benchmarking network topology will be an independent test setup 359 and MUST NOT be connected to devices that may forward the test 360 traffic into a production network, or misroute traffic to the test 361 management network. See [RFC6815]. 363 Further, benchmarking is performed on a "black-box" basis, relying 364 solely on measurements observable external to the DUT/SUT. 366 Special capabilities SHOULD NOT exist in the DUT/SUT specifically for 367 benchmarking purposes. Any implications for network security arising 368 from the DUT/SUT SHOULD be identical in the lab and in production 369 networks. 371 8. IANA Considerations 373 This memo makes no requests of IANA. 375 9. Acknowledgements 377 Thanks to Trevor Cooper, Sridhar Rao, and Martin Klozik of the VSPERF 378 project for many contributions to the testing [VSPERF-b2b]. Yoshiaki 379 Itou has also investigated the topic, and made useful suggestions. 381 10. References 383 10.1. Normative References 385 [RFC1242] Bradner, S., "Benchmarking Terminology for Network 386 Interconnection Devices", RFC 1242, DOI 10.17487/RFC1242, 387 July 1991, . 389 [RFC1944] Bradner, S. and J. McQuaid, "Benchmarking Methodology for 390 Network Interconnect Devices", RFC 1944, 391 DOI 10.17487/RFC1944, May 1996, 392 . 394 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 395 Requirement Levels", BCP 14, RFC 2119, 396 DOI 10.17487/RFC2119, March 1997, 397 . 399 [RFC2544] Bradner, S. and J. McQuaid, "Benchmarking Methodology for 400 Network Interconnect Devices", RFC 2544, 401 DOI 10.17487/RFC2544, March 1999, 402 . 404 [RFC5180] Popoviciu, C., Hamza, A., Van de Velde, G., and D. 405 Dugatkin, "IPv6 Benchmarking Methodology for Network 406 Interconnect Devices", RFC 5180, DOI 10.17487/RFC5180, May 407 2008, . 409 [RFC6201] Asati, R., Pignataro, C., Calabria, F., and C. Olvera, 410 "Device Reset Characterization", RFC 6201, 411 DOI 10.17487/RFC6201, March 2011, 412 . 414 [RFC6815] Bradner, S., Dubray, K., McQuaid, J., and A. Morton, 415 "Applicability Statement for RFC 2544: Use on Production 416 Networks Considered Harmful", RFC 6815, 417 DOI 10.17487/RFC6815, November 2012, 418 . 420 [RFC6985] Morton, A., "IMIX Genome: Specification of Variable Packet 421 Sizes for Additional Testing", RFC 6985, 422 DOI 10.17487/RFC6985, July 2013, 423 . 425 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 426 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 427 May 2017, . 429 10.2. Informative References 431 [OPNFV-2017] 432 Cooper, T., Morton, A., and S. Rao, "Dataplane 433 Performance, Capacity, and Benchmarking in OPNFV", June 434 2017, 435 . 438 [RFC8239] Avramov, L. and J. Rapp, "Data Center Benchmarking 439 Methodology", RFC 8239, DOI 10.17487/RFC8239, August 2017, 440 . 442 [VSPERF-b2b] 443 Morton, A., "Back2Back Testing Time Series (from CI)", 444 June 2017, . 448 Author's Address 450 Al Morton 451 AT&T Labs 452 200 Laurel Avenue South 453 Middletown,, NJ 07748 454 USA 456 Phone: +1 732 420 1571 457 Fax: +1 732 368 1192 458 Email: acmorton@att.com