idnits 2.17.1 draft-morton-bmwg-b2b-frame-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 : ---------------------------------------------------------------------------- -- 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 (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 (February 5, 2018) is 2243 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 (~~), 1 warning (==), 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) February 5, 2018 5 Intended status: Informational 6 Expires: August 9, 2018 8 Updates for the Back-to-back Frame Benchmark in RFC 2544 9 draft-morton-bmwg-b2b-frame-01 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 provisions 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", "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 https://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 August 9, 2018. 43 Copyright Notice 45 Copyright (c) 2018 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 (https://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 . . . . . . . . . . . . . . . . . . . . . . . . 2 61 2. Scope and Goals . . . . . . . . . . . . . . . . . . . . . . . 3 62 3. Motivation . . . . . . . . . . . . . . . . . . . . . . . . . 3 63 4. Prerequisites . . . . . . . . . . . . . . . . . . . . . . . . 5 64 5. Back-to-back Frames . . . . . . . . . . . . . . . . . . . . . 5 65 5.1. Preparing the list of Frame sizes . . . . . . . . . . . . 5 66 5.2. Test for a Single Frame Size . . . . . . . . . . . . . . 6 67 5.3. Test Repetition . . . . . . . . . . . . . . . . . . . . . 6 68 5.4. Benchmark Calculations . . . . . . . . . . . . . . . . . 6 69 6. Reporting . . . . . . . . . . . . . . . . . . . . . . . . . . 7 70 7. Security Considerations . . . . . . . . . . . . . . . . . . . 8 71 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 8 72 9. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 8 73 10. References . . . . . . . . . . . . . . . . . . . . . . . . . 8 74 10.1. Normative References . . . . . . . . . . . . . . . . . . 8 75 10.2. Informative References . . . . . . . . . . . . . . . . . 9 76 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 9 78 1. Introduction 80 The IETF's fundamental Benchmarking Methodologies are defined 81 in[RFC2544], supported by the terms and definitions in [RFC1242], and 82 [RFC2544] actually obsoletes an earlier specification, [RFC1944]. 83 Over time, the benchmarking community has updated [RFC2544] several 84 times, including the Device Reset Benchmark [RFC6201], and the 85 important Applicability Statement [RFC6815] concerning use outside 86 the Isolated Test Environment (ITE) required for accurate 87 benchmarking. Other specifications implicitly update [RFC2544], such 88 as the IPv6 Benchmarking Methodologies in [RFC5180]. 90 Recent testing experience with the Back-to-back Frame test and 91 Benchmark in Section 26.4 of [RFC2544] indicates that an update is 92 warranted [OPNFV-2017] [VSPERF-b2b]. This memo describes the 93 rationale and provides the updated method. 95 [RFC2544] provides its own Requirements Language consistent with 96 [RFC2119], since [RFC1944] predates [RFC2119]. Thus, the 97 requirements presented in this memo are expressed in [RFC2119] terms, 98 and intended for those performing/reporting laboratory tests to 99 improve clarity and repeatability, and for those designing devices 100 that facilitate these tests. 102 2. Scope and Goals 104 The scope of this memo is to define an updated method to 105 unambiguously perform tests, measure the benchmark(s), and report the 106 results for Back-to-back Frames (presently described Section 26.4 of 107 [RFC2544]). 109 The goal is to provide more efficient test procedures where possible, 110 and to expand reporting with additional interpretation of the 111 results. 113 [RFC2544] Benchmarks rely on test conditions with constant frame 114 sizes, with the goal of understanding what network device capability 115 has been tested. Tests with the smallest size stress the header 116 processing capacity, and tests with the largest size stress the 117 overall bit processing capacity. Tests with sizes in-between may 118 determine the transition between these two capacities. However, 119 conditions simultaneously sending multiple frame sizes, such as those 120 described in [RFC6985], MUST NOT be used in Back-to-back Frame 121 testing. 123 3. Motivation 125 Section 3.1 of [RFC1242] describes the rationale for the Back-to-back 126 Frames Benchmark. To summarize, there are several reasons that 127 devices on a network produce bursts of frames at the minimum allowed 128 spacing, and it is therefore worthwhile to understand the Device 129 Under Test (DUT) limit on the length of such bursts in practice. 130 Also, [RFC1242] states: 132 "Tests of this parameter are intended to determine the extent 133 of data buffering in the device." 135 After this test was defined, there have been occasional discussions 136 of the stability and repeatability of the results, both over time and 137 across labs. Fortunately, the Open Platform for Network Function 138 Virtualization (OPNFV) VSPERF project's Continuous Integration (CI) 139 testing routinely repeats Back-to-back Frame tests to verify that 140 test functionality has been maintained through development of the 141 test control programs. These tests were used as a basis to evaluate 142 stability and repeatability, even across lab set-ups when the test 143 platform was migrated to new DUT hardware at the end of 2016. 145 When the VSPERF CI results were examined [VSPERF-b2b], several 146 aspects of the results were considered notable: 148 1. Back-to-back Frame Benchmark was very consistent for some fixed 149 frame sizes, and somewhat variable for others. 151 2. The Back-to-back Frame length reported for large frame sizes was 152 unexpectedly long, and no explanation or measurement limit 153 condition was indicated. 155 3. Calculation of the extent of buffer time in the DUT helped to 156 explain the results observed with all frame sizes (for example, 157 some frame sizes cannot exceed the frame header processing rate 158 of the DUT and therefore no buffering occurs, therefore the 159 results depended on the test equipment and not the DUT). 161 4. It was found that the actual buffer time in the DUT could be 162 estimated using results from the Throughput tests conducted 163 according to Section 26.1 of [RFC2544], because it appears that 164 the DUT's frame processing rate may tend to increase the 165 estimate. 167 Further, if the Throughput tests of Section 26.1 of [RFC2544] are 168 conducted as a prerequisite test, the number of frame sizes required 169 for Back-to-back Frame Benchmarking can be reduced to one or more of 170 the small frame sizes, or the results for large frame sizes can be 171 noted as invalid in the results if tested anyway (these are the frame 172 sizes for which the back-to-back frame rate cannot exceed the exceed 173 the frame header processing rate of the DUT and no buffering occurs). 175 [VSPERF-b2b] provides the details of the calculation to estimate the 176 actual buffer storage available in the DUT, using results from the 177 Throughput tests for each frame size, and the maximum theoretical 178 frame rate for the DUT links (which constrain the minimum frame 179 spacing). Knowledge of approximate buffer storage size (in time or 180 bytes) may be useful to estimate whether frame losses will occur if 181 DUT forwarding is temporarily suspended in a production deployment, 182 due to an unexpected interruption of frame processing (an 183 interruption of duration greater than the estimated buffer would 184 certainly cause lost frames). 186 4. Prerequisites 188 The Test Setup MUST be consistent with Figure 1 of [RFC2544], or 189 Figure 2 when the tester's sender and recover are different devices. 190 Other mandatory testing aspects described in [RFC2544] MUST be 191 included, unless explicitly modified in the next section. 193 The ingress and egress link speeds and link layer protocols MUST be 194 specified and used to compute the maximum theoretical frame rate when 195 respecting the minimum inter-frame gap. 197 The test results for the Throughput Benchmark conducted according to 198 Section 26.1 of [RFC2544] for all [RFC2544]-RECOMMENDED frame sizes 199 MUST be available to reduce the tested frame size list, or to note 200 invalid results for individual frame sizes (because the burst length 201 may be essentially infinite for large frame sizes). 203 Note that: 205 o the Throughput and the Back-to-back Frame measurement 206 configuration traffic characteristics (unidirectional or bi- 207 directional) MUST match. 209 o the Throughput measurement MUST be under zero-loss conditions, 210 according to Section 26.1 of [RFC2544]. 212 The Back-to-back Benchmark described in Section 3.1 of [RFC1242] MUST 213 be measured directly by the tester. Additional measurement 214 requirements are described below in Section 5. 216 5. Back-to-back Frames 218 Objective: To characterize the ability of a DUT to process back-to- 219 back frames as defined in [RFC1242]. 221 The Procedure follows. 223 5.1. Preparing the list of Frame sizes 225 From the list of RECOMMENDED Frame sizes (Section 9 of [RFC2544]), 226 select the subset of Frame sizes whose measured Throughput was less 227 than the maximum theoretical Frame Rate. These are the only Frame 228 sizes where it is possible to produce a burst of frames that cause 229 the DUT buffers to fill and eventually overflow, producing one or 230 more discarded frames. 232 5.2. Test for a Single Frame Size 234 Each trial in the test requires the tester to send a burst of frames 235 (after idle time) with the minimum inter-frame gap, and to count the 236 corresponding frames forwarded by the DUT. 238 The duration of the trial MUST be at least 2 seconds, to allow DUT 239 buffers to deplete. 241 If all frames have been received, the tester increases the length of 242 the burst and performs another trial. 244 If the received frame count is less than the number of frames in the 245 burst, then the limit of DUT processing and buffering may have been 246 exceeded, and the burst length is reduced for the next trial. 248 @@@@ Should a particular search algorithm be included? 250 @@@@ Should the search include trial repetition whenever frame loss 251 is observed, to avoid the effects of background loss (un-related to 252 buffer overflow)? 254 The Back-to-back Frame value is the longest burst of frames that the 255 DUT can successfully process and buffer without frame loss, as 256 determined from the series of trials. The tester may impose a 257 (configurable) minimum step size for burst length, and the step size 258 MUST be reported with the results (as this influences the accuracy 259 and variation of test results). 261 5.3. Test Repetition 263 The test MUST be repeated N times for each frame size in the subset 264 list, and each Back-to-back Frame value made available for further 265 processing (below). 267 5.4. Benchmark Calculations 269 For each Frame size, calculate the following summary statistics for 270 Back-to-back Frame values over the N tests: 272 o Average (Benchmark) 274 o Minimum 276 o Maximum 278 o Standard Deviation 279 Further, calculate the Implied DUT Buffer Time and the Corrected DUT 280 Buffer Time in seconds, as follows: 282 Implied DUT Buffer Time = 284 Average num of Back-to-back Frames / Max Theoretical Frame Rate 286 Corrected DUT Buffer Time = 288 Measured Throughput 289 Implied DUT Buffer Time * -------------------------- 290 Max Theoretical Frame Rate 292 6. Reporting 294 The back-to-back results SHOULD be reported in the format of a table 295 with a row for each of the tested frame sizes. There SHOULD be 296 columns for the frame size and for the resultant average frame count 297 for each type of data stream tested. 299 The number of tests Averaged for the Benchmark, N, MUST be reported. 301 The Minimum, Maximum, and Standard Deviation across all complete 302 tests SHOULD also be reported. 304 The Corrected DUT Buffer Time SHOULD also be reported. 306 If the tester operates using a maximum burst length in frames, then 307 this maximum length SHOULD be reported. 309 +--------------+----------------+----------------+------------------+ 310 | Frame Size, | Ave B2B | Min,Max,StdDev | Corrected Buff | 311 | octets | Length, frames | | Time, Sec | 312 +--------------+----------------+----------------+------------------+ 313 | 64 | 26000 | 25500,27000,20 | 0.00004 | 314 +--------------+----------------+----------------+------------------+ 316 Back-to-Back Frame Results 318 Static and configuration parameters: 320 Number of test repetitions, N 322 Minimum Step Size (during searches), in frames. 324 7. Security Considerations 326 Benchmarking activities as described in this memo are limited to 327 technology characterization using controlled stimuli in a laboratory 328 environment, with dedicated address space and the other constraints 329 [RFC2544]. 331 The benchmarking network topology will be an independent test setup 332 and MUST NOT be connected to devices that may forward the test 333 traffic into a production network, or misroute traffic to the test 334 management network. See [RFC6815]. 336 Further, benchmarking is performed on a "black-box" basis, relying 337 solely on measurements observable external to the DUT/SUT. 339 Special capabilities SHOULD NOT exist in the DUT/SUT specifically for 340 benchmarking purposes. Any implications for network security arising 341 from the DUT/SUT SHOULD be identical in the lab and in production 342 networks. 344 8. IANA Considerations 346 This memo makes no requests of IANA. 348 9. Acknowledgements 350 Thanks to Trevor Cooper, Sridhar Rao, and Martin Klozik of the VSPERF 351 project for many contributions to the testing [VSPERF-b2b]. 353 10. References 355 10.1. Normative References 357 [RFC1242] Bradner, S., "Benchmarking Terminology for Network 358 Interconnection Devices", RFC 1242, DOI 10.17487/RFC1242, 359 July 1991, . 361 [RFC1944] Bradner, S. and J. McQuaid, "Benchmarking Methodology for 362 Network Interconnect Devices", RFC 1944, 363 DOI 10.17487/RFC1944, May 1996, 364 . 366 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 367 Requirement Levels", BCP 14, RFC 2119, 368 DOI 10.17487/RFC2119, March 1997, 369 . 371 [RFC2544] Bradner, S. and J. McQuaid, "Benchmarking Methodology for 372 Network Interconnect Devices", RFC 2544, 373 DOI 10.17487/RFC2544, March 1999, 374 . 376 [RFC5180] Popoviciu, C., Hamza, A., Van de Velde, G., and D. 377 Dugatkin, "IPv6 Benchmarking Methodology for Network 378 Interconnect Devices", RFC 5180, DOI 10.17487/RFC5180, May 379 2008, . 381 [RFC6201] Asati, R., Pignataro, C., Calabria, F., and C. Olvera, 382 "Device Reset Characterization", RFC 6201, 383 DOI 10.17487/RFC6201, March 2011, 384 . 386 [RFC6815] Bradner, S., Dubray, K., McQuaid, J., and A. Morton, 387 "Applicability Statement for RFC 2544: Use on Production 388 Networks Considered Harmful", RFC 6815, 389 DOI 10.17487/RFC6815, November 2012, 390 . 392 [RFC6985] Morton, A., "IMIX Genome: Specification of Variable Packet 393 Sizes for Additional Testing", RFC 6985, 394 DOI 10.17487/RFC6985, July 2013, 395 . 397 10.2. Informative References 399 [OPNFV-2017] 400 Cooper, T., Morton, A., and S. Rao, "Dataplane 401 Performance, Capacity, and Benchmarking in OPNFV", June 402 2017, 403 . 406 [VSPERF-b2b] 407 Morton, A., "Back2Back Testing Time Series (from CI)", 408 June 2017, . 412 Author's Address 413 Al Morton 414 AT&T Labs 415 200 Laurel Avenue South 416 Middletown,, NJ 07748 417 USA 419 Phone: +1 732 420 1571 420 Fax: +1 732 368 1192 421 Email: acmorton@att.com