idnits 2.17.1 draft-vsperf-bmwg-vswitch-opnfv-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 : ---------------------------------------------------------------------------- ** There are 17 instances of too long lines in the document, the longest one being 2 characters in excess of 72. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (October 14, 2015) is 3115 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Unused Reference: 'RFC2330' is defined on line 784, but no explicit reference was found in the text == Unused Reference: 'RFC2680' is defined on line 798, but no explicit reference was found in the text == Unused Reference: 'RFC2681' is defined on line 803, but no explicit reference was found in the text == Unused Reference: 'RFC3393' is defined on line 812, but no explicit reference was found in the text == Unused Reference: 'RFC3432' is defined on line 817, but no explicit reference was found in the text == Unused Reference: 'RFC4689' is defined on line 826, but no explicit reference was found in the text == Unused Reference: 'RFC5357' is defined on line 836, but no explicit reference was found in the text == Unused Reference: 'RFC5905' is defined on line 841, but no explicit reference was found in the text == Unused Reference: 'RFC6049' is defined on line 873, but no explicit reference was found in the text == Unused Reference: 'RFC6248' is defined on line 877, but no explicit reference was found in the text == Unused Reference: 'RFC6390' is defined on line 882, but no explicit reference was found in the text ** Obsolete normative reference: RFC 2679 (Obsoleted by RFC 7679) ** Obsolete normative reference: RFC 2680 (Obsoleted by RFC 7680) == Outdated reference: A later version (-05) exists of draft-ietf-bmwg-virtual-net-01 Summary: 3 errors (**), 0 flaws (~~), 13 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group M. Tahhan 3 Internet-Draft B. O'Mahony 4 Intended status: Informational Intel 5 Expires: April 16, 2016 A. Morton 6 AT&T Labs 7 October 14, 2015 9 Benchmarking Virtual Switches in OPNFV 10 draft-vsperf-bmwg-vswitch-opnfv-01 12 Abstract 14 This memo describes the progress of the Open Platform for NFV (OPNFV) 15 project on virtual switch performance "VSWITCHPERF". This project 16 intends to build on the current and completed work of the 17 Benchmarking Methodology Working Group in IETF, by referencing 18 existing literature. The Benchmarking Methodology Working Group has 19 traditionally conducted laboratory characterization of dedicated 20 physical implementations of internetworking functions. Therefore, 21 this memo begins to describe the additional considerations when 22 virtual switches are implemented in general-purpose hardware. The 23 expanded tests and benchmarks are also influenced by the OPNFV 24 mission to support virtualization of the "telco" infrastructure. 26 Requirements Language 28 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 29 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 30 document are to be interpreted as described in RFC 2119 [RFC2119]. 32 Status of This Memo 34 This Internet-Draft is submitted in full conformance with the 35 provisions of BCP 78 and BCP 79. 37 Internet-Drafts are working documents of the Internet Engineering 38 Task Force (IETF). Note that other groups may also distribute 39 working documents as Internet-Drafts. The list of current Internet- 40 Drafts is at http://datatracker.ietf.org/drafts/current/. 42 Internet-Drafts are draft documents valid for a maximum of six months 43 and may be updated, replaced, or obsoleted by other documents at any 44 time. It is inappropriate to use Internet-Drafts as reference 45 material or to cite them other than as "work in progress." 47 This Internet-Draft will expire on April 16, 2016. 49 Copyright Notice 51 Copyright (c) 2015 IETF Trust and the persons identified as the 52 document authors. All rights reserved. 54 This document is subject to BCP 78 and the IETF Trust's Legal 55 Provisions Relating to IETF Documents 56 (http://trustee.ietf.org/license-info) in effect on the date of 57 publication of this document. Please review these documents 58 carefully, as they describe your rights and restrictions with respect 59 to this document. Code Components extracted from this document must 60 include Simplified BSD License text as described in Section 4.e of 61 the Trust Legal Provisions and are provided without warranty as 62 described in the Simplified BSD License. 64 Table of Contents 66 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 67 2. Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 68 3. Benchmarking Considerations . . . . . . . . . . . . . . . . . 4 69 3.1. Comparison with Physical Network Functions . . . . . . . 4 70 3.2. Continued Emphasis on Black-Box Benchmarks . . . . . . . 4 71 3.3. New Configuration Parameters . . . . . . . . . . . . . . 4 72 3.4. Flow classification . . . . . . . . . . . . . . . . . . . 6 73 3.5. Benchmarks using Baselines with Resource Isolation . . . 7 74 4. VSWITCHPERF Specification Summary . . . . . . . . . . . . . . 8 75 5. 3x3 Matrix Coverage . . . . . . . . . . . . . . . . . . . . . 16 76 5.1. Speed of Activation . . . . . . . . . . . . . . . . . . . 17 77 5.2. Accuracy of Activation section . . . . . . . . . . . . . 17 78 5.3. Reliability of Activation . . . . . . . . . . . . . . . . 17 79 5.4. Scale of Activation . . . . . . . . . . . . . . . . . . . 17 80 5.5. Speed of Operation . . . . . . . . . . . . . . . . . . . 17 81 5.6. Accuracy of Operation . . . . . . . . . . . . . . . . . . 17 82 5.7. Reliability of Operation . . . . . . . . . . . . . . . . 17 83 5.8. Scalability of Operation . . . . . . . . . . . . . . . . 18 84 5.9. Summary . . . . . . . . . . . . . . . . . . . . . . . . . 18 85 6. Security Considerations . . . . . . . . . . . . . . . . . . . 18 86 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 19 87 8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 19 88 9. References . . . . . . . . . . . . . . . . . . . . . . . . . 19 89 9.1. Normative References . . . . . . . . . . . . . . . . . . 19 90 9.2. Informative References . . . . . . . . . . . . . . . . . 21 91 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 21 93 1. Introduction 95 Benchmarking Methodology Working Group (BMWG) has traditionally 96 conducted laboratory characterization of dedicated physical 97 implementations of internetworking functions. The Black-box 98 Benchmarks of Throughput, Latency, Forwarding Rates and others have 99 served our industry for many years. Now, Network Function 100 Virtualization (NFV) has the goal to transform how internetwork 101 functions are implemented, and therefore has garnered much attention. 103 This memo describes the progress of the Open Platform for NFV (OPNFV) 104 project on virtual switch performance characterization, 105 "VSWITCHPERF". This project intends to build on the current and 106 completed work of the Benchmarking Methodology Working Group in IETF, 107 by referencing existing literature. For example, currently the most 108 often referenced RFC is [RFC2544] (which depends on [RFC1242]) and 109 foundation of the benchmarking work in OPNFV is common and strong. 111 See https://wiki.opnfv.org/ 112 characterize_vswitch_performance_for_telco_nfv_use_cases for more 113 background, and the OPNFV website for general information: 114 https://www.opnfv.org/ 116 The authors note that OPNFV distinguishes itself from other open 117 source compute and networking projects through its emphasis on 118 existing "telco" services as opposed to cloud-computing. There are 119 many ways in which telco requirements have different emphasis on 120 performance dimensions when compared to cloud computing: support for 121 and transfer of isochronous media streams is one example. 123 Note also that the move to NFV Infrastructure has resulted in many 124 new benchmarking initiatives across the industry, and the authors are 125 currently doing their best to maintain alignment with many other 126 projects, and this Internet Draft is evidence of the efforts. 128 2. Scope 130 The primary purpose and scope of the memo is to inform BMWG of work- 131 in-progress that builds on the body of extensive literature and 132 experience. Additionally, once the initial information conveyed here 133 is received, this memo may be expanded to include more detail and 134 commentary from both BMWG and OPNFV communities, under BMWG's 135 chartered work to characterize the NFV Infrastructure (a virtual 136 switch is an important aspect of that infrastructure). 138 3. Benchmarking Considerations 140 This section highlights some specific considerations (from 141 [I-D.ietf-bmwg-virtual-net])related to Benchmarks for virtual 142 switches. The OPNFV project is sharing its present view on these 143 areas, as they develop their specifications in the Level Test Design 144 (LTD) document. 146 3.1. Comparison with Physical Network Functions 148 To compare the performance of virtual designs and implementations 149 with their physical counterparts, identical benchmarks are needed. 150 BMWG has developed specifications for many network functions this 151 memo re-uses existing benchmarks through references, and expands them 152 during development of new methods. A key configuration aspect is the 153 number of parallel cores required to achieve comparable performance 154 with a given physical device, or whether some limit of scale was 155 reached before the cores could achieve the comparable level. 157 It's unlikely that the virtual switch will be the only application 158 running on the SUT, so CPU utilization, Cache utilization, and Memory 159 footprint should also be recorded for the virtual implementations of 160 internetworking functions. 162 3.2. Continued Emphasis on Black-Box Benchmarks 164 External observations remain essential as the basis for Benchmarks. 165 Internal observations with fixed specification and interpretation 166 will be provided in parallel to assist the development of operations 167 procedures when the technology is deployed. 169 3.3. New Configuration Parameters 171 A key consideration when conducting any sort of benchmark is trying 172 to ensure the consistency and repeatability of test results. When 173 benchmarking the performance of a vSwitch there are many factors that 174 can affect the consistency of results, one key factor is matching the 175 various hardware and software details of the SUT. This section lists 176 some of the many new parameters which this project believes are 177 critical to report in order to achieve repeatability. 179 Hardware details including: 181 o Platform details 183 o Processor details 185 o Memory information (type and size) 186 o Number of enabled cores 188 o Number of cores used for the test 190 o Number of physical NICs, as well as their details (manufacturer, 191 versions, type and the PCI slot they are plugged into) 193 o NIC interrupt configuration 195 o BIOS version, release date and any configurations that were 196 modified 198 o CPU microcode level 200 o Memory DIMM configurations (quad rank performance may not be the 201 same as dual rank) in size, freq and slot locations 203 o PCI configuration parameters (payload size, early ack option...) 205 o Power management at all levels (ACPI sleep states, processor 206 package, OS...) 208 Software details including: 210 o OS parameters and behavior (text vs graphical no one typing at the 211 console on one system) 213 o OS version (for host and VNF) 215 o Kernel version (for host and VNF) 217 o GRUB boot parameters (for host and VNF) 219 o Hypervisor details (Type and version) 221 o Selected vSwitch, version number or commit id used 223 o vSwitch launch command line if it has been parameterised 225 o Memory allocation to the vSwitch 227 o which NUMA node it is using, and how many memory channels 229 o DPDK or any other SW dependency version number or commit id used 231 o Memory allocation to a VM - if it's from Hugpages/elsewhere 232 o VM storage type: snapshot/independent persistent/independent non- 233 persistent 235 o Number of VMs 237 o Number of Virtual NICs (vNICs), versions, type and driver 239 o Number of virtual CPUs and their core affinity on the host 241 o Number vNIC interrupt configuration 243 o Thread affinitization for the applications (including the vSwitch 244 itself) on the host 246 o Details of Resource isolation, such as CPUs designated for Host/ 247 Kernel (isolcpu) and CPUs designated for specific processes 248 (taskset). - Test duration. - Number of flows. 250 Test Traffic Information: 252 o Traffic type - UDP, TCP, IMIX / Other 254 o Packet Sizes 256 o Deployment Scenario 258 3.4. Flow classification 260 Virtual switches group packets into flows by processing and matching 261 particular packet or frame header information, or by matching packets 262 based on the input ports. Thus a flow can be thought of a sequence 263 of packets that have the same set of header field values or have 264 arrived on the same port. Performance results can vary based on the 265 parameters the vSwitch uses to match for a flow. The recommended 266 flow classification parameters for any vSwitch performance tests are: 267 the input port, the source IP address, the destination IP address and 268 the Ethernet protocol type field. It is essential to increase the 269 flow timeout time on a vSwitch before conducting any performance 270 tests that do not measure the flow setup time. Normally the first 271 packet of a particular stream will install the flow in the virtual 272 switch which adds an additional latency, subsequent packets of the 273 same flow are not subject to this latency if the flow is already 274 installed on the vSwitch. 276 3.5. Benchmarks using Baselines with Resource Isolation 278 This outline describes measurement of baseline with isolated 279 resources at a high level, which is the intended approach at this 280 time. 282 1. Baselines: 284 * Optional: Benchmark platform forwarding capability without a 285 vswitch or VNF for at least 72 hours (serves as a means of 286 platform validation and a means to obtain the base performance 287 for the platform in terms of its maximum forwarding rate and 288 latency). 290 Benchmark platform forwarding capability 292 __ 293 +--------------------------------------------------+ | 294 | +------------------------------------------+ | | 295 | | | | | 296 | | Simple Forwarding App | | Host 297 | | | | | 298 | +------------------------------------------+ | | 299 | | NIC | | | 300 +---+------------------------------------------+---+ __| 301 ^ : 302 | | 303 : v 304 +--------------------------------------------------+ 305 | | 306 | traffic generator | 307 | | 308 +--------------------------------------------------+ 310 * Benchmark VNF forwarding capability with direct connectivity 311 (vSwitch bypass, e.g., SR/IOV) for at least 72 hours (serves 312 as a means of VNF validation and a means to obtain the base 313 performance for the VNF in terms of its maximum forwarding 314 rate and latency). The metrics gathered from this test will 315 serve as a key comparison point for vSwitch bypass 316 technologies performance and vSwitch performance. 318 Benchmark VNF forwarding capability 320 __ 321 +--------------------------------------------------+ | 322 | +------------------------------------------+ | | 323 | | | | | 324 | | VNF | | | 325 | | | | | 326 | +------------------------------------------+ | | 327 | | Passthrough/SR-IOV | | Host 328 | +------------------------------------------+ | | 329 | | NIC | | | 330 +---+------------------------------------------+---+ __| 331 ^ : 332 | | 333 : v 334 +--------------------------------------------------+ 335 | | 336 | traffic generator | 337 | | 338 +--------------------------------------------------+ 340 * Benchmarking with isolated resources alone, with other 341 resources (both HW&SW) disabled Example, vSw and VM are SUT 343 * Benchmarking with isolated resources alone, leaving some 344 resources unused 346 * Benchmark with isolated resources and all resources occupied 348 2. Next Steps 350 * Limited sharing 352 * Production scenarios 354 * Stressful scenarios 356 4. VSWITCHPERF Specification Summary 358 The overall specification in preparation is referred to as a Level 359 Test Design (LTD) document, which will contain a suite of performance 360 tests. The base performance tests in the LTD are based on the pre- 361 existing specifications developed by BMWG to test the performance of 362 physical switches. These specifications include: 364 o [RFC2544] Benchmarking Methodology for Network Interconnect 365 Devices 367 o [RFC2889] Benchmarking Methodology for LAN Switching 369 o [RFC6201] Device Reset Characterization 371 o [RFC5481] Packet Delay Variation Applicability Statement 373 Some of the above/newer RFCs are being applied in benchmarking for 374 the first time, and represent a development challenge for test 375 equipment developers. Fortunately, many members of the testing 376 system community have engaged on the VSPERF project, including an 377 open source test system. 379 In addition to this, the LTD also re-uses the terminology defined by: 381 o [RFC2285] Benchmarking Terminology for LAN Switching Devices 383 o [RFC5481] Packet Delay Variation Applicability Statement 385 Specifications to be included in future updates of the LTD include: 387 o [RFC3918] Methodology for IP Multicast Benchmarking 389 o [RFC4737] Packet Reordering Metrics 391 As one might expect, the most fundamental internetworking 392 characteristics of Throughput and Latency remain important when the 393 switch is virtualized, and these benchmarks figure prominently in the 394 specification. 396 When considering characteristics important to "telco" network 397 functions, we must begin to consider additional performance metrics. 398 In this case, the project specifications have referenced metrics from 399 the IETF IP Performance Metrics (IPPM) literature. This means that 400 the [RFC2544] test of Latency is replaced by measurement of a metric 401 derived from IPPM's [RFC2679], where a set of statistical summaries 402 will be provided (mean, max, min, etc.). Further metrics planned to 403 be benchmarked include packet delay variation as defined by [RFC5481] 404 , reordering, burst behaviour, DUT availability, DUT capacity and 405 packet loss in long term testing at Throughput level, where some low- 406 level of background loss may be present and characterized. 408 Tests have been (or will be) designed to collect the metrics below: 410 o Throughput Tests to measure the maximum forwarding rate (in frames 411 per second or fps) and bit rate (in Mbps) for a constant load (as 412 defined by RFC1242) without traffic loss. 414 o Packet and Frame Delay Distribution Tests to measure average, min 415 and max packet and frame delay for constant loads. 417 o Packet Delay Tests to understand latency distribution for 418 different packet sizes and over an extended test run to uncover 419 outliers. 421 o Scalability Tests to understand how the virtual switch performs as 422 the number of flows, active ports, complexity of the forwarding 423 logic's configuration... it has to deal with increases. 425 o Stream Performance Tests (TCP, UDP) to measure bulk data transfer 426 performance, i.e. how fast systems can send and receive data 427 through the switch. 429 o Control Path and Datapath Coupling Tests, to understand how 430 closely coupled the datapath and the control path are as well as 431 the effect of this coupling on the performance of the DUT 432 (example: delay of the initial packet of a flow). 434 o CPU and Memory Consumption Tests to understand the virtual 435 switch's footprint on the system, usually conducted as auxiliary 436 measurements with benchmarks above. They include: CPU 437 utilization, Cache utilization and Memory footprint. 439 Future/planned test specs include: 441 o Request/Response Performance Tests (TCP, UDP) which measure the 442 transaction rate through the switch. 444 o Noisy Neighbour Tests, to understand the effects of resource 445 sharing on the performance of a virtual switch. 447 o Tests derived from examination of ETSI NFV Draft GS IFA003 448 requirements [IFA003] on characterization of acceleration 449 technologies applied to vswitches. 451 The flexibility of deployment of a virtual switch within a network 452 means that the BMWG IETF existing literature needs to be used to 453 characterize the performance of a switch in various deployment 454 scenarios. The deployment scenarios under consideration include: 456 Physical port to virtual switch to physical port 458 __ 459 +--------------------------------------------------+ | 460 | +--------------------+ | | 461 | | | | | 462 | | v | | Host 463 | +--------------+ +--------------+ | | 464 | | phy port | vSwitch | phy port | | | 465 +---+--------------+------------+--------------+---+ __| 466 ^ : 467 | | 468 : v 469 +--------------------------------------------------+ 470 | | 471 | traffic generator | 472 | | 473 +--------------------------------------------------+ 475 Physical port to virtual switch to VNF to virtual switch to physical 476 port 478 __ 479 +---------------------------------------------------+ | 480 | | | 481 | +-------------------------------------------+ | | 482 | | Application | | | 483 | +-------------------------------------------+ | | 484 | ^ : | | 485 | | | | | Guest 486 | : v | | 487 | +---------------+ +---------------+ | | 488 | | logical port 0| | logical port 1| | | 489 +---+---------------+-----------+---------------+---+ __| 490 ^ : 491 | | 492 : v __ 493 +---+---------------+----------+---------------+---+ | 494 | | logical port 0| | logical port 1| | | 495 | +---------------+ +---------------+ | | 496 | ^ : | | 497 | | | | | Host 498 | : v | | 499 | +--------------+ +--------------+ | | 500 | | phy port | vSwitch | phy port | | | 501 +---+--------------+------------+--------------+---+ __| 502 ^ : 503 | | 504 : v 505 +--------------------------------------------------+ 506 | | 507 | traffic generator | 508 | | 509 +--------------------------------------------------+ 511 Physical port to virtual switch to VNF to virtual switch to VNF to 512 virtual switch to physical port 514 __ 515 +----------------------+ +----------------------+ | 516 | Guest 1 | | Guest 2 | | 517 | +---------------+ | | +---------------+ | | 518 | | Application | | | | Application | | | 519 | +---------------+ | | +---------------+ | | 520 | ^ | | | ^ | | | 521 | | v | | | v | | Guests 522 | +---------------+ | | +---------------+ | | 523 | | logical ports | | | | logical ports | | | 524 | | 0 1 | | | | 0 1 | | | 525 +---+---------------+--+ +---+---------------+--+__| 526 ^ : ^ : 527 | | | | 528 : v : v _ 529 +---+---------------+---------+---------------+--+ | 530 | | 0 1 | | 3 4 | | | 531 | | logical ports | | logical ports | | | 532 | +---------------+ +---------------+ | | 533 | ^ | ^ | | | Host 534 | | |-----------------| v | | 535 | +--------------+ +--------------+ | | 536 | | phy ports | vSwitch | phy ports | | | 537 +---+--------------+----------+--------------+---+_| 538 ^ : 539 | | 540 : v 541 +--------------------------------------------------+ 542 | | 543 | traffic generator | 544 | | 545 +--------------------------------------------------+ 547 Physical port to virtual switch to VNF 549 __ 550 +---------------------------------------------------+ | 551 | | | 552 | +-------------------------------------------+ | | 553 | | Application | | | 554 | +-------------------------------------------+ | | 555 | ^ | | 556 | | | | Guest 557 | : | | 558 | +---------------+ | | 559 | | logical port 0| | | 560 +---+---------------+-------------------------------+ __| 561 ^ 562 | 563 : __ 564 +---+---------------+------------------------------+ | 565 | | logical port 0| | | 566 | +---------------+ | | 567 | ^ | | 568 | | | | Host 569 | : | | 570 | +--------------+ | | 571 | | phy port | vSwitch | | 572 +---+--------------+------------ -------------- ---+ __| 573 ^ 574 | 575 : 576 +--------------------------------------------------+ 577 | | 578 | traffic generator | 579 | | 580 +--------------------------------------------------+ 582 VNF to virtual switch to physical port 584 __ 585 +---------------------------------------------------+ | 586 | | | 587 | +-------------------------------------------+ | | 588 | | Application | | | 589 | +-------------------------------------------+ | | 590 | : | | 591 | | | | Guest 592 | v | | 593 | +---------------+ | | 594 | | logical port | | | 595 +-------------------------------+---------------+---+ __| 596 : 597 | 598 v __ 599 +------------------------------+---------------+---+ | 600 | | logical port | | | 601 | +---------------+ | | 602 | : | | 603 | | | | Host 604 | v | | 605 | +--------------+ | | 606 | vSwitch | phy port | | | 607 +-------------------------------+--------------+---+ __| 608 : 609 | 610 v 611 +--------------------------------------------------+ 612 | | 613 | traffic generator | 614 | | 615 +--------------------------------------------------+ 617 VNF to virtual switch to VNF 619 __ 620 +----------------------+ +----------------------+ | 621 | Guest 1 | | Guest 2 | | 622 | +---------------+ | | +---------------+ | | 623 | | Application | | | | Application | | | 624 | +---------------+ | | +---------------+ | | 625 | | | | ^ | | 626 | v | | | | | Guests 627 | +---------------+ | | +---------------+ | | 628 | | logical ports | | | | logical ports | | | 629 | | 0 | | | | 0 | | | 630 +---+---------------+--+ +---+---------------+--+__| 631 : ^ 632 | | 633 v : _ 634 +---+---------------+---------+---------------+--+ | 635 | | 1 | | 1 | | | 636 | | logical ports | | logical ports | | | 637 | +---------------+ +---------------+ | | 638 | | ^ | | Host 639 | L-----------------+ | | 640 | | | 641 | vSwitch | | 642 +------------------------------------------------+_| 644 A set of Deployment Scenario figures is available on the VSPERF Test 645 Methodology Wiki page [TestTopo]. 647 5. 3x3 Matrix Coverage 649 This section organizes the many existing test specifications into the 650 "3x3" matrix (introduced in [I-D.ietf-bmwg-virtual-net]). Because 651 the LTD specification ID names are quite long, this section is 652 organized into lists for each occupied cell of the matrix (not all 653 are occupied, also the matrix has grown to 3x4 to accommodate scale 654 metrics when displaying the coverage of many metrics/benchmarks). 656 The tests listed below assess the activation of paths in the data 657 plane, rather than the control plane. 659 A complete list of tests with short summaries is available on the 660 VSPERF "LTD Test Spec Overview" Wiki page [LTDoverV]. 662 5.1. Speed of Activation 664 o Activation.RFC2889.AddressLearningRate 666 o PacketLatency.InitialPacketProcessingLatency 668 5.2. Accuracy of Activation section 670 o CPDP.Coupling.Flow.Addition 672 5.3. Reliability of Activation 674 o Throughput.RFC2544.SystemRecoveryTime 676 o Throughput.RFC2544.ResetTime 678 5.4. Scale of Activation 680 o Activation.RFC2889.AddressCachingCapacity 682 5.5. Speed of Operation 684 o Throughput.RFC2544.PacketLossRate 686 o CPU.RFC2544.0PacketLoss 688 o Throughput.RFC2544.PacketLossRateFrameModification 690 o Throughput.RFC2544.BackToBackFrames 692 o Throughput.RFC2889.MaxForwardingRate 694 o Throughput.RFC2889.ForwardPressure 696 o Throughput.RFC2889.BroadcastFrameForwarding 698 5.6. Accuracy of Operation 700 o Throughput.RFC2889.ErrorFramesFiltering 702 o Throughput.RFC2544.Profile 704 5.7. Reliability of Operation 706 o Throughput.RFC2889.Soak 708 o Throughput.RFC2889.SoakFrameModification 709 o PacketDelayVariation.RFC3393.Soak 711 5.8. Scalability of Operation 713 o Scalability.RFC2544.0PacketLoss 715 o MemoryBandwidth.RFC2544.0PacketLoss.Scalability 717 5.9. Summary 719 |------------------------------------------------------------------------| 720 | | | | | | 721 | | SPEED | ACCURACY | RELIABILITY | SCALE | 722 | | | | | | 723 |------------------------------------------------------------------------| 724 | | | | | | 725 | Activation | X | X | X | X | 726 | | | | | | 727 |------------------------------------------------------------------------| 728 | | | | | | 729 | Operation | X | X | X | X | 730 | | | | | | 731 |------------------------------------------------------------------------| 732 | | | | | | 733 | De-activation | | | | | 734 | | | | | | 735 |------------------------------------------------------------------------| 737 6. Security Considerations 739 Benchmarking activities as described in this memo are limited to 740 technology characterization of a Device Under Test/System Under Test 741 (DUT/SUT) using controlled stimuli in a laboratory environment, with 742 dedicated address space and the constraints specified in the sections 743 above. 745 The benchmarking network topology will be an independent test setup 746 and MUST NOT be connected to devices that may forward the test 747 traffic into a production network, or misroute traffic to the test 748 management network. 750 Further, benchmarking is performed on a "black-box" basis, relying 751 solely on measurements observable external to the DUT/SUT. 753 Special capabilities SHOULD NOT exist in the DUT/SUT specifically for 754 benchmarking purposes. Any implications for network security arising 755 from the DUT/SUT SHOULD be identical in the lab and in production 756 networks. 758 7. IANA Considerations 760 No IANA Action is requested at this time. 762 8. Acknowledgements 764 The authors acknowledge 766 9. References 768 9.1. Normative References 770 [NFV.PER001] 771 "Network Function Virtualization: Performance and 772 Portability Best Practices", Group Specification ETSI GS 773 NFV-PER 001 V1.1.1 (2014-06), June 2014. 775 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 776 Requirement Levels", BCP 14, RFC 2119, 777 DOI 10.17487/RFC2119, March 1997, 778 . 780 [RFC2285] Mandeville, R., "Benchmarking Terminology for LAN 781 Switching Devices", RFC 2285, DOI 10.17487/RFC2285, 782 February 1998, . 784 [RFC2330] Paxson, V., Almes, G., Mahdavi, J., and M. Mathis, 785 "Framework for IP Performance Metrics", RFC 2330, 786 DOI 10.17487/RFC2330, May 1998, 787 . 789 [RFC2544] Bradner, S. and J. McQuaid, "Benchmarking Methodology for 790 Network Interconnect Devices", RFC 2544, 791 DOI 10.17487/RFC2544, March 1999, 792 . 794 [RFC2679] Almes, G., Kalidindi, S., and M. Zekauskas, "A One-way 795 Delay Metric for IPPM", RFC 2679, DOI 10.17487/RFC2679, 796 September 1999, . 798 [RFC2680] Almes, G., Kalidindi, S., and M. Zekauskas, "A One-way 799 Packet Loss Metric for IPPM", RFC 2680, 800 DOI 10.17487/RFC2680, September 1999, 801 . 803 [RFC2681] Almes, G., Kalidindi, S., and M. Zekauskas, "A Round-trip 804 Delay Metric for IPPM", RFC 2681, DOI 10.17487/RFC2681, 805 September 1999, . 807 [RFC2889] Mandeville, R. and J. Perser, "Benchmarking Methodology 808 for LAN Switching Devices", RFC 2889, 809 DOI 10.17487/RFC2889, August 2000, 810 . 812 [RFC3393] Demichelis, C. and P. Chimento, "IP Packet Delay Variation 813 Metric for IP Performance Metrics (IPPM)", RFC 3393, 814 DOI 10.17487/RFC3393, November 2002, 815 . 817 [RFC3432] Raisanen, V., Grotefeld, G., and A. Morton, "Network 818 performance measurement with periodic streams", RFC 3432, 819 DOI 10.17487/RFC3432, November 2002, 820 . 822 [RFC3918] Stopp, D. and B. Hickman, "Methodology for IP Multicast 823 Benchmarking", RFC 3918, DOI 10.17487/RFC3918, October 824 2004, . 826 [RFC4689] Poretsky, S., Perser, J., Erramilli, S., and S. Khurana, 827 "Terminology for Benchmarking Network-layer Traffic 828 Control Mechanisms", RFC 4689, DOI 10.17487/RFC4689, 829 October 2006, . 831 [RFC4737] Morton, A., Ciavattone, L., Ramachandran, G., Shalunov, 832 S., and J. Perser, "Packet Reordering Metrics", RFC 4737, 833 DOI 10.17487/RFC4737, November 2006, 834 . 836 [RFC5357] Hedayat, K., Krzanowski, R., Morton, A., Yum, K., and J. 837 Babiarz, "A Two-Way Active Measurement Protocol (TWAMP)", 838 RFC 5357, DOI 10.17487/RFC5357, October 2008, 839 . 841 [RFC5905] Mills, D., Martin, J., Ed., Burbank, J., and W. Kasch, 842 "Network Time Protocol Version 4: Protocol and Algorithms 843 Specification", RFC 5905, DOI 10.17487/RFC5905, June 2010, 844 . 846 [RFC6201] Asati, R., Pignataro, C., Calabria, F., and C. Olvera, 847 "Device Reset Characterization", RFC 6201, 848 DOI 10.17487/RFC6201, March 2011, 849 . 851 9.2. Informative References 853 [I-D.ietf-bmwg-virtual-net] 854 Morton, A., "Considerations for Benchmarking Virtual 855 Network Functions and Their Infrastructure", draft-ietf- 856 bmwg-virtual-net-01 (work in progress), September 2015. 858 [IFA003] "https://docbox.etsi.org/ISG/NFV/Open/Drafts/ 859 IFA003_Acceleration_-_vSwitch_Spec/". 861 [LTDoverV] 862 "LTD Test Spec Overview https://wiki.opnfv.org/wiki/ 863 vswitchperf_test_spec_review". 865 [RFC1242] Bradner, S., "Benchmarking Terminology for Network 866 Interconnection Devices", RFC 1242, DOI 10.17487/RFC1242, 867 July 1991, . 869 [RFC5481] Morton, A. and B. Claise, "Packet Delay Variation 870 Applicability Statement", RFC 5481, DOI 10.17487/RFC5481, 871 March 2009, . 873 [RFC6049] Morton, A. and E. Stephan, "Spatial Composition of 874 Metrics", RFC 6049, DOI 10.17487/RFC6049, January 2011, 875 . 877 [RFC6248] Morton, A., "RFC 4148 and the IP Performance Metrics 878 (IPPM) Registry of Metrics Are Obsolete", RFC 6248, 879 DOI 10.17487/RFC6248, April 2011, 880 . 882 [RFC6390] Clark, A. and B. Claise, "Guidelines for Considering New 883 Performance Metric Development", BCP 170, RFC 6390, 884 DOI 10.17487/RFC6390, October 2011, 885 . 887 [TestTopo] 888 "Test Topologies https://wiki.opnfv.org/vsperf/ 889 test_methodology". 891 Authors' Addresses 893 Maryam Tahhan 894 Intel 896 Email: maryam.tahhan@intel.com 897 Billy O'Mahony 898 Intel 900 Email: billy.o.mahony@intel.com 902 Al Morton 903 AT&T Labs 904 200 Laurel Avenue South 905 Middletown,, NJ 07748 906 USA 908 Phone: +1 732 420 1571 909 Fax: +1 732 368 1192 910 Email: acmorton@att.com 911 URI: http://home.comcast.net/~acmacm/