1 <?xml version="1.0" encoding="US-ASCII"?>
2 <!DOCTYPE rfc SYSTEM "rfc2629.dtd">
4 <?rfc tocompact="yes"?>
6 <?rfc tocindent="yes"?>
12 <?rfc subcompact="no"?>
13 <rfc category="info" docName="draft-vsperf-bmwg-vswitch-opnfv-02"
16 <title abbrev="Benchmarking vSwitches">Benchmarking Virtual Switches in
19 <author fullname="Maryam Tahhan" initials="M." surname="Tahhan">
20 <organization>Intel</organization>
39 <email>maryam.tahhan@intel.com</email>
45 <author fullname="Billy O'Mahony" initials="B." surname="O'Mahony">
46 <organization>Intel</organization>
65 <email>billy.o.mahony@intel.com</email>
71 <author fullname="Al Morton" initials="A." surname="Morton">
72 <organization>AT&T Labs</organization>
76 <street>200 Laurel Avenue South</street>
78 <city>Middletown,</city>
84 <country>USA</country>
87 <phone>+1 732 420 1571</phone>
89 <facsimile>+1 732 368 1192</facsimile>
91 <email>acmorton@att.com</email>
93 <uri>http://home.comcast.net/~acmacm/</uri>
97 <date day="21" month="March" year="2016"/>
100 <t>This memo describes the progress of the Open Platform for NFV (OPNFV)
101 project on virtual switch performance "VSWITCHPERF". This project
102 intends to build on the current and completed work of the Benchmarking
103 Methodology Working Group in IETF, by referencing existing literature.
104 The Benchmarking Methodology Working Group has traditionally conducted
105 laboratory characterization of dedicated physical implementations of
106 internetworking functions. Therefore, this memo begins to describe the
107 additional considerations when virtual switches are implemented in
108 general-purpose hardware. The expanded tests and benchmarks are also
109 influenced by the OPNFV mission to support virtualization of the "telco"
113 <note title="Requirements Language">
114 <t>The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
115 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
116 document are to be interpreted as described in <xref
117 target="RFC2119">RFC 2119</xref>.</t>
124 <section title="Introduction">
125 <t>Benchmarking Methodology Working Group (BMWG) has traditionally
126 conducted laboratory characterization of dedicated physical
127 implementations of internetworking functions. The Black-box Benchmarks
128 of Throughput, Latency, Forwarding Rates and others have served our
129 industry for many years. Now, Network Function Virtualization (NFV) has
130 the goal to transform how internetwork functions are implemented, and
131 therefore has garnered much attention.</t>
133 <t>This memo summarizes the progress of the Open Platform for NFV
134 (OPNFV) project on virtual switch performance characterization,
135 "VSWITCHPERF", through the Brahmaputra (second) release <xref
136 target="BrahRel"/>. This project intends to build on the current and
137 completed work of the Benchmarking Methodology Working Group in IETF, by
138 referencing existing literature. For example, currently the most often
139 referenced RFC is <xref target="RFC2544"/> (which depends on <xref
140 target="RFC1242"/>) and foundation of the benchmarking work in OPNFV is
141 common and strong.</t>
144 https://wiki.opnfv.org/characterize_vswitch_performance_for_telco_nfv_use_cases
145 for more background, and the OPNFV website for general information:
146 https://www.opnfv.org/</t>
148 <t>The authors note that OPNFV distinguishes itself from other open
149 source compute and networking projects through its emphasis on existing
150 "telco" services as opposed to cloud-computing. There are many ways in
151 which telco requirements have different emphasis on performance
152 dimensions when compared to cloud computing: support for and transfer of
153 isochronous media streams is one example.</t>
155 <t>Note also that the move to NFV Infrastructure has resulted in many
156 new benchmarking initiatives across the industry. The authors are
157 currently doing their best to maintain alignment with many other
158 projects, and this Internet Draft is one part of the efforts. We
159 acknowledge the early work in <xref
160 target="I-D.huang-bmwg-virtual-network-performance"/>, and useful
161 discussion with the authors.</t>
164 <section title="Scope">
165 <t>The primary purpose and scope of the memo is to inform the industry
166 of work-in-progress that builds on the body of extensive BMWG literature
167 and experience, and describe the extensions needed for benchmarking
168 virtual switches. Inital feedback indicates that many of these
169 extensions may be applicable beyond the current scope (to hardware
170 switches in the NFV Infrastructure and to virtual routers, for example).
171 Additionally, this memo serves as a vehicle to include more detail and
172 commentary from BMWG and other Open Source communities, under BMWG's
173 chartered work to characterize the NFV Infrastructure (a virtual switch
174 is an important aspect of that infrastructure).</t>
177 <section title="Benchmarking Considerations">
178 <t>This section highlights some specific considerations (from <xref
179 target="I-D.ietf-bmwg-virtual-net"/>)related to Benchmarks for virtual
180 switches. The OPNFV project is sharing its present view on these areas,
181 as they develop their specifications in the Level Test Design (LTD)
184 <section title="Comparison with Physical Network Functions">
185 <t>To compare the performance of virtual designs and implementations
186 with their physical counterparts, identical benchmarks are needed.
187 BMWG has developed specifications for many network functions this memo
188 re-uses existing benchmarks through references, and expands them
189 during development of new methods. A key configuration aspect is the
190 number of parallel cores required to achieve comparable performance
191 with a given physical device, or whether some limit of scale was
192 reached before the cores could achieve the comparable level.</t>
194 <t>It's unlikely that the virtual switch will be the only application
195 running on the SUT, so CPU utilization, Cache utilization, and Memory
196 footprint should also be recorded for the virtual implementations of
197 internetworking functions.</t>
200 <section title="Continued Emphasis on Black-Box Benchmarks">
201 <t>External observations remain essential as the basis for Benchmarks.
202 Internal observations with fixed specification and interpretation will
203 be provided in parallel to assist the development of operations
204 procedures when the technology is deployed.</t>
207 <section title="New Configuration Parameters">
208 <t>A key consideration when conducting any sort of benchmark is trying
209 to ensure the consistency and repeatability of test results. When
210 benchmarking the performance of a vSwitch there are many factors that
211 can affect the consistency of results, one key factor is matching the
212 various hardware and software details of the SUT. This section lists
213 some of the many new parameters which this project believes are
214 critical to report in order to achieve repeatability.</t>
216 <t>Hardware details including:</t>
218 <t><list style="symbols">
219 <t>Platform details</t>
221 <t>Processor details</t>
223 <t>Memory information (type and size)</t>
225 <t>Number of enabled cores</t>
227 <t>Number of cores used for the test</t>
229 <t>Number of physical NICs, as well as their details
230 (manufacturer, versions, type and the PCI slot they are plugged
233 <t>NIC interrupt configuration</t>
235 <t>BIOS version, release date and any configurations that were
238 <t>CPU microcode level</t>
240 <t>Memory DIMM configurations (quad rank performance may not be
241 the same as dual rank) in size, freq and slot locations</t>
243 <t>PCI configuration parameters (payload size, early ack
246 <t>Power management at all levels (ACPI sleep states, processor
248 </list>Software details including:</t>
250 <t><list style="symbols">
251 <t>OS parameters and behavior (text vs graphical no one typing at
252 the console on one system)</t>
254 <t>OS version (for host and VNF)</t>
256 <t>Kernel version (for host and VNF)</t>
258 <t>GRUB boot parameters (for host and VNF)</t>
260 <t>Hypervisor details (Type and version)</t>
262 <t>Selected vSwitch, version number or commit id used</t>
264 <t>vSwitch launch command line if it has been parameterised</t>
266 <t>Memory allocation to the vSwitch</t>
268 <t>which NUMA node it is using, and how many memory channels</t>
270 <t>DPDK or any other SW dependency version number or commit id
273 <t>Memory allocation to a VM - if it's from Hugpages/elsewhere</t>
275 <t>VM storage type: snapshot/independent persistent/independent
280 <t>Number of Virtual NICs (vNICs), versions, type and driver</t>
282 <t>Number of virtual CPUs and their core affinity on the host</t>
284 <t>Number vNIC interrupt configuration</t>
286 <t>Thread affinitization for the applications (including the
287 vSwitch itself) on the host</t>
289 <t>Details of Resource isolation, such as CPUs designated for
290 Host/Kernel (isolcpu) and CPUs designated for specific processes
291 (taskset). - Test duration. - Number of flows.</t>
294 <t>Test Traffic Information:<list style="symbols">
295 <t>Traffic type - UDP, TCP, IMIX / Other</t>
299 <t>Deployment Scenario</t>
305 <section title="Flow classification">
306 <t>Virtual switches group packets into flows by processing and
307 matching particular packet or frame header information, or by matching
308 packets based on the input ports. Thus a flow can be thought of a
309 sequence of packets that have the same set of header field values or
310 have arrived on the same port. Performance results can vary based on
311 the parameters the vSwitch uses to match for a flow. The recommended
312 flow classification parameters for any vSwitch performance tests are:
313 the input port, the source IP address, the destination IP address and
314 the Ethernet protocol type field. It is essential to increase the flow
315 timeout time on a vSwitch before conducting any performance tests that
316 do not measure the flow setup time. Normally the first packet of a
317 particular stream will install the flow in the virtual switch which
318 adds an additional latency, subsequent packets of the same flow are
319 not subject to this latency if the flow is already installed on the
323 <section title="Benchmarks using Baselines with Resource Isolation">
324 <t>This outline describes measurement of baseline with isolated
325 resources at a high level, which is the intended approach at this
328 <t><list style="numbers">
329 <t>Baselines: <list style="symbols">
330 <t>Optional: Benchmark platform forwarding capability without
331 a vswitch or VNF for at least 72 hours (serves as a means of
332 platform validation and a means to obtain the base performance
333 for the platform in terms of its maximum forwarding rate and
335 <preamble>Benchmark platform forwarding
336 capability</preamble>
338 <artwork align="right"><![CDATA[ __
339 +--------------------------------------------------+ |
340 | +------------------------------------------+ | |
342 | | Simple Forwarding App | | Host
344 | +------------------------------------------+ | |
346 +---+------------------------------------------+---+ __|
350 +--------------------------------------------------+
352 | traffic generator |
354 +--------------------------------------------------+]]></artwork>
359 <t>Benchmark VNF forwarding capability with direct
360 connectivity (vSwitch bypass, e.g., SR/IOV) for at least 72
361 hours (serves as a means of VNF validation and a means to
362 obtain the base performance for the VNF in terms of its
363 maximum forwarding rate and latency). The metrics gathered
364 from this test will serve as a key comparison point for
365 vSwitch bypass technologies performance and vSwitch
366 performance. <figure align="right">
367 <preamble>Benchmark VNF forwarding capability</preamble>
369 <artwork><![CDATA[ __
370 +--------------------------------------------------+ |
371 | +------------------------------------------+ | |
375 | +------------------------------------------+ | |
376 | | Passthrough/SR-IOV | | Host
377 | +------------------------------------------+ | |
379 +---+------------------------------------------+---+ __|
383 +--------------------------------------------------+
385 | traffic generator |
387 +--------------------------------------------------+]]></artwork>
392 <t>Benchmarking with isolated resources alone, with other
393 resources (both HW&SW) disabled Example, vSw and VM are
396 <t>Benchmarking with isolated resources alone, leaving some
399 <t>Benchmark with isolated resources and all resources
403 <t>Next Steps<list style="symbols">
404 <t>Limited sharing</t>
406 <t>Production scenarios</t>
408 <t>Stressful scenarios</t>
414 <section title="VSWITCHPERF Specification Summary">
415 <t>The overall specification in preparation is referred to as a Level
416 Test Design (LTD) document, which will contain a suite of performance
417 tests. The base performance tests in the LTD are based on the
418 pre-existing specifications developed by BMWG to test the performance of
419 physical switches. These specifications include:</t>
421 <t><list style="symbols">
422 <t><xref target="RFC2544"/> Benchmarking Methodology for Network
423 Interconnect Devices</t>
425 <t><xref target="RFC2889"/> Benchmarking Methodology for LAN
428 <t><xref target="RFC6201"/> Device Reset Characterization</t>
430 <t><xref target="RFC5481"/> Packet Delay Variation Applicability
434 <t>Some of the above/newer RFCs are being applied in benchmarking for
435 the first time, and represent a development challenge for test equipment
436 developers. Fortunately, many members of the testing system community
437 have engaged on the VSPERF project, including an open source test
440 <t>In addition to this, the LTD also re-uses the terminology defined
443 <t><list style="symbols">
444 <t><xref target="RFC2285"/> Benchmarking Terminology for LAN
445 Switching Devices</t>
447 <t><xref target="RFC5481"/> Packet Delay Variation Applicability
453 <t>Specifications to be included in future updates of the LTD
454 include:<list style="symbols">
455 <t><xref target="RFC3918"/> Methodology for IP Multicast
458 <t><xref target="RFC4737"/> Packet Reordering Metrics</t>
461 <t>As one might expect, the most fundamental internetworking
462 characteristics of Throughput and Latency remain important when the
463 switch is virtualized, and these benchmarks figure prominently in the
466 <t>When considering characteristics important to "telco" network
467 functions, we must begin to consider additional performance metrics. In
468 this case, the project specifications have referenced metrics from the
469 IETF IP Performance Metrics (IPPM) literature. This means that the <xref
470 target="RFC2544"/> test of Latency is replaced by measurement of a
471 metric derived from IPPM's <xref target="RFC2679"/>, where a set of
472 statistical summaries will be provided (mean, max, min, etc.). Further
473 metrics planned to be benchmarked include packet delay variation as
474 defined by <xref target="RFC5481"/> , reordering, burst behaviour, DUT
475 availability, DUT capacity and packet loss in long term testing at
476 Throughput level, where some low-level of background loss may be present
477 and characterized.</t>
479 <t>Tests have been (or will be) designed to collect the metrics
482 <t><list style="symbols">
483 <t>Throughput Tests to measure the maximum forwarding rate (in
484 frames per second or fps) and bit rate (in Mbps) for a constant load
485 (as defined by <xref target="RFC1242"/>) without traffic loss.</t>
487 <t>Packet and Frame Delay Distribution Tests to measure average, min
488 and max packet and frame delay for constant loads.</t>
490 <t>Packet Delay Tests to understand latency distribution for
491 different packet sizes and over an extended test run to uncover
494 <t>Scalability Tests to understand how the virtual switch performs
495 as the number of flows, active ports, complexity of the forwarding
496 logic’s configuration… it has to deal with
499 <t>Stream Performance Tests (TCP, UDP) to measure bulk data transfer
500 performance, i.e. how fast systems can send and receive data through
503 <t>Control Path and Datapath Coupling Tests, to understand how
504 closely coupled the datapath and the control path are as well as the
505 effect of this coupling on the performance of the DUT (example:
506 delay of the initial packet of a flow).</t>
508 <t>CPU and Memory Consumption Tests to understand the virtual
509 switch’s footprint on the system, usually conducted as
510 auxiliary measurements with benchmarks above. They include: CPU
511 utilization, Cache utilization and Memory footprint.</t>
513 <t>The so-called "Soak" tests, where the selected test is conducted
514 over a long period of time (with an ideal duration of 24 hours, and
515 at least 6 hours). The purpose of soak tests is to capture transient
516 changes in performance which may occur due to infrequent processes
517 or the low probability coincidence of two or more processes. The
518 performance must be evaluated periodically during continuous
519 testing, and this results in use of <xref target="RFC2889"/> Frame
520 Rate metrics instead of <xref target="RFC2544"/> Throughput (which
521 requires stopping traffic to allow time for all traffic to exit
522 internal queues).</t>
525 <t>Future/planned test specs include:<list style="symbols">
526 <t>Request/Response Performance Tests (TCP, UDP) which measure the
527 transaction rate through the switch.</t>
529 <t>Noisy Neighbour Tests, to understand the effects of resource
530 sharing on the performance of a virtual switch.</t>
532 <t>Tests derived from examination of ETSI NFV Draft GS IFA003
533 requirements <xref target="IFA003"/> on characterization of
534 acceleration technologies applied to vswitches.</t>
535 </list>The flexibility of deployment of a virtual switch within a
536 network means that the BMWG IETF existing literature needs to be used to
537 characterize the performance of a switch in various deployment
538 scenarios. The deployment scenarios under consideration include:</t>
541 <preamble>Physical port to virtual switch to physical
544 <artwork><![CDATA[ __
545 +--------------------------------------------------+ |
546 | +--------------------+ | |
549 | +--------------+ +--------------+ | |
550 | | phy port | vSwitch | phy port | | |
551 +---+--------------+------------+--------------+---+ __|
555 +--------------------------------------------------+
557 | traffic generator |
559 +--------------------------------------------------+]]></artwork>
563 <preamble>Physical port to virtual switch to VNF to virtual switch
564 to physical port</preamble>
566 <artwork><![CDATA[ __
567 +---------------------------------------------------+ |
569 | +-------------------------------------------+ | |
570 | | Application | | |
571 | +-------------------------------------------+ | |
575 | +---------------+ +---------------+ | |
576 | | logical port 0| | logical port 1| | |
577 +---+---------------+-----------+---------------+---+ __|
581 +---+---------------+----------+---------------+---+ |
582 | | logical port 0| | logical port 1| | |
583 | +---------------+ +---------------+ | |
587 | +--------------+ +--------------+ | |
588 | | phy port | vSwitch | phy port | | |
589 +---+--------------+------------+--------------+---+ __|
593 +--------------------------------------------------+
595 | traffic generator |
597 +--------------------------------------------------+]]></artwork>
599 <preamble>Physical port to virtual switch to VNF to virtual switch
600 to VNF to virtual switch to physical port</preamble>
602 <artwork><![CDATA[ __
603 +----------------------+ +----------------------+ |
604 | Guest 1 | | Guest 2 | |
605 | +---------------+ | | +---------------+ | |
606 | | Application | | | | Application | | |
607 | +---------------+ | | +---------------+ | |
609 | | v | | | v | | Guests
610 | +---------------+ | | +---------------+ | |
611 | | logical ports | | | | logical ports | | |
612 | | 0 1 | | | | 0 1 | | |
613 +---+---------------+--+ +---+---------------+--+__|
617 +---+---------------+---------+---------------+--+ |
618 | | 0 1 | | 3 4 | | |
619 | | logical ports | | logical ports | | |
620 | +---------------+ +---------------+ | |
622 | | |-----------------| v | |
623 | +--------------+ +--------------+ | |
624 | | phy ports | vSwitch | phy ports | | |
625 +---+--------------+----------+--------------+---+_|
629 +--------------------------------------------------+
631 | traffic generator |
633 +--------------------------------------------------+]]></artwork>
635 <preamble>Physical port to virtual switch to VNF</preamble>
637 <artwork><![CDATA[ __
638 +---------------------------------------------------+ |
640 | +-------------------------------------------+ | |
641 | | Application | | |
642 | +-------------------------------------------+ | |
646 | +---------------+ | |
647 | | logical port 0| | |
648 +---+---------------+-------------------------------+ __|
652 +---+---------------+------------------------------+ |
653 | | logical port 0| | |
654 | +---------------+ | |
658 | +--------------+ | |
659 | | phy port | vSwitch | |
660 +---+--------------+------------ -------------- ---+ __|
664 +--------------------------------------------------+
666 | traffic generator |
668 +--------------------------------------------------+]]></artwork>
670 <preamble>VNF to virtual switch to physical port</preamble>
672 <artwork><![CDATA[ __
673 +---------------------------------------------------+ |
675 | +-------------------------------------------+ | |
676 | | Application | | |
677 | +-------------------------------------------+ | |
681 | +---------------+ | |
682 | | logical port | | |
683 +-------------------------------+---------------+---+ __|
687 +------------------------------+---------------+---+ |
688 | | logical port | | |
689 | +---------------+ | |
693 | +--------------+ | |
694 | vSwitch | phy port | | |
695 +-------------------------------+--------------+---+ __|
699 +--------------------------------------------------+
701 | traffic generator |
703 +--------------------------------------------------+]]></artwork>
705 <preamble>VNF to virtual switch to VNF</preamble>
707 <artwork><![CDATA[ __
708 +----------------------+ +----------------------+ |
709 | Guest 1 | | Guest 2 | |
710 | +---------------+ | | +---------------+ | |
711 | | Application | | | | Application | | |
712 | +---------------+ | | +---------------+ | |
715 | +---------------+ | | +---------------+ | |
716 | | logical ports | | | | logical ports | | |
717 | | 0 | | | | 0 | | |
718 +---+---------------+--+ +---+---------------+--+__|
722 +---+---------------+---------+---------------+--+ |
724 | | logical ports | | logical ports | | |
725 | +---------------+ +---------------+ | |
727 | L-----------------+ | |
730 +------------------------------------------------+_|]]></artwork>
733 <t>A set of Deployment Scenario figures is available on the VSPERF Test
734 Methodology Wiki page <xref target="TestTopo"/>.</t>
737 <section title="3x3 Matrix Coverage">
738 <t>This section organizes the many existing test specifications into the
739 "3x3" matrix (introduced in <xref target="I-D.ietf-bmwg-virtual-net"/>).
740 Because the LTD specification ID names are quite long, this section is
741 organized into lists for each occupied cell of the matrix (not all are
742 occupied, also the matrix has grown to 3x4 to accommodate scale metrics
743 when displaying the coverage of many metrics/benchmarks). The current
744 version of the LTD specification is available <xref target="LTD"/>.</t>
746 <t>The tests listed below assess the activation of paths in the data
747 plane, rather than the control plane.</t>
749 <t>A complete list of tests with short summaries is available on the
750 VSPERF "LTD Test Spec Overview" Wiki page <xref target="LTDoverV"/>.</t>
752 <section title="Speed of Activation">
753 <t><list style="symbols">
754 <t>Activation.RFC2889.AddressLearningRate</t>
756 <t>PacketLatency.InitialPacketProcessingLatency</t>
760 <section title="Accuracy of Activation section">
761 <t><list style="symbols">
762 <t>CPDP.Coupling.Flow.Addition</t>
766 <section title="Reliability of Activation">
767 <t><list style="symbols">
768 <t>Throughput.RFC2544.SystemRecoveryTime</t>
770 <t>Throughput.RFC2544.ResetTime</t>
774 <section title="Scale of Activation">
775 <t><list style="symbols">
776 <t>Activation.RFC2889.AddressCachingCapacity</t>
780 <section title="Speed of Operation">
781 <t><list style="symbols">
782 <t>Throughput.RFC2544.PacketLossRate</t>
784 <t>CPU.RFC2544.0PacketLoss</t>
786 <t>Throughput.RFC2544.PacketLossRateFrameModification</t>
788 <t>Throughput.RFC2544.BackToBackFrames</t>
790 <t>Throughput.RFC2889.MaxForwardingRate</t>
792 <t>Throughput.RFC2889.ForwardPressure</t>
794 <t>Throughput.RFC2889.BroadcastFrameForwarding</t>
798 <section title="Accuracy of Operation">
799 <t><list style="symbols">
800 <t>Throughput.RFC2889.ErrorFramesFiltering</t>
802 <t>Throughput.RFC2544.Profile</t>
806 <section title="Reliability of Operation">
807 <t><list style="symbols">
808 <t>Throughput.RFC2889.Soak</t>
810 <t>Throughput.RFC2889.SoakFrameModification</t>
812 <t>PacketDelayVariation.RFC3393.Soak</t>
816 <section title="Scalability of Operation">
817 <t><list style="symbols">
818 <t>Scalability.RFC2544.0PacketLoss</t>
820 <t>MemoryBandwidth.RFC2544.0PacketLoss.Scalability</t>
824 <section title="Summary">
826 <artwork><![CDATA[|------------------------------------------------------------------------|
828 | | SPEED | ACCURACY | RELIABILITY | SCALE |
830 |------------------------------------------------------------------------|
832 | Activation | X | X | X | X |
834 |------------------------------------------------------------------------|
836 | Operation | X | X | X | X |
838 |------------------------------------------------------------------------|
840 | De-activation | | | | |
842 |------------------------------------------------------------------------|]]></artwork>
847 <section title="Security Considerations">
848 <t>Benchmarking activities as described in this memo are limited to
849 technology characterization of a Device Under Test/System Under Test
850 (DUT/SUT) using controlled stimuli in a laboratory environment, with
851 dedicated address space and the constraints specified in the sections
854 <t>The benchmarking network topology will be an independent test setup
855 and MUST NOT be connected to devices that may forward the test traffic
856 into a production network, or misroute traffic to the test management
859 <t>Further, benchmarking is performed on a "black-box" basis, relying
860 solely on measurements observable external to the DUT/SUT.</t>
862 <t>Special capabilities SHOULD NOT exist in the DUT/SUT specifically for
863 benchmarking purposes. Any implications for network security arising
864 from the DUT/SUT SHOULD be identical in the lab and in production
868 <section anchor="IANA" title="IANA Considerations">
869 <t>No IANA Action is requested at this time.</t>
872 <section title="Acknowledgements">
873 <t>The authors appreciate and acknowledge comments from Scott Bradner,
874 Marius Georgescu, Ramki Krishnan, Doug Montgomery, Martin Klozik,
875 Christian Trautman, and others for their reviews.</t>
880 <references title="Normative References">
883 <?rfc include="reference.RFC.2119"?>
887 <?rfc include="reference.RFC.2330"?>
889 <?rfc include='reference.RFC.2544'?>
891 <?rfc include="reference.RFC.2679"?>
893 <?rfc include='reference.RFC.2680'?>
895 <?rfc include='reference.RFC.3393'?>
897 <?rfc include='reference.RFC.3432'?>
899 <?rfc include='reference.RFC.2681'?>
901 <?rfc include='reference.RFC.5905'?>
903 <?rfc include='reference.RFC.4689'?>
905 <?rfc include='reference.RFC.4737'?>
907 <?rfc include='reference.RFC.5357'?>
909 <?rfc include='reference.RFC.2889'?>
911 <?rfc include='reference.RFC.3918'?>
913 <?rfc include='reference.RFC.6201'?>
915 <?rfc include='reference.RFC.2285'?>
917 <reference anchor="NFV.PER001">
919 <title>Network Function Virtualization: Performance and Portability
920 Best Practices</title>
922 <author fullname="ETSI NFV" initials="" surname="">
926 <date month="June" year="2014"/>
929 <seriesInfo name="Group Specification"
930 value="ETSI GS NFV-PER 001 V1.1.1 (2014-06)"/>
936 <references title="Informative References">
937 <?rfc include='reference.RFC.1242'?>
939 <?rfc include='reference.RFC.5481'?>
941 <?rfc include='reference.RFC.6049'?>
943 <?rfc include='reference.RFC.6248'?>
945 <?rfc include='reference.RFC.6390'?>
947 <?rfc include='reference.I-D.ietf-bmwg-virtual-net'?>
949 <?rfc include='reference.I-D.huang-bmwg-virtual-network-performance'?>
951 <reference anchor="TestTopo">
953 <title>Test Topologies
954 https://wiki.opnfv.org/vsperf/test_methodology</title>
964 <reference anchor="LTDoverV">
966 <title>LTD Test Spec Overview
967 https://wiki.opnfv.org/wiki/vswitchperf_test_spec_review</title>
977 <reference anchor="LTD">
979 <title>LTD Test Specification
980 http://artifacts.opnfv.org/vswitchperf/docs/requirements/index.html</title>
990 <reference anchor="BrahRel">
992 <title>Brahmaputra, Second OPNFV Release
993 https://www.opnfv.org/brahmaputra</title>
1003 <reference anchor="IFA003">
1005 <title>https://docbox.etsi.org/ISG/NFV/Open/Drafts/IFA003_Acceleration_-_vSwitch_Spec/</title>