IMIX Genome: Specification of Variable Packet Sizes for Additional Testing
A short written name that spells out the exact sequence of packet sizes used in a load test.
Morton, A., "IMIX Genome: Specification of Variable Packet Sizes for Additional Testing", RFC 6985, DOI 10.17487/RFC6985, July 2013, <https://www.rfc-editor.org/info/rfc6985>.
@misc{rfc6985,
series = {Request for Comments},
number = 6985,
howpublished = {RFC 6985},
publisher = {RFC Editor},
doi = {10.17487/RFC6985},
url = {https://www.rfc-editor.org/info/rfc6985},
author = {A. Morton},
title = {{IMIX Genome: Specification of Variable Packet Sizes for Additional Testing}},
pagetotal = 10,
year = 2013,
month = jul,
} RFC text
Contents
Internet Engineering Task Force (IETF) A. Morton
Request for Comments: 6985 AT&T Labs
Category: Informational July 2013
ISSN: 2070-1721
IMIX Genome: Specification of Variable Packet Sizes
for Additional Testing
Abstract
Benchmarking methodologies have always relied on test conditions with
constant packet sizes, with the goal of understanding what network
device capability has been tested. Tests with a constant packet size
reveal device capabilities but differ significantly from the
conditions encountered in operational deployment, so additional tests
are sometimes conducted with a mixture of packet sizes, or "IMIX"
("Internet Mix"). The mixture of sizes a networking device will
encounter is highly variable and depends on many factors. An IMIX
suited for one networking device and deployment will not be
appropriate for another. However, the mix of sizes may be known, and
the tester may be asked to augment the fixed-size tests. To address
this need and the perpetual goal of specifying repeatable test
conditions, this document defines a way to specify the exact
repeating sequence of packet sizes from the usual set of fixed sizes
and from other forms of mixed-size specification.
Status of This Memo
This document is not an Internet Standards Track specification; it is
published for informational purposes.
This document is a product of the Internet Engineering Task Force
(IETF). It represents the consensus of the IETF community. It has
received public review and has been approved for publication by the
Internet Engineering Steering Group (IESG). Not all documents
approved by the IESG are a candidate for any level of Internet
Standard; see Section 2 of RFC 5741.
Information about the current status of this document, any errata,
and how to provide feedback on it may be obtained at
http://www.rfc-editor.org/info/rfc6985.
Copyright Notice
Copyright (c) 2013 IETF Trust and the persons identified as the
document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents
(http://trustee.ietf.org/license-info) in effect on the date of
publication of this document. Please review these documents
carefully, as they describe your rights and restrictions with respect
to this document. Code Components extracted from this document must
include Simplified BSD License text as described in Section 4.e of
the Trust Legal Provisions and are provided without warranty as
described in the Simplified BSD License.
Table of Contents
1. Introduction ....................................................2
2. Requirements Language ...........................................3
3. Scope and Goals .................................................3
4. Specification of the IMIX Genome ................................4
5. Specification of a Custom IMIX ..................................6
6. Reporting Long or Pseudorandom Packet Sequences .................7
6.1. Run-Length Encoding ........................................7
6.2. Table of Proportions .......................................7
6.3. Deterministic Algorithm ....................................7
6.4. Pseudorandom Length Algorithm ..............................8
6.5. Pseudorandom Sequence Algorithm ............................8
7. Security Considerations .........................................8
8. Acknowledgements ................................................8
9. References ......................................................9
9.1. Normative References .......................................9
9.2. Informative References .....................................9
1. Introduction
This memo defines a method to unambiguously specify the sequence of
packet sizes used in a load test.
Benchmarking methodologies [RFC2544] have always relied on test
conditions with constant packet sizes, with the goal of understanding
what network device capability has been tested. Tests with the
smallest size stress the header processing capacity, and tests with
the largest size stress the overall bit-processing capacity. Tests
with sizes in between may determine the transition between these two
capacities.
Streams of constant packet size differ significantly from the
conditions encountered in operational deployment, so additional tests
are sometimes conducted with a mixture of packet sizes. The set of
sizes used is often called an Internet Mix, or "IMIX" [Spirent]
[IXIA] [Agilent].
The mixture of sizes a networking device will encounter is highly
variable and depends on many factors. An IMIX suited for one
networking device and deployment will not be appropriate for another.
However, the mix of sizes may be known, and the tester may be asked
to augment the fixed-size tests. The references above cite the
original studies and their methodologies. Similar methods can be
used to determine new size mixes present on a link or network. We
note that the architecture for IP Flow Information Export [RFC5470]
provides one method to gather packet-size information on private
networks.
To address this need and the perpetual goal of specifying repeatable
test conditions, this memo proposes a way to specify the exact
repeating sequence of packet sizes from the usual set of fixed sizes:
the IMIX Genome. Other, less exact forms of size specification are
also recommended for extremely complicated or customized size mixes.
We apply the term "genome" to infer that the entire test packet-size
sequence can be replicated if this information is known -- a parallel
to the information needed for biological replication.
This memo takes the position that it cannot be proven for all
circumstances that the sequence of packet sizes does not affect the
test result; thus, a standardized specification of sequence is
valuable.
2. Requirements Language
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
document are to be interpreted as described in RFC 2119 [RFC2119].
3. Scope and Goals
This memo defines a method to unambiguously specify the sequence of
packet sizes that have been used in a load test, assuming that a
relevant mix of sizes is known to the tester and the length of the
repeating sequence is not very long (<100 packets).
The IMIX Genome will allow an exact sequence of packet sizes to be
communicated as a single-line name, resolving the current ambiguity
with results that simply refer to "IMIX". This aspect is critical
because no ability has been demonstrated to extrapolate results from
one IMIX to another IMIX -- and certainly no ability to extrapolate
results to other circumstances -- even when the mix varies only
slightly from another IMIX.
While documentation of the exact sequence is ideal, the memo also
covers the case where the sequence of sizes is very long or may be
generated by a pseudorandom process.
It is a colossal non-goal to standardize one or more versions of the
IMIX. This topic has been discussed on many occasions on the BMWG
mailing list [IMIXonList]. The goal is to enable customization with
minimal constraints while fostering repeatable testing once the
fixed-size testing is complete. Thus, the requirements presented in
this specification, expressed in [RFC2119] terms, are intended for
those performing/reporting laboratory tests to improve clarity and
repeatability.
4. Specification of the IMIX Genome
The IMIX Genome is specified in the following format:
IMIX - 123456...x
where each number is replaced by the letter corresponding to the size
of the packet at that position in the sequence. The following table
gives the letter encoding for the [RFC2544] standard sizes (64, 128,
256, 512, 1024, 1280, and 1518 bytes) and "jumbo" sizes (2112, 9000,
and 16000 bytes). Note that the 4-octet Ethernet frame check
sequence may fail to detect bit errors in the larger jumbo frames
[Jumbo1] [Jumbo2].
+--------------+--------------------+
| Size (Bytes) | Genome Code Letter |
+--------------+--------------------+
| 64 | a |
| 128 | b |
| 256 | c |
| 512 | d |
| 1024 | e |
| 1280 | f |
| 1518 | g |
| 2112 | h |
| 9000 | i |
| 16000 | j |
| MTU | z |
+--------------+--------------------+
For example, a five-packet sequence with sizes 64,64,64,1280,1518
would be designated:
IMIX - aaafg
If z (MTU) is used, the tester MUST specify the length of the MTU in
the report.
While this approach allows some flexibility, there are also
constraints.
o Packet sizes not defined by RFC 2544 would need to be approximated
by those available in the table.
o The genome for very long sequences can become undecipherable by
humans.
Some questions testers must ask and answer when using the IMIX Genome
are:
1. Multiple source-destination address pairs: Is the IMIX sequence
applicable to each pair, across multiple pairs in sets, or across
all pairs?
2. Multiple tester ports: Is the IMIX sequence applicable to each
port, across multiple ports in sets, or across all ports?
The chosen configuration would be expressed in the following general
form:
+-------------------+------------------------+---------------------+
| Source Address + | Destination Address + | Corresponding IMIX |
| Port AND/OR Blade | Port AND/OR Blade | |
+-------------------+------------------------+---------------------+
| x.x.x.x Blade2 | y.y.y.y Blade3 | IMIX - aaafg |
+-------------------+------------------------+---------------------+
where testers can specify the IMIX used between any two entities in
the test architecture (and "Blade" is a component in a multi-
component device chassis).
5. Specification of a Custom IMIX
This section describes how to specify an IMIX with locally selected
packet sizes.
The custom IMIX is specified in the following format:
CUSTOM IMIX - 123456...x
where each number is replaced by the letter corresponding to the size
of the packet at that position in the sequence. The tester MUST
complete the following table, giving the letter encoding for each
size used, where each set of three lower-case letters would be
replaced by the integer size in octets.
+--------------+--------------------+
| Size (Bytes) | Custom Code Letter |
+--------------+--------------------+
| aaa | A |
| bbb | B |
| ccc | C |
| ddd | D |
| eee | E |
| fff | F |
| ggg | G |
| etc. | up to Z |
+--------------+--------------------+
For example, a five-packet sequence with sizes
aaa=64,aaa=64,aaa=64,ggg=1020,ggg=1020 would be designated:
CUSTOM IMIX - AAAGG
6. Reporting Long or Pseudorandom Packet Sequences
When the IMIX Genome cannot be used (when the sheer length of the
sequence would make the genome unmanageable), five options are
possible, as noted in the following subsections.
6.1. Run-Length Encoding
When a sequence can be decomposed into a series of short repeating
sequences, then a run-length encoding approach MAY be specified as
shown in the table below (using the single lower-case letter Genome
Codes from Section 4):
+------------------------------+----------------------+
| Count of Repeating Sequences | Packet-Size Sequence |
+------------------------------+----------------------+
| 20 | abcd |
| 5 | ggga |
| 10 | dcba |
+------------------------------+----------------------+
The run-length encoding approach is also applicable to the custom
IMIX as described in Section 5 (where the single upper-case letter
Genome Codes would be used instead).
6.2. Table of Proportions
When the sequence is designed to vary within some proportional
constraints, a table simply giving the proportions of each size MAY
be used instead.
+-----------+---------------------+---------------------------+
| IP Length | Percentage of Total | Length(s) at Other Layers |
+-----------+---------------------+---------------------------+
| 64 | 23 | 82 |
| 128 | 67 | 146 |
| 1000 | 10 | 1018 |
+-----------+---------------------+---------------------------+
Note that the table of proportions also allows non-standard packet
sizes but trades the short genome specification and ability to
specify the exact sequence for other flexibilities.
6.3. Deterministic Algorithm
If a deterministic packet-size generation method is used (such as a
monotonic increase by 1 octet from start value to MTU), then the
generation algorithm SHOULD be reported.
6.4. Pseudorandom Length Algorithm
If a pseudorandom length generation capability is used, then the
generation algorithm SHOULD be reported with the results along with
the seed value used. We also recognize the opportunity to randomize
inter-packet spacing from a test sender as well as the size, and both
spacing and length pseudorandom generation algorithms and seeds
SHOULD be reported when used.
6.5. Pseudorandom Sequence Algorithm
Finally, we note another possibility: a pseudorandom sequence
generates an index to the table of packet lengths, and the generation
algorithm SHOULD be reported with the results, along with the seed
value if used.
7. Security Considerations
Benchmarking activities as described in this memo are limited to
technology characterization using controlled stimuli in a laboratory
environment, with dedicated address space and other constraints
[RFC2544].
The benchmarking network topology will be an independent test setup
and MUST NOT be connected to devices that may forward the test
traffic into a production network or misroute traffic to the test
management network.
Further, benchmarking is performed on a "black-box" basis, relying
solely on measurements observable external to the Device Under Test
(DUT) or System Under Test (SUT).
Special capabilities SHOULD NOT exist in the DUT/SUT specifically for
benchmarking purposes. Any implications for network security arising
from the DUT/SUT SHOULD be identical in the lab and in production
networks.
8. Acknowledgements
Thanks to Sarah Banks, Aamer Akhter, Steve Maxwell, and Scott Bradner
for their reviews and comments. Ilya Varlashkin suggested the
run-length encoding approach in Section 6.1.
9. References
9.1. Normative References
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119, March 1997.
[RFC2544] Bradner, S. and J. McQuaid, "Benchmarking Methodology for
Network Interconnect Devices", RFC 2544, March 1999.
9.2. Informative References
[Agilent] Agilent, "The Journal of Internet Test Methodologies",
September 2007, <http://www.ixiacom.com/pdfs/test_plans/
agilent_journal_of_internet_test_methodologies.pdf>.
[IMIXonList]
IETF Benchmarking Methodology Working Group, "Discussion
on IMIX", October 2003, <http://www.ietf.org/mail-archive/
web/bmwg/current/msg00691.html>.
[IXIA] IXIA, "Testing PPPoX and L2TP Broadband Access Devices",
2004, <http://www.ixiacom.com/library/test_plans/
display?skey=testing_pppox>.
[Jumbo1] Dykstra, P., "Gigabit Ethernet Jumbo Frames, and why you
should care", WareOnEarth Communications, Inc., December
1999, <http://sd.wareonearth.com/~phil/jumbo.html>.
[Jumbo2] Mathis, M., "The Ethernet CRC limits packets to about
12 kBytes. (NOT)", Pittsburgh Supercomputing Center,
April 2003,
<http://staff.psc.edu/mathis/MTU/arguments.html#crc>.
[RFC5470] Sadasivan, G., Brownlee, N., Claise, B., and J. Quittek,
"Architecture for IP Flow Information Export", RFC 5470,
March 2009.
[Spirent] Spirent, "Test Methodology Journal: IMIX (Internet Mix)
Journal", January 2006, <http://gospirent.com/whitepaper/
IMIX%20Test%20Methodolgy%20Journal.pdf>.
Author's Address
Al Morton
AT&T Labs
200 Laurel Avenue South
Middletown, NJ 07748
USA
Phone: +1 732 420 1571
Fax: +1 732 368 1192
EMail: acmorton@att.com
URI: http://home.comcast.net/~acmacm/
ISO Plain Language
On this page
§1Why Fixed Sizes Are Not Enough
Benchmarking methodologies have always relied on test conditions with constant packet sizes, with the goal of understanding what network device capability has been tested. Tests with the smallest size stress the header processing capacity, and tests with the largest size stress the overall bit-processing capacity. Sizes in between may determine the transition between those two capacities.
Streams of constant packet size differ significantly from the conditions met in operational deployment, so additional tests are sometimes conducted with a mixture of packet sizes. The set of sizes used is often called an Internet Mix, or IMIX. The mixture a device will encounter is highly variable and depends on many factors, so an IMIX suited to one device and deployment will not be appropriate for another. The mix may nevertheless be known, and the tester may be asked to augment the fixed-size tests. The RFC also notes that the architecture for IP Flow Information Export in RFC 5470 provides one way to gather packet-size information on private networks.
The proposal is a way to specify the exact repeating sequence of packet sizes from the usual set of fixed sizes, called the IMIX Genome. The name is meant to convey that the entire test packet-size sequence can be replicated if the information is known, much as the information needed for biological replication does. The memo takes the position that it cannot be proven for all circumstances that the sequence of packet sizes does not affect the test result, so a standardised specification of that sequence is valuable.
Testing with one packet size at a time tells you what a device can do. Testing with a realistic mixture tells you what it will actually meet. The trouble is that "a realistic mixture" is not the same in every lab, so the results cannot be compared.
§3What This Covers and What It Does Not
The method defines how to specify unambiguously the sequence of packet sizes used in a load test, on the assumption that a relevant mix of sizes is known to the tester and that the length of the repeating sequence is not very long, in other words under 100 packets. The genome allows an exact sequence to be communicated as a single-line name, which resolves the current ambiguity in results that simply refer to IMIX.
The RFC is emphatic that this aspect matters. No ability has been demonstrated to extrapolate results from one IMIX to another, and certainly none to extrapolate results to other circumstances, even when the mix varies only slightly.
Two limits are set on the ambition. Documenting the exact sequence is ideal, but the memo also covers the case where the sequence of sizes is very long or may be generated by a pseudorandom process. And it is called a colossal non-goal to standardise one or more versions of the IMIX itself; that has been discussed many times on the benchmarking working group's mailing list. The goal is to enable customisation with minimal constraints while fostering repeatable testing once the fixed-size testing is complete. The requirements are written for those performing and reporting laboratory tests, in order to improve clarity and repeatability.
The method is for short repeating sequences of packet sizes. It does not try to say which mixture anyone should use, because that choice depends on the network being represented.
§4The Genome Code
The IMIX Genome is written in the form IMIX - 123456...x, where each number is replaced by the letter that corresponds to the size of the packet at that position in the sequence. The letters cover the standard sizes of RFC 2544 and the larger jumbo sizes: a for 64 bytes, b for 128, c for 256, d for 512, e for 1024, f for 1280, g for 1518, h for 2112, i for 9000 and j for 16000, with z standing for the maximum transmission unit. A note warns that the four-octet Ethernet frame check sequence may fail to detect bit errors in the larger jumbo frames.
A five-packet sequence with sizes 64, 64, 64, 1280 and 1518 is therefore written IMIX - aaafg. If the letter z is used, the tester must specify the length of the maximum transmission unit in the report, since the code alone does not carry it.
The approach allows some flexibility but has constraints. Packet sizes not defined by RFC 2544 would have to be approximated by the sizes available in the table, and the genome for very long sequences can become undecipherable by people. The RFC also asks testers two questions they must answer and record: whether the sequence applies to each source and destination address pair, across multiple pairs in sets, or across all pairs; and whether it applies to each tester port, across multiple ports in sets, or across all ports. The chosen configuration is expressed as a table of source address and port or blade, destination address and port or blade, and the corresponding IMIX. A blade is a component in a multi-component device chassis.
Each letter is one packet size. Writing the sequence out as letters tells the whole story in a few characters: aaafg means a 64-byte packet, another, another, then 1280 bytes and 1518 bytes, over and over.
- a64 bytes — An RFC 2544 standard size.
- b128 bytes — An RFC 2544 standard size.
- c256 bytes — An RFC 2544 standard size.
- d512 bytes — An RFC 2544 standard size.
- e1024 bytes — An RFC 2544 standard size.
- f1280 bytes — An RFC 2544 standard size.
- g1518 bytes — An RFC 2544 standard size.
- h2112 bytes — A jumbo size.
- i9000 bytes — A jumbo size.
- j16000 bytes — A jumbo size.
- zMTU — The maximum transmission unit; its length must be given in the report.
IMIX - aaafg§5Custom Sizes
When the sizes needed are not in the standard table, a custom IMIX is written in the form CUSTOM IMIX - 123456...x, again with one letter per position in the sequence, but using capital letters. The tester must complete a table giving the letter for each size used, where each set of three lower-case letters is replaced by the integer size in octets. So aaa maps to A, bbb to B, ccc to C, ddd to D, eee to E, fff to F, ggg to G, and so on up to Z.
The example in the RFC is a five-packet sequence with sizes aaa = 64, aaa = 64, aaa = 64, ggg = 1020 and ggg = 1020, which is written CUSTOM IMIX - AAAGG. The custom form therefore carries two pieces of information: the order of the sizes, and the table that says what each letter means.
If the sizes you need are not standard, you name them yourself with capital letters and publish the list, so that the code and the list together describe the test exactly.
§6Long or Random Sequences
When the genome cannot be used, because the sheer length of the sequence would make it unmanageable, five options are available. The first is run-length encoding, which may be specified when the sequence decomposes into a series of short repeating sequences: a table gives a count of repeating sequences and the packet-size sequence, for example a count of 20 for the sequence abcd, a count of 5 for ggga and a count of 10 for dcba. The same approach works for the custom form, with capital letters.
The second is a table of proportions, used when the sequence varies within proportional constraints. It gives the IP length, the percentage of the total and the length at other layers, for example 64 bytes at 23 per cent and 82 bytes at other layers, 128 bytes at 67 per cent and 146 bytes at other layers, and 1000 bytes at 10 per cent and 1018 bytes at other layers. This also allows non-standard packet sizes, but it trades the short genome specification and the ability to state the exact sequence for other flexibilities.
The third is a deterministic algorithm, such as a monotonic increase by one octet from a start value up to the maximum transmission unit, in which case the generation algorithm should be reported. The fourth is a pseudorandom length algorithm, where the algorithm should be reported along with the seed value; the RFC also recognises the opportunity to randomise the spacing between packets as well as the size, and says that both the spacing and length algorithms and seeds should be reported when used. The fifth is a pseudorandom sequence that generates an index into the table of packet lengths, where again the algorithm should be reported with the results, along with the seed value if one is used.
Very long or random sequences cannot be written out letter by letter. Instead you describe how they are made: how many times a short pattern repeats, what share of the traffic is each size, or the algorithm and seed that produce them.
- Run-length encodingGive a count and a short pattern that repeats that many times.
- Table of proportionsGive the share of the traffic at each size, plus the length at other layers.
- Deterministic algorithmReport the rule that generates the sizes, such as a step from a start value to the MTU.
- Pseudorandom lengthReport the algorithm and the seed value used to pick the lengths.
- Pseudorandom sequenceReport the algorithm and seed that pick an index into the table of lengths.
§7Keeping the Test Off the Production Network
Benchmarking as described in this memo is limited to technology characterisation using controlled stimuli in a laboratory environment, with dedicated address space and the other constraints of RFC 2544. The benchmarking network topology is an independent test setup and must not be connected to devices that may forward the test traffic into a production network or misroute traffic to the test management network.
Benchmarking is also performed on a black-box basis, relying solely on measurements observable from outside the device or system under test. Special capabilities should not exist in the device under test specifically for benchmarking purposes, and any implications for network security arising from the device should be identical in the lab and in production networks.
The test lab has to stay separate from the real network, and the device under test must not get special treatment that would only exist for testing. What the test reveals about security should be true in the lab and in production alike.