draft-ietf-bmwg-lanswitch-03.txt   draft-ietf-bmwg-lanswitch-04.txt 
Network Working Group R. Mandeville Network Working Group R. Mandeville
INTERNET-DRAFT European Network Laboratories INTERNET-DRAFT European Network Laboratories
Expiration Date: Jul 1997 Feb 1997 Expiration Date: September 1997 March 1997
Benchmarking Terminology for LAN Switching Devices Benchmarking Terminology for LAN Switching Devices
< draft-ietf-bmwg-lanswitch-04.txt >
< draft-ietf-bmwg-lanswitch-03.txt >
Status of this Document Status of this Document
This document is an Internet-Draft. Internet-Drafts are working documents of This document is an Internet-Draft. Internet-Drafts are working
the Internet Engineering Task Force (IETF), its areas, and its working documents of the Internet Engineering Task Force (IETF), its areas,
groups. Note that other groups may also distribute working documents as and its working groups. Note that other groups may also distribute
Internet-Drafts. working documents as Internet-Drafts.
Internet-Drafts are draft documents valid for a maximum of six months and Internet-Drafts are draft documents valid for a maximum of six
may be updated, replaced, or obsoleted by other documents at any time. It is months and may be updated, replaced, or obsoleted by other
inappropriate to use Internet- Drafts as reference material or to cite them documents at any time. It is inappropriate to use Internet- Drafts as
other than as ``work in progress.'' reference material or to cite them other than as ``work in progress.''
To learn the current status of any Internet-Draft, please check the To learn the current status of any Internet-Draft, please check the
``1id-abstracts.txt'' listing contained in the Internet-Drafts Shadow ``1id-abstracts.txt'' listing contained in the Internet-Drafts Shadow
Directories on ds.internic.net (US East Coast), nic.nordu.net (Europe), Directories on ds.internic.net (US East Coast), nic.nordu.net
ftp.isi.edu (US West Coast), or munnari.oz.au (Pacific Rim). (Europe), ftp.isi.edu (US West Coast), or munnari.oz.au (Pacific
Rim).
Distribution of this document is unlimited. Please send comments to Distribution of this document is unlimited. Please send comments to
bmwg@harvard.edu or to the editor. bmwg@harvard.edu or to the editor.
Abstract Abstract
The purpose of this draft is to define and discuss benchmarking terminology The purpose of this draft is to define and discuss benchmarking
for local area switching devices. It is meant to extend the terminology terminology for local area switching devices. It is meant to extend the
already defined for network interconnect devices in RFCs 1242 and 1944 by terminology already defined for network interconnect devices in
the Benchmarking Methodology Working Group (BMWG) of the Internet RFCs 1242 and 1944 by the Benchmarking Methodology Working
Engineering Task Force (IETF) and prepare the way for a discussion on Group (BMWG) of the Internet Engineering Task Force (IETF) and
benchmarking methodology for local area switches. prepare the way for a discussion on benchmarking methodology for
local area switches.
LAN switches are one of the principal sources of new bandwidth in the local LAN switches are one of the principal sources of new bandwidth in
area. The multiplicity of products brought to market makes it desirable to the local area. The multiplicity of products brought to market makes
define a set of terms to be used when evaluating the performance it desirable to define a set of terms to be used when evaluating the
characteristics of local area switching devices. Well-defined terminology performance characteristics of local area switching devices. Well-
will help in providing the user community with complete, reliable and defined terminology will help in providing the user community with
comparable data on LAN switches. complete, reliable and comparable data on LAN switches.
1. Introduction 1. Introduction
The purpose of this draft is to discuss and define terminology for the The purpose of this draft is to discuss and define terminology for the
benchmarking of local area network switches. Although it might be found benchmarking of local area network switches. Although it might be
useful to apply some of the terms defined here to a broader range of network found useful to apply some of the terms defined here to a broader
interconnect devices, this draft primarily deals with devices which switch range of network interconnect devices, this draft primarily deals with
frames at the Medium Access Control (MAC) layer. It defines terms in devices which switch frames at the Medium Access Control (MAC)
relation to throughput, latency, address handling and filtering. layer. It defines terms in relation to forwarding performance, latency,
address handling and filtering.
2. Term definitions 2. Term definitions
A previous document, "Benchmarking Terminology for Network Interconnect RFC 1242 "Benchmarking Terminology for Network Interconnect
Devices" (RFC 1242), defined many of the terms that are used in this Devices" defines many of the terms that are used in this document.
document. The terminology document should be consulted before attempting to The terminology document should be consulted before attempting to
make use of this document. A more recent document, "Benchmarking Methodology make use of this document. RFC 1944 "Benchmarking Methodology
for Network Interconnect Devices" (RFC 1944), defined a number of test for Network Interconnect Devices" defines a number of test
procedures which are directly applicable to switches. Since it discusses a procedures which are directly applicable to switches. Since it
number of terms relevant to benchmarking switches it should also be consulted. discusses a number of terms relevant to benchmarking switches it
A number of new terms applicable to benchmarking switches are defined below should also be consulted.
using the format for definitions set out in Section 2 of RFC 1242. RFCs 1242 A number of new terms applicable to benchmarking switches are
and 1944 already contain discussions of some of these terms. defined below using the format for definitions set out in Section 2 of
RFC 1242. RFCs 1242 and 1944 already contain discussions of
some of these terms. The new terms are defined in groups.
2. 1. Reminder of RFC 1242 definition format 2. 1. Reminder of RFC 1242 definition format
Term to be defined. (e.g., Latency) Term to be defined. (e.g., Latency)
Definition: Definition:
The specific definition for the term. The specific definition for the term.
Discussion: Discussion:
A brief discussion about the term, it's application A brief discussion about the term, its application
and any restrictions on measurement procedures. and any restrictions on measurement procedures.
Measurement units: Measurement units:
The units used to report measurements of this The units used to report measurements of this
term, if applicable. term, if applicable.
Issues: Issues:
List of issues or conditions that effect this term. List of issues or conditions that effect this term.
See Also: See Also:
List of other terms that are relevant to the discussion List of other terms that are relevant to the discussion
of this term. of this term.
2.2. Unidirectional traffic 3. Index of terms defined
3.1 Devices
3.1.1 Device under test (DUT)
3.1.2 System under test (SUT)
3.2 Traffic patterns
3.2.1 Unidirectional traffic
3.2.2 Bidirectional traffic
3.2.3 Partially meshed traffic
3.2.4 Fully meshed traffic
3.3 Bursts
3.3.1 Burst
3.3.2 Burst size
3.3.3 Inter-burst gap (IBG)
3.4 Loads
3.4.1 Intended load
3.4.2 Offered load
3.4.3 Maximum offered load
3.4.4 Overloading
3.5 Forwarding rates
3.5.1 Forwarding rate at maximum load
3.5.2 Maximum forwarding rate
3.6 Congestion control
3.6.1 Backpressure
3.6.2 Forward pressure
3.6.3 Head of line blocking
3.7 Address handling
3.7.1 Address caching capacity
3.7.2 Address learning rate
3.7.3 Flood count
3.8 Errored frame filtering
3.8.1 Errored frames
3.9 Broadcasts
3.9.1 Broadcast forwarding rate at maximum load
3.9.2 Broadcast latency
3.1 Devices
This group applies to all types of networking devcies.
3.1.1 Device under test (DUT)
Definition: Definition:
The network forwarding device to which stimuli is offered and
response measured.
Single or multiple streams of frames forwarded in one direction only from Discussion:
one or more ports of a switching device designated as input ports to one or A single stand-alone or modular unit generally equipped with its own
more other ports of the device designated as output ports. power supply.
Measurement units:
n/a
Issues:
See Also:
System under test (SUT) (3.1.2)
3.1.2 System Under Test (SUT).
Definition:
The collective set of network devices to which stimuli is offered as a
single entity and response measured.
Discussion: Discussion:
This definition conforms to the discussion in section 16 of RFC 1944 on A system under test may be comprised of a variety of networking
multi-port testing which describes how unidirectional traffic can be offered devices. Some devices may be active in the forwarding decision-
to ports of a device to measure throughput. making process, such as routers or switches; other devices may be
passive such as CSU/DSUs. Regardless of constituent components,
the system is treated as a singular entity to which stimuli is offered
and response measured.
Measurement units:
n/a
Issues:
See Also:
Device under test (DUT) (3.1.1)
3.2 Traffic patterns
This group applies to the distribution of frames forwarded by any
DUT/SUT.
3.2.1 Unidirectional traffic
Definition:
Single or multiple streams of frames forwarded in one direction only
from a set of input ports to a set of output ports.
Discussion:
This definition conforms to the discussion in section 16 of RFC
1944 on multi-port testing which describes how unidirectional traffic
can be offered to ports of a device to measure throughput.
Unidirectional traffic is also appropriate for: Unidirectional traffic is also appropriate for:
- the measurement of the minimum inter-frame gap - the measurement of the minimum inter-frame gap
- the creation of many-to-one or one-to-many port overload - the creation of many-to-one or one-to-many port overload
- the detection of head of line blocking - the detection of head of line blocking
- the measurement of throughput when congestion control mechanisms are active - the measurement of throughput when congestion control
mechanisms are active
Unidirectional traffic can be used to load the ports of a switching device Unidirectional traffic can be used to load the ports of a DUT/SUT in
in different ways. For example unidirectional traffic can be sent to two or different ways. For example unidirectional traffic can be sent to two
more input ports from an external source and switched by the device under or more input ports from an external source and forwarded by the
test to a single output port (n-to-1) or such traffic can be sent to a DUT/SUT to a single output port (n-to-1) or such traffic can be sent
single input port and switched by the device under test to two or more to a single input port and forwarded by the DUT/SUT to two or more
output ports (1-to-n). Such patterns can be combined to test for head of output ports (1-to-n). Such patterns can be combined to test for head
line blocking or to measure throughput when congestion control mechanisms of line blocking or to measure throughput when congestion control
are active. mechanisms are active.
When devices are equipped with ports running at different media rates the When a DUT/SUT is equipped with ports running at different media
number of input streams required to load or overload an output port or ports rates the number of input streams required to load or overload an
will vary. output port or ports will vary.
The measurement of the minimum inter-frame gap serves to detect violations It should be noted that measurement of the minimum inter-frame gap
of the IEEE 802.3 standard. serves to detect violations of the IEEE 802.3 standard.
Issues: Issues:
half duplex / full duplex half duplex / full duplex
Measurement units: Measurement units:
n/a n/a
See Also: See Also:
bidirectional traffic (2.3) bidirectional traffic (3.2.2)
meshed traffic (2.4) partially meshed traffic (3.2.3)
fully meshed traffic (3.2.4)
2.3. Bidirectional traffic 3.2.2 Bidirectional traffic
Definition: Definition:
Two or more streams of frames forwarded in opposite directions between at Two or more streams of frames forwarded in opposite directions
least two or more ports of a switching device. between two or more ports of a DUT/SUT.
Discussion: Discussion:
This definition conforms to the discussions in sections 14 and 16 of RFC This definition conforms to the discussions in sections 14 and 16 of
1944 on bidirectional traffic and multi-port testing. RFC 1944 on bidirectional traffic and multi-port testing.
Bidirectional traffic MUST be offered when measuring throughput on full Bidirectional traffic MUST be offered when measuring throughput
duplex ports of a switching device. on full duplex ports of a switching device.
Issues: Issues:
truncated binary exponential back-off algorithm truncated binary exponential back-off algorithm
Measurement units: Measurement units:
n/a n/a
See Also: See Also:
unidirectional traffic (2.2) unidirectional traffic (3.2.1)
meshed traffic (2.4) partially meshed traffic (3.2.3)
fully meshed traffic (3.2.4)
2.4. Meshed traffic 3.2.3 Partially meshed traffic
Definition: Definition:
Multiple streams of frames switched simultaneously between all of a Streams of frames forwarded between a set of input ports and a set of
designated number of ports of a switching device such that each of the ports output ports of a DUT/SUT with a one to many, many to one or many
under test will both send frames to and receive frames from all of the other to many mapping of input ports to output ports.
ports under test.
Discussion: Discussion:
This definition follows from the discussions in sections 14 and 16 of RFC This definition follows from the discussions in sections 14 and 16 of
1944 on bidirectional traffic and multi-port testing and readily extends to RFC 1944 on bidirectional traffic and multi-port testing and readily
configurations with multiple switching devices linked together over backbone extends to configurations with multiple switching devices linked
connections. together over backbone connections. Meshed traffic can be
As with bidirectional multi-port traffic, meshed traffic exercises both the unidirectional or bidirectional.
transmission and reception sides of the ports of a switching device. Since
ports are not divided into two groups every port forwards frames to and
receives frames from every other port. The total number of individual
streams when traffic is meshed over n switched ports equals n x (n - 1).
This compares with n x (n / 2) such streams in a bidirectional multi-port
test. It should be noted that bidirectional multiport traffic can load
backbone connections linking together two switching devices more than meshed
traffic.
Meshed traffic on half duplex ports is inherently bursty since ports must
interrupt transmission whenever they receive frames. Bursty meshed traffic
which is characteristic of real network traffic simultaneously exercises
many of the component parts of a switching device such as input and output
buffers, buffer allocation mechanisms, aggregate switching capacity,
processing speed and behavior of the medium access controller.
When offering bursty meshed traffic to a device under test a number of
variables have to be considered. These include frame size, the number of
frames within bursts, the interval between bursts as well as the
distribution of load between incoming and outgoing traffic. The terms burst,
burst size and inter-burst gap are defined in sections 2.5, 2.6 and 2.7
below. Load and balanced load are defined in sections 2.8 and 2.10 below.
Measurement units: Measurement units:
n/a n/a
Issues: Issues:
half duplex / full duplex half duplex / full duplex
See Also: See Also:
unidirectional traffic (2.2) unidrectional traffic (3.2.1)
bidirectional traffic (2.3) bidirectional traffic (3.2.2)
burst (2.5) fully meshed traffic (3.2.4)
burst size (2.6) bursts (3.3)
inter-burst gap (2.7)
load (2.8)
balanced load (2.10)
2.5 Burst 3.2.4 Fully meshed traffic
Definition: Definition:
A sequence of frames transmitted with the minimum inter-frame gap allowed by Streams of frames switched simultaneously between all of a
the medium. designated number of ports of a device such that each of the ports
under test will both send frames to and receive frames from all of the
other ports under test.
Discussion: Discussion:
This definition follows from discussions in section 3.16 of RFC 1242 and As with bidirectional multi-port traffic, meshed traffic exercises both
section 21 of RFC 1944 which describes cases where it is useful to consider the transmission and reception sides of the ports of a switching
isolated frames as single frame bursts. device. Since ports are not divided into two groups every port
forwards frames to and receives frames from every other port. The
total number of individual streams when traffic is meshed over n
switched ports equals n x (n - 1). This compares with n x (n / 2) such
streams in a bidirectional multi-port test. It should be noted that
bidirectional multiport traffic can load backbone connections linking
together two switching devices more than meshed traffic.
It should be noted that bidirectional meshed traffic on half duplex
ports is inherently bursty since ports must interrupt transmission
whenever they receive frames. This kind of bursty meshed traffic is
characteristic of real network traffic and can be advantageously used
to diagnose a DUT/SUT by exercising many of its component parts
simultaneously. Additional inspection may be warranted to correlate
the frame forwarding capacity of a DUT/SUT when offered meshed
traffic and the behavior of individual elements such as input and
output buffers, buffer allocation mechanisms, aggregate switching
capacity, processing speed or medium access control.
When offering bursty meshed traffic to a DUT/SUT a number of
variables have to be considered. These include frame size, the number
of frames within bursts, the interval between bursts as well as the
distribution of load between incoming and outgoing traffic. Terms
related to bursts are defined in section 3.3 below.
Measurement units: Measurement units:
n/a n/a
Issues: Issues:
half duplex / full duplex
See Also: See Also:
burst size (2.6) unidrectional traffic (3.2.1)
bidirectional traffic (3.2.2)
partially meshed traffic (3.2.4)
bursts (3.3)
2.6 Burst size 3.3 Bursts
This group applies to the intervals defining traffic forwarded by
DUT/SUT.
3.3.1 Burst
Definition:
A sequence of frames transmitted with the minimum inter-frame gap
allowed by the medium.
Discussion:
This definition follows from discussions in section 3.16 of RFC
1242 and section 21 of RFC 1944 which describes cases where it is
useful to consider isolated frames as single frame bursts.
Measurement units:
n/a
Issues:
See Also:
burst size (3.3.2)
inter-burst gap (IBG) (3.2.3)
3.2.2 Burst size
Definition: Definition:
The number of frames in a burst. The number of frames in a burst.
Discussion: Discussion:
Burst size can range from one to infinity. In unidirectional streams there Burst size can range from one to infinity. In unidirectional streams
is no theoretical limit to burst length. When traffic is bidirectional or there is no theoretical limit to burst length. When traffic is
meshed bursts on half duplex media are finite since ports interrupt bidirectional or meshed bursts on half duplex media are finite since
transmission intermittently to receive frames. ports interrupt transmission intermittently to receive frames.
On real networks burst size will normally increase with window size. This On real networks burst size will normally increase with window size.
makes it desirable to test devices with small as well as large burst sizes. This makes it desirable to test devices with small as well as large
burst sizes.
Measurement units: Measurement units:
number of N-octet frames number of N-octet frames
Issues: Issues:
See Also: See Also:
burst (2.5) burst (3.3.1)
inter-burst gap (IBG) (3.2.3)
2.7 Inter-burst gap (IBG) 3.2.3 Inter-burst gap (IBG)
Definition: Definition:
The interval between two bursts. The interval between two bursts.
Discussion: Discussion:
This definition conforms to the discussion in section 20 of RFC 1944 on This definition conforms to the discussion in section 20 of RFC
bursty traffic. 1944 on bursty traffic.
Bidirectional and meshed streams of traffic are inherently bursty since Bidirectional and meshed streams of traffic are inherently bursty
ports share their time between receiving and transmitting frames. External since ports share their time between receiving and transmitting
sources offering bursty traffic for a given frame size and burst size must frames. External sources offering bursty traffic for a given frame size
adjust the inter-burst gap to achieve a specified rate of transmission. and burst size must adjust the inter-burst gap to achieve a specified
rate of transmission.
Measurement units: Measurement units:
nanoseconds nanoseconds
microseconds microseconds
milliseconds milliseconds
seconds seconds
Issues: Issues:
See Also: See Also:
burst size (2.6) burst (3.3.1)
burst size (3.2.2)
2.8 Port load 3.4 Loads
This group applies to the rates that traffic is offered to any
DUT/SUT.
3.4.1 Intended load
Definition: Definition:
The total number of frames per second that a switched port can be observed The number of frames per second that an external source attempts to
to transmit and receive in response to an offered load. transmit to a DUT/SUT for forwarding to a specified output port or
ports.
Discussion: Discussion:
Port load can be expressed in a number of ways: bits per second, frames per Collisions on CSMA/CD links or the action of congestion control
second with the frame size specified or as a percentage of the maximum frame mechanisms can effect the rate at which an external source of traffic
rate allowed by the medium for a given frame size. In the case of transmits frames to a DUT/SUT. This makes it useful to distinguish
bidirectional or meshed traffic port load is the sum of the frames the load that an external source attempts to apply to a DUT/SUT and
transmitted and received on a port per second. The load on an Ethernet port the load it is observed or measured to apply.
which is transmitting and receiving a total of 7440 64-byte frames per In the case of Ethernet an external source of traffic must implement
second equals 50% given that the maximum rate of transmission on an Ethernet the truncated binary exponential back-off algorithm to ensure that it
is 14880 64-byte frames per second. is accessing the medium legally.
Measurement units: Measurement units:
bits per second bits per second
frames per second with the frame size specified N-octets per second
as a percentage of the maximum frame rate allowed by the medium for a given (N-octets per second / media_maximum-octets per second) x 100
frame size.
Issues: Issues:
See Also: See Also:
bidirectional traffic (2.3) offered load (3.4.2)
meshed traffic (2.4)
offered load (2.9)
overload (2.12)
2.9 Offered load 3.4.2 Offered load
The total number of frames per second that an external source attempts to
transmit or address to a port of a device under test. Definition:
The number of frames per second that an external source can be
observed or measured to transmit to a DUT/SUT for forwarding to a
specified output port or ports.
Discussion: Discussion:
Collisions on CSMA/CD links or the action of congestion control mechanisms The load which an external device can be observed to apply to a
can reduce the rate of transmission of external sources sending or DUT/SUT may be less than the load the external device attempts to
addressing traffic to a port under test. This makes it useful to distinguish apply due to collisions or the action of congestion control
port load, the load that can be observed on a port, from the load that an mechanisms.
external source offers, that is, attempts to send or address to the port. Frames which are not successfully transmitted by an external source
In the case of Ethernet an external source of traffic must implement the of traffic to a DUT/SUT MUST NOT be counted as transmitted
truncated binary exponential back-off algorithm when executing bidirectional frames when measuring the forwarding rate of a DUT/SUT.
and meshed performance tests to ensure that it is accessing the medium legally. The frame count on a port of a DUT/SUT may exceed the rate at
Frames which are not successfully transmitted by an external source of which an external device offers frames due to the presence of
traffic to the device under test MUST NOT be counted as transmitted frames spanning tree BPDUs (Bridge Protocol Data Units) on 802.1D-
in performance benchmarks. compliant switches or SNMP frames. Such frames should be treated
The frame count on a port of a device under test may exceed the rate at as modifiers as described in section 11 of RFC 1944.
which an external device offers frames due to the presence of spanning tree
BPDUs (Bridge Protocol Data Units) on 802.1D-compliant switches or SNMP
frames. If such frames cannot be inhibited, they MUST be left out of frame
counts in performance benchmarks.
Measurement units: Measurement units:
bits per second bits per second
N-octets per second N-octets per second
(N-octets per second / media_maximum-octets per second) x 100 (N-octets per second / media_maximum-octets per second) x 100
Issues: Issues:
token ring token ring
See also: See also:
port load (2.8) intended load (3.4.1)
2.10 Balanced load 3.4.3 Maximum offered load
Definition: Definition:
Port load when the total number of frames per second that an external source The highest number of frames per second that an external source can
attempts to transmit to a port of a device under test equals the total transmit to a DUT/SUT for forwarding to a specified output port or
number of frames per second that the external source attempts to address to ports.
the same port.
Discussion: Discussion:
There is room for varying the balance between incoming and outgoing traffic The maximum load that an external device can apply to a DUT/SUT
when loading ports with bidirectional and meshed traffic. A balanced load on may not equal the maximum load allowed by the medium. This will
all ports will help avoid unwanted or inadvertent port overloading in be the case when an external source lacks the resources to transmit
throughput tests. frames at the minimum legal inter-frame gap or when it has sufficient
resources to transmit frames below the minimum legal inter-frame
gap. Moreover, maximum load may vary with respect to parameters
other than a medium's maximum theoretical utilization. For example,
on those media employing tokens, maximum load may vary as a
function of Token Rotation Time, Token Holding Time, or the ability
to chain multiple frames to a single token.
The maximum load that an external device applies to a DUT/SUT
MUST be specified when measuring forwarding rates.
Measurement units: Measurement units:
bits per second bits per second
N-octets per second N-octets per second
(N-octets per second / media_maximum-octets per second) x 100 (N-octets per second / media_maximum-octets per second) x 100
Issues: Issues:
See also: See Also:
port load (2.8) offered load (3.4.2)
offered load (2.9)
2.11 Maximum load 3.4.4 Overloading
Definition: Definition:
Port load when the offered load equals the maximum rate allowed by the medium. Attempting to load a DUT/SUT in excess of the maximum rate of
transmission allowed by the medium.
Discussion: Discussion:
Maximum load in balanced bidirectional and meshed traffic tests requires the Overloading can serve to exercise buffers and buffer allocation
number of frames per second an external source attempts to transmit to a algorithms as well as congestion control mechanisms.
port and the number of frames per second the source attempts to address to The number of input port or ports required to overload one or more
the same port to be equally divided and total to the maximum rate of output ports of a DUT/SUT will vary according to the media rates of
transmission allowed by the medium. the ports involved. An external source can also overload a port by
transmitting frames below the minimum inter-frame gap. This can
serve to determine whether a device respects the minimum inter-frame
gap. Overloading can be achieved with unidirectional, bidirectional
and meshed traffic.
Measurement units: Measurement units:
bits per second N-octets per second
frames per second with the frame size specified (N-octets per second / media_maximum-octets per second) x 100
as a percentage of the maximum frame rate allowed by the medium for a given N-octet frames per second
frame size.
Issues: Issues:
See Also: See Also:
bidirectional traffic (2.3) offered load (3.4.2)
meshed traffic (2.4)
port load (2.8)
offered load (2.9)
2.12 Overload 3.5 Forwarding rates
This group applies to the rates at which traffic is forwarded by any
DUT/SUT in response a stimulus.
3.5.1 Forwarding rate
Definition: Definition:
Loading a port or ports in excess of the maximum rate of transmission The number of frames per second that a device can be observed to
allowed by the medium. successfully transmit to the correct destination port in response to a
specified offered load.
Discussion: Discussion:
Overloading can serve to exercise input and output buffers, buffer Unlike throughput defined in section 3.17 of RFC 1242, forwarding
allocation algorithms and congestion control mechanisms. rate makes no explicit reference to frame loss. Forwarding rate,
Port overloading with unidirectional traffic requires a minimum of two input which must only be sampled on the output side of the ports under
and one output ports when the medium rate of all ports is the same. The test, can be measured for unidirectional, bidirectional or meshed
number of input ports will vary according to the media rates of the output traffic and should be sampled in fixed time intervals of one second. If
port or ports under test. longer or shorter intervals are used they should be cited when
Port overloading with bidirectional and meshed traffic requires the offered reporting a device's forwarding rate.
load on each port to exceed the maximum rate of transmission allowed by the It should be noted that the forwarding rate of a DUT/SUT may be
medium. sensitive to the action of congestion control mechanisms.
Measurement units: Measurement units:
N-octet frames per second N-octet frames per second
Issues: Issues:
See Also: See Also:
bidirectional traffic (2.3) offered load (3.4.2)
meshed traffic (2.4) forwarding rate at maximum offered load (3.5.2)
port load (2.8) maximum forwarding rate (3.5.3)
2.13 Forwarding rate 3.5.2 Forwarding rate at maximum offered load
Definition: Definition:
The number of frames per second that a device can be observed to deliver to The number of frames per second that a device can be observed to
the correct output port in response to an offered load. successfully transmit to the correct destination port in response to the
maximum offered load.
Discussion: Discussion:
Forwarding rate does not take frame loss into account and must only be Forwarding rate at maximum offered load may be less than the
sampled on the output side of the ports under test. It can be measured on maximum rate at which a device can be observed to successfully
devices offered unidirectional, bidirectional or meshed traffic with forward traffic. This will be the case when the ability of a device to
balanced loads to help avoid unwanted or inadvertent port overloading in forward frames degenerates when offered traffic at maximum load.
throughput tests. Maximum offered load must be cited when reporting forwarding rate
The forwarding rates of switching devices which exhibit no frame loss may at maximum offered load.
decrease when congestion control mechanisms are active.
Measurement units: Measurement units:
N-octet frames per second N-octet frames per second
Issues: Issues:
See Also: See Also:
port load (2.8) maximum offered load (3.4.3)
offered load (2.9) forwarding rate (3.5.1)
forwarding rate at maximum load (2.14) maximum forwarding rate (3.5.3)
2.14 Forwarding rate at maximum load 3.5.3 Maximum forwarding rate
Definition: Definition:
Forwarding rate when the offered load equals the maximum rate allowed by the The highest forwarding rate of a DUT/SUT taken from an iterative set
medium. of forwarding rate measurements.
Discussion: Discussion:
Forwarding rate at maximum load may be less than the maximum rate at which a The forwarding rate of a device may degenerate before maximum load
device can be observed to successfully forward traffic. is reached. The load applied to a device must be cited when reporting
maximum forwarding rate.
Measurement units: Measurement units:
N-octet frames per second N-octet frames per second
Issues: Issues:
See Also: See Also:
maximum load (2.11) offered load (3.4.2)
forwarding rate (2.13) forwarding rates (3.5.1)
forwarding rate at maximum load (3.5.2)
2.15 Backpressure 3.6 Congestion
This group applies to the behavior of a DUT/SUT when congestion
or contention is present.
3.6.1 Backpressure
Definition: Definition:
Techniques whereby switching devices attempt to avoid frame loss by impeding Any technique used by a DUT/SUT to attempt to avoid frame loss by
external sources of traffic from transmitting frames to congested ports. impeding external sources of traffic from transmitting frames to
congested ports.
Discussion: Discussion:
Some switches send jam signals, for example preamble bits, back to traffic Some switches send jam signals, for example preamble bits, back to
sources when their transmit and/or receive buffers start to overfill. traffic sources when their transmit amd/or receive buffers start to
Switches implementing full duplex Ethernet links may use IEEE 802.3x Flow overfill. Switches implementing full duplex Ethernet links may use
Control for the same purpose. Such devices may incur no frame loss when IEEE 802.3x Flow Control for the same purpose. Such devices may
external sources attempt to offer traffic to congested or overloaded ports. incur no frame loss when external sources attempt to offer traffic to
Jamming and flow control normally slow all traffic transmitted to congested congested or overloaded ports.
input ports including traffic intended for uncongested output ports. It shoulkd be noted that jamming and other flow control methods
may slow all traffic transmitted to congested input ports including
traffic intended for uncongested output ports.
Measurement units: Measurement units:
frame loss on congested port or ports frame loss on congested port or ports
N--octet frames per second between the port applying backpressure and an N--octet frames per second between the port applying backpressure
uncongested and an uncongested
destination port destination port
Issues: Issues:
jamming not explicitly described in standards jamming not explicitly described in standards
See Also: See Also:
forward pressure (2.16) forward pressure (3.6.2)
2.16 Forward pressure 3.6.2 Forward pressure
Definition: Definition:
An illegal technique whereby a device retransmits buffered frames without Methods which depart from or otherwise violate a defined
waiting for the interval calculated by the normal operation of the back-off standardized protocol in an attempt to increase the forwarding
algorithm. performance of a DUT/SUT.
Discussion: Discussion:
Some switches illegally inhibit or abort the truncated binary exponential A DUT/SUT may be found to inhibit or abort backoff algorithms in
backoff algorithm and force access to the medium to avoid frame loss. order to force access to the medium when contention occurs. It
The backoff algorithm should be fair whether the device under test is in a should be noted that the backoff algorithm should be fair whether the
congested or an uncongested state. DUT/SUT is in a congested or an uncongested state. Transmission
below the minimum inter-frame gap or the disregard of flow control
primitives fall into this category.
Measurement units: Measurement units:
intervals in microseconds between transmission retries during 16 successive intervals between frames in microseconds
collisions. intervals in microseconds between transmission retries during 16
successive collisions.
Issues: Issues:
truncated binary exponential backoff algorithm truncated binary exponential backoff algorithm
See also: See also:
backpressure (2.15) backpressure (3.6.1)
2.17 Head of line blocking 3.6.3 Head of line blocking
Definition: Definition:
Frame loss observed on an uncongested output port whenever frames are Frame loss observed on an uncongested output port whenever frames
received from an input port which is also attempting to forward frames to a are received from an input port which is also attempting to forward
congested output port. frames to a congested output port.
Discussion: Discussion:
It is important to verify that a switch does not slow transmission or drop It is important to verify that a switch does not slow transmission or
frames on ports which are not congested whenever overloading on one of its drop frames on ports which are not congested whenever overloading
other ports occurs. on one of its other ports occurs.
Measurement units: Measurement units:
frame loss recorded on an uncongested port when receiving frames from a port frame loss recorded on an uncongested port when receiving frames
which is also forwarding frames to a congested port. from a port which is also forwarding frames to a congested port.
Issues: Issues:
input buffers input buffers
See Also: See Also:
unidirectional traffic (2.2) unidirectional traffic (3.2.1)
2.18 Address handling 3.7 Address handling
This group applies to the process of address resolution which enables
a DUT/SUT to forward frames to the correct destination.
3.7.1 Address caching capacity
Definition: Definition:
The number of MAC addresses per n ports, per module or per device which a The number of MAC addresses per n ports, per module or per device
switch can cache and successfully forward frames to without flooding or that a DUT/SUT can cache and successfully forward frames to
dropping frames. without flooding or dropping frames.
Discussion: Discussion:
Users building networks will want to know how many nodes they can connect to Users building networks will want to know how many nodes they can
a switch. This makes it necessary to verify the number of MAC addresses that connect to a DUT/SUT. This makes it necessary to verify the number
can be assigned per n ports, per module and per chassis before a switch of MAC addresses that can be assigned per n ports, per module and
begins flooding frames. per chassis before a DUT/SUT begins flooding frames.
Measurement units: Measurement units:
number of MAC addresses number of MAC addresses per n ports, per module and/or per chassis
Issues: Issues:
See Also: See Also:
Address learning rate (2.19) Address learning rate (3.7.2)
2.19 Address learning rate 3.7.2 Address learning rate
Definition: Definition:
The maximum rate at which a switch can learn new MAC addresses before The maximum rate at which a switch can learn new MAC addresses
starting to flood or drop frames. before starting to flood or drop frames.
Discussion: Discussion:
Users may want to know how long it takes a switch to build its address Users may want to know how long it takes a switch to build its
tables. This information is useful to have when considering how long it address tables. This information is useful to have when considering
takes a network to come up when many users log on in the morning or after a how long it takes a network to come up when many users log on in
network crash. the morning or after a network crash.
Measurement units: Measurement units:
frames per second with each successive frame sent to the switch containing a frames per second with each successive frame sent to the switch
different source address. containing a different source address.
Issues: Issues:
See Also: address handling (2.18) See Also: address caching capacity (3.7.1)
2.20 Flooding 3.7.3 Flood count
Definition: Definition:
Frames received on ports which do not correspond to the destination MAC Frames forwarded to ports which do not correspond to the
address information. destination MAC address information when traffic is offered to a
DUT/SUT for forwarding.
Discussion: Discussion:
When recording throughput statistics it is important to check that frames When recording throughput statistics it is important to check that
have been forwarded to their proper destinations. Flooded frames MUST NOT be frames have been forwarded to their proper destinations. Flooded
counted as received frames. Both known and unknown unicast frames can be frames MUST NOT be counted as received frames. Both known and
flooded. unknown unicast frames can be flooded.
Measurement units: Measurement units:
N-octet valid frames per second N-octet valid frames
Issues: Issues:
Spanning tree BPDUs. Spanning tree BPDUs.
See Also: See Also:
address caching capacity (3.7.1)
2.21 Illegal frames 3.8 Errored frame filtering
This group applies to frames with errors which a DUT/SUT may
filter.
3.8.1 Errored frames
Definition: Definition:
Frames which are over-sized, under-sized, misaligned or with an errored Frames which are over-sized, under-sized, misaligned or with an
Frame Check Sequence. errored Frame Check Sequence.
Discussion: Discussion:
Switches, unlike IEEE 802.1d compliant brdiges, do not necessarily filter Switches, unlike IEEE 802.1d compliant bridges, do not necessarily
all types of illegal frames. Some switches, for example, which do not store filter all types of illegal frames. Some switches, for example, which
frames before forwarding them to their destination ports may not filter do not store frames before forwarding them to their destination ports
over-sized frames (jabbers) or verify the validity of the Frame Check may not filter over-sized frames (jabbers) or verify the validity of the
Sequence field. Other illegal frames are under-sized frames (runts) and Frame Check Sequence field. Other illegal frames are under-sized
misaligned frames. frames (runts) and misaligned frames.
Measurement units: Measurement units:
N-octet frames filtered or not filtered n/a
Issues: Issues:
See Also: See Also:
2.22 Maximum broadcast forwarding rate 3.9 Broadcasts
This group applies to MAC layer and network layer broadcast
frames.
3.9.1 Broadcast forwarding rate
Definition: Definition:
The number of broadcast frames per second that a switch can be observed to The number of broadcast frames per second that a DUT/SUT can be
deliver to all ports at maximum load. observed to deliver to all ports located within a broadcast domain in
response to a specified offered load.
Discussion: Discussion:
There is no standard forwarding mechanism used by switches to forward There is no standard forwarding mechanism used by switches to
broadcast frames. It is useful to determine the broadcast forwarding rate forward broadcast frames. It is useful to determine the broadcast
for frames switched between ports on the same card, ports on different cards forwarding rate for frames switched between ports on the same card,
in the same chassis and ports on different chassis linked together over ports on different cards in the same chassis and ports on different
backbone connections. chassis linked together over backbone connections. The terms
maximum broadcast forwarding rate and broadcast forwarding rate at
maximum load follow directly from the terms already defined for
forwarding rate measurements in section 3.5 above.
Measurement units: Measurement units:
N-octet frames per second N-octet frames per second
Issues: Issues:
See Also: See Also:
broadcast latency (2.23) forwarding rate at maximum load (3.5.2)
maximum forwarding rate (3.5.3)
broadcast latency (3.9.2)
2.23 Broadcast latency 3.9.2 Broadcast latency
Definition: Definition:
The time required by a switch to forward a broadcast frame to each port The time required by a DUT/SUT to forward a broadcast frame to
located within a broadcast domain. each port located within a broadcast domain.
Discussion: Discussion:
Since there is no standard way for switches to process broadcast frames, Since there is no standard way for switches to process broadcast
broadcast latency may not be the same on all receiving ports of a switching frames, broadcast latency may not be the same on all receiving ports
device. The latency measurements SHOULD be bit oriented as described in 3.8 of a switching device. The latency measurements SHOULD be bit
of RFC 1242. It is useful to determine broadcast latency for frames oriented as described in 3.8 of RFC 1242. It is useful to determine
forwarded between ports on the same card, ports on different cards in the broadcast latency for frames forwarded between ports on the same
same chassis and ports on different chassis linked together over backbone card, ports on different cards in the same chassis and ports on
connections. different chassis linked together over backbone connections.
Measurement units: Measurement units:
nanoseconds nanoseconds
microseconds microseconds
milliseconds milliseconds
seconds seconds
Issues: Issues:
See Also: See Also:
broadcast forwarding rate (2.20) broadcast forwarding rate (3.9.1)
3. Index of definitions
2.1 Reminder of RFC 1242 definition format
Traffic patterns
2.2 Unidirectional traffic
2.3 Bidirectional traffic
2.4 Meshed traffic
Bursts
2.5 Burst
2.6 Burst size
2.7 Inter-burst gap (IBG)
Port loads
2.8 Port load
2.9 Offered load
2.10 Balanced load
2.11 Maximum load
2.12 Overload
Forwarding rates
2.13 Forwarding rate
2.14 Forwarding rate at maximum load
Congestion control
2.15 Backpressure
2.16 Forward pressure
2.17 Head of line blocking
Address handling
2.18 Address handling
2.19 Address learning rate
2.20 Flooding
Filtering
2.21 Illegal frames
Broadcasts 4. References:
2.22 Maximum broadcast forwarding rate 1. RFC 1242 "Benchmarking Terminology for Network Interconnect
2.23 Broadcast latency Devices"
2. RFC 1944 "Benchmarking Methodology for Network Interconnect
Devices"
4. Acknowledgments 5. Acknowledgments
In order of appearance Jean-Christophe Bestaux of European Network A special thanks goes to the IETF BenchMark WorkGroup for the
Laboratories, Ajay Shah of Wandel & Goltermann, Henry Hamon of Netcom many suggestions it collectively made to help complete this draft.
Systems, Stan Kopek of Digital Equipment Corporation, Kevin Dubray of Bay Kevin Dubray (Bay Networks), Jean-Christophe Bestaux (ENL),
Networks, and Doug Ruby of Prominet were all instrumental in getting this Ajay Shah ( WG), Henry Hamon (Netcom Systems), Stan Kopek
draft done. (3Com) and Doug Ruby (Prominet) all provided valuable input at
A special thanks goes to the IETF BenchMark WorkGroup for the many various stages of this project.
suggestions it collectively made to help shape this draft.
The editor The editor
Bob Mandeville Bob Mandeville
5. Editor's Address 5. Editor's Address
Robert Mandeville, ENL Robert Mandeville
European Network Laboratories ENL (European Network Laboratories)
6 Parc Ariane, le Mercure 35, rue Beaubourg
Blvd des Chenes 75003 Paris
78284 Guyancourt
France France
mobile phone: +33 6 07 47 67 10
phone: +33 1 39 44 12 05
fax: + 33 1 39 44 12 06
email: bob.mandeville@eunet.fr email: bob.mandeville@eunet.fr
Phone/voice mail: +33 1 39 44 12 05
Fax: +33 1 39 44 12 06
Mobile phone: +33 6 07 47 67 10
 End of changes. 

This html diff was produced by rfcdiff 1.23, available from http://www.levkowetz.com/ietf/tools/rfcdiff/