draft-ietf-bmwg-dsmterm-02.txt   draft-ietf-bmwg-dsmterm-03.txt 
Network Working Group Jerry Perser Network Working Group Jerry Perser
INTERNET-DRAFT Spirent INTERNET-DRAFT Spirent
Expires in: May 2002 David Newman Expires in: December 2002 David Newman
Network Test Network Test
Sumit Khurana Sumit Khurana
Telcordia Telcordia
Shobha Erramilli Shobha Erramilli
QNetworx QNetworx
Scott Poretsky Scott Poretsky
Avici Systems Avici Systems
November 2001 June 2002
Terminology for Benchmarking Network-layer Terminology for Benchmarking Network-layer
Traffic Control Mechanisms Traffic Control Mechanisms
<draft-ietf-bmwg-dsmterm-02.txt> <draft-ietf-bmwg-dsmterm-03.txt>
Status of this Memo Status of this Memo
This document is an Internet-Draft and is in full conformance with This document is an Internet-Draft and is in full conformance with
all provisions of Section 10 of RFC2026. all provisions of Section 10 of RFC2026.
Internet-Drafts are working documents of the Internet Engineering Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF), its areas, and its working groups. Note that Task Force (IETF), its areas, and its working groups. Note that
other groups may also distribute working documents as Internet- other groups may also distribute working documents as Internet-
Drafts. Drafts.
skipping to change at page 1, line 52 skipping to change at page 1, line 52
Table of Contents Table of Contents
1. Introduction .............................................. 2 1. Introduction .............................................. 2
2. Existing definitions ...................................... 3 2. Existing definitions ...................................... 3
3. Term definitions ............................................3 3. Term definitions ............................................3
3.1 Configuration Terms 3.1 Configuration Terms
3.1.1 Classification .........................................3 3.1.1 Classification .........................................3
3.1.2 Codepoint Set ..........................................4 3.1.2 Codepoint Set ..........................................4
3.1.3 Congestion .............................................4 3.1.3 Congestion .............................................4
3.1.4 Congestion Management ..................................5 3.1.4 Congestion Management ..................................5
3.2 Vectors ....................................................6 3.1.5 Flow ...................................................6
3.2.1 Intended Vector ........................................6 3.2 Measurement Terms
Network-layer Traffic Control Mechanisms Network-layer Traffic Control Mechanisms
3.2.2 Offered Vector .........................................6 3.2.1 Channel Capacity .......................................7
3.2.3 Expected Vectors 3.2.2 Conforming .............................................7
3.2.3.1 Expected Forwarding Vector ...........................7 3.2.3 Nonconforming ..........................................8
3.2.3.2 Expected Loss Vector .................................8 3.2.4 Delay ..................................................8
3.2.3.3 Expected Sequence Vector .............................8 3.2.6 Undifferentiated Response ................................9
3.2.3.4 Expected Delay Vector ................................9 3.3 Sequence Tracking
3.2.3.5 Expected Jitter Vector ..............................10 3.3.1 In-sequence Packet .....................................9
3.2.4 Output Vectors 3.3.2 Out-of-order Packet ...................................10
3.2.4.1 Forwarding Vector ...................................11 3.3.3 Duplicate Packet ......................................11
3.2.4.2 Loss Vector .........................................11 3.3.4 Stream ................................................11
3.2.4.3 Sequence Vector .....................................12 3.3.5 Test Sequence number .................................12
3.2.4.4 Delay Vector ........................................13 3.4 Vectors ...................................................12
3.2.4.5 Jitter Vector .......................................14 3.4.1 Intended Vector .......................................12
3.3 Measurement Terms 3.4.2 Offered Vector ........................................13
3.3.1 Channel Capacity ......................................15 3.4.3 Expected Vectors
3.3.2 Conforming ............................................15 3.4.3.1 Expected Forwarding Vector ........................13
3.3.3 Nonconforming .........................................16 3.4.3.2 Expected Loss Vector ..............................14
3.3.4 Delay .................................................16 3.4.3.3 Expected Sequence Vector ..........................14
3.3.5 Flow ..................................................17 3.4.3.4 Expected Instantaneous Delay Vector ...............15
3.3.6 Stream ................................................18 3.4.3.5 Expected Average Delay Vector .....................16
3.3.7 Test Sequence number ..................................19 3.4.3.6 Expected Maximum Delay Vector .....................17
3.3.8 Undifferentiated Response .............................19 3.4.3.7 Expected Minimum Delay Vector .....................17
4. Security Considerations ....................................20 3.4.3.8 Expected Instantaneous Delay Variation Vector .....18
5. References .................................................20 3.4.3.9 Expected Average Delay Variation Vector ...........19
6. Author's Address ...........................................21 3.4.3.10 Expected Peak-to-peak Delay Variation Vector .....19
7. Full Copyright Statement ...................................22 3.4.4 Output Vectors
3.4.4.1 Forwarding Vector .................................20
3.4.4.2 Loss Vector .......................................20
3.4.4.3 Sequence Vector ...................................21
3.4.4.4 Instantaneous Delay Vector ........................22
3.4.4.5 Average Delay Vector ..............................23
3.4.4.6 Maximum Delay Vector ..............................23
3.4.4.7 Minimum Delay Vector ..............................24
3.4.4.8 Instantaneous Delay Variation Vector ..............25
3.4.4.9 Average Delay Variation Vector ....................26
3.4.4.10 Peak-to-peak Delay Variation Vector ..............27
4. Security Considerations ....................................28
5. References .................................................28
6. Author's Address ...........................................29
7. Full Copyright Statement ...................................30
1. Introduction 1. Introduction
This document describes terminology for the benchmarking of This document describes terminology for the benchmarking of
devices that implement traffic control based on IP precedence or devices that implement traffic control based on IP precedence or
diff-serv code point criteria. diff-serv code point criteria.
New terminology is needed because most existing measurements New terminology is needed because most existing measurements
assume the absence of congestion and only a single per-hop- assume the absence of congestion and only a single per-hop-
behavior. This document introduces several new terms that will behavior. This document introduces several new terms that will
allow measurements to be taken during periods of congestion. allow measurements to be taken during periods of congestion.
Network-layer Traffic Control Mechanisms
Another key difference from existing terminology is the definition Another key difference from existing terminology is the definition
of measurements as observed on egress as well as ingress of a of measurements as observed on egress as well as ingress of a
device/system under test. Again, the existence of congestion device/system under test. Again, the existence of congestion
requires the addition of egress measurements as well as those requires the addition of egress measurements as well as those
taken on ingress; without observing traffic leaving a taken on ingress; without observing traffic leaving a
device/system it is not possible to say whether traffic-control device/system it is not possible to say whether traffic-control
mechanisms effectively dealt with congestion. mechanisms effectively dealt with congestion.
The principal measurements introduced in this document are vectors The principal measurements introduced in this document are vectors
for rate, delay, and jitter, all of which can be observed with or for rate, delay, and jitter, all of which can be observed with or
without congestion of the DUT/SUT. without congestion of the DUT/SUT.
Network-layer Traffic Control Mechanisms
This document describes only those terms relevant to measuring This document describes only those terms relevant to measuring
behavior of a device or a group of devices using one of these two behavior of a device or a group of devices using one of these two
mechanisms. End-to-end and service-level measurements are beyond mechanisms. End-to-end and service-level measurements are beyond
the scope of this document. the scope of this document.
2. Existing definitions 2. Existing definitions
RFC 1242 "Benchmarking Terminology for Network Interconnect RFC 1242 "Benchmarking Terminology for Network Interconnect
Devices" and RFC 2285 "Benchmarking Terminology for LAN Switching Devices" and RFC 2285 "Benchmarking Terminology for LAN Switching
Devices" should be consulted before attempting to make use of this Devices" should be consulted before attempting to make use of this
skipping to change at page 3, line 44 skipping to change at page 4, line 5
3. Term definitions 3. Term definitions
3.1 Configuration Terms 3.1 Configuration Terms
3.1.1 Classification 3.1.1 Classification
Definition: Definition:
Selection of packets based on the contents of packet header Selection of packets based on the contents of packet header
according to defined rules. according to defined rules.
Network-layer Traffic Control Mechanisms
Discussion: Discussion:
Packets can be selected based on the DS field or IP Packets can be selected based on the DS field or IP
Precedence in the packet header. Classification can also be Precedence in the packet header. Classification can also be
based on Multi-Field (MF) criteria such as IP Source and based on Multi-Field (MF) criteria such as IP Source and
destination addresses, protocol and port number. destination addresses, protocol and port number.
Classification determines the per-hop behaviors and traffic Classification determines the per-hop behaviors and traffic
conditioning functions such as shaping and dropping that are conditioning functions such as shaping and dropping that are
to be applied to the packet. to be applied to the packet.
Measurement units: Measurement units:
Network-layer Traffic Control Mechanisms
n/a n/a
Issues:
See Also: See Also:
3.1.2 Codepoint Set 3.1.2 Codepoint Set
Definition: Definition:
The set of all DS Code-points or IP precedence values used The set of all DS Code-points or IP precedence values used
during the test duration. during the test duration.
Discussion: Discussion:
Describes all the code-point markings associated with packets Describes all the code-point markings associated with packets
that are input to the DUT/SUT. For each entry in the that are input to the DUT/SUT. For each entry in the
codepoint set, there are associated vectors describing the codepoint set, there are associated vectors describing the
rate of traffic containing that particular DSCP or IP rate of traffic, delay, loss, or jitter containing that
precedence value. particular DSCP or IP precedence value.
The treatment that a packet belonging to a particular code- The treatment that a packet belonging to a particular code-
point gets is subject to the DUT classifying packets to map point gets is subject to the DUT classifying packets to map
to the correct PHB. Moreover, the forwarding treatment in to the correct PHB. Moreover, the forwarding treatment in
general is also dependent on the complete set of offered general is also dependent on the complete set of offered
vectors. vectors.
Measurement Units: Measurement Units:
n/a n/a
skipping to change at page 4, line 46 skipping to change at page 5, line 4
3.1.3 Congestion 3.1.3 Congestion
Definition: Definition:
A condition in which one or more egress interfaces are A condition in which one or more egress interfaces are
offered more packets than are forwarded at any given instant. offered more packets than are forwarded at any given instant.
Discussion: Discussion:
This condition is a superset of the overload definition [2]. This condition is a superset of the overload definition [2].
The overload definition assumes the congestion is introduced The overload definition assumes the congestion is introduced
Network-layer Traffic Control Mechanisms
strictly by the tester on ingress of a DUT/SUT. That may or strictly by the tester on ingress of a DUT/SUT. That may or
may not be the case here. may not be the case here.
Another difference is that with multiple-DUT measurements, Another difference between congestion and overload occurs
congestion may occur at multiple points. For example, when the SUT comprises multiple elements, in that congestion
multiple edge devices collectively may congest a core device. may occur at multiple points. Consider an SUT comprising
In contrast, overload [1] deals only with overload on multiple edge devices exchanging traffic with a single core
device. Depending on traffic patterns, the edge devices may
induce congestion on multiple egress interfaces on the core
device. In contrast, overload [1] deals only with overload on
ingress. ingress.
Network-layer Traffic Control Mechanisms
Ingress observations alone are not sufficient to cover all Ingress observations alone are not sufficient to cover all
cases in which congestion may occur. A device with an cases in which congestion may occur. A device with an
infinite amount of memory could buffer an infinite amount of infinite amount of memory could buffer an infinite amount of
packets, and eventually forward all of them. However, these packets, and eventually forward all of them. However, these
packets may or may not be forwarded during the test duration. packets may or may not be forwarded during the test duration.
Even though ingress interfaces accept all packets without Even though ingress interfaces accept all packets without
loss, this hypothetical device may still be congested. loss, this hypothetical device may still be congested.
The definition presented here explicitly defines congestion The definition presented here explicitly defines congestion
as an event observable on egress interfaces. Regardless of as an event observable on egress interfaces. Regardless of
skipping to change at page 5, line 21 skipping to change at page 5, line 32
packets may or may not be forwarded during the test duration. packets may or may not be forwarded during the test duration.
Even though ingress interfaces accept all packets without Even though ingress interfaces accept all packets without
loss, this hypothetical device may still be congested. loss, this hypothetical device may still be congested.
The definition presented here explicitly defines congestion The definition presented here explicitly defines congestion
as an event observable on egress interfaces. Regardless of as an event observable on egress interfaces. Regardless of
internal architecture, any device that cannot forward packets internal architecture, any device that cannot forward packets
on one or more egress interfaces is congested. on one or more egress interfaces is congested.
Measurement units: Measurement units:
n/a n/a
Issues:
See Also: See Also:
3.1.4 Congestion Management 3.1.4 Congestion Management
Definition: Definition:
An implementation of one or more per-hop-behaviors to avoid An implementation of one or more per-hop-behaviors to avoid
or minimize the condition of congestion. or minimize the condition of congestion.
Discussion: Discussion:
Congestion management may seek either to control congestion Congestion management may seek either to control congestion
or avoid it altogether. Such mechanisms classify packets or avoid it altogether. Such mechanisms classify packets
based upon IP Precedence or DSCP settings in a packet's IP based upon IP Precedence or DSCP settings in a packet's IP
header. header.
Congestion avoidance mechanisms seek to prevent congestion Congestion avoidance mechanisms seek to prevent congestion
before it actually occurs. before it actually occurs.
Congestion control mechanisms gives one or more service Congestion control mechanisms gives one or flows (with a
classes preferential treatment over other classes during discrete IP Precedence or DSCP value) preferential treatment
periods of congestion. over other classes during periods of congestion.
Measurement units: Measurement units:
n/a
Network-layer Traffic Control Mechanisms
See Also:
3.1.5 Flow
Definition:
A flow is a one or more of packets sharing a common intended
pair of source and destination interfaces.
Discussion:
Packets are grouped by the ingress and egress interfaces they
use on a given DUT/SUT.
A flow can contain multiple source IP addresses and/or
destination IP addresses. All packets in a flow must enter
on the same ingress interface and exit on the same egress
interface, and have some common network layer content.
Microflows [3] are a subset of flows. As defined in [3],
microflows require application-to-application measurement. In
contrast, flows use lower-layer classification criteria.
Since this document focuses on network-layer classification
criteria, we concentrate here on the use of network-layer
identifiers in describing a flow. Flow identifiers also may
reside at the data-link, transport, or application layers of
the ISO model. However, identifiers other than those at the
network layer are out of scope for this document.
A flow may contain a single code point/IP precedence value or
may contain multiple values destined for a single egress
interface. This is determined by the test methodology.
Measurement units:
n/a n/a
Issues: See Also:
Microflow [3]
Streams
Network-layer Traffic Control Mechanisms
3.2 Measurement Terms
3.2.1 Channel Capacity
Definition:
The maximum forwarding rate [2] at which none of the offered
packets are dropped by the DUT/SUT.
Discussion:
Channel capacity measures the packet rate at the egress
interface(s) of the DUT/SUT. In contrast, throughput as
defined in RFC 1242 measures the packet rate is based on the
ingress interface(s) of the DUT/SUT.
Ingress-based measurements do not account for congestion of
the DUT/SUT. Channel capacity, as an egress measurement, does
take congestion into account.
Understanding channel capacity is a necessary precursor to
any measurement involving congestion. Throughput numbers can
be higher than channel capacity because of queueing.
This measurement differs from forwarding rate at maximum
offered load (FRMOL) [2] in that it is intolerant of loss.
Measurement units:
N-octet packets per second
See Also: See Also:
Throughput [1]
Forwarding Rate at Maximum Offered Load [2]
3.2.2 Conforming
Definition:
Packets which lie within specific rate, delay, or jitter
bounds.
Discussion:
A DUT/SUT may be configured to allow a given traffic class to
consume a given amount of bandwidth, or to fall within
predefined delay or jitter boundaries. All packets that lie
within specified bounds are then said to be conforming,
whereas those outside the bounds are nonconforming.
Measurement units:
n/a
See Also:
Expected Vector
Network-layer Traffic Control Mechanisms Network-layer Traffic Control Mechanisms
3.2 Vectors Forwarding Vector
Offered Vector
Nonconforming
3.2.3 Nonconforming
Definition:
Packets that do not lie within specific rate, delay, or
jitter bounds.
Discussion:
A DUT/SUT may be configured to allow a given traffic class to
consume a given amount of bandwidth, or to fall within
predefined delay or jitter boundaries. All packets that do
not lie within these bounds are then said to be
nonconforming.
Measurement units:
n/a
See Also:
Expected Vector
Forwarding Vector
Offered Vector
Conforming
3.2.4 Delay
Definition:
The time interval starting when the last bit of the input IP
packet reaches the input port of the DUT/SUT and ending when
the last bit of the output IP packet is seen on the output
port of the DUT/SUT.
Discussion:
Delay is measured the same regardless of the type of DUT/SUT.
Latency [1] require some knowledge of whether the DUT/SUT is
a "store and forward" or a "bit forwarding" device. The fact
that a DUT/SUT's technology has a lower delay than another
technology should be visible.
By specifying the metric to be inside the Internet protocol,
the tester is relieved from specifying the start/end for
every data link layer protocol that IP runs on. This avoids
determining if the start/end delimiters are included in the
frame. Also heterogeneous data link protocol can be used in
a test.
The measurement point at the end closely simulates the way an
internet datagram is processed. An internet datagram is not
Network-layer Traffic Control Mechanisms
passed up or down the stack unless it is complete.
Completion occurs once the last bit of the IP packet has been
received.
Delay can be run at any offered load. Recommend at or below
the channel capacity for non-congested delay. For congested
delay, run the offered load above the channel capacity.
Measurement units:
Seconds.
See Also:
Latency [1]
3.2.6 Undifferentiated Response
Definition:
The vector(s) obtained when mechanisms used to support diff-
serv or IP precedence are disabled.
Discussion:
Enabling diff-serv or IP precedence mechanisms may impose
additional processing overhead for packets. This overhead may
degrade performance even when traffic belonging to only one
class, the best-effort class, is offered to the device.
Measurements with "undifferentiated response" should be made
to establish a baseline.
The vector(s) obtained with DSCPs or IP precedence enabled
can be compared to the undifferentiated response to determine
the effect of differentiating traffic.
Measurement units:
n/a
3.3 Sequence Tracking
3.3.1 In-sequence Packet
Definition:
A received packet with the expected Test Sequence number.
Discussion:
In-sequence is done on a stream level. As packets are
received on a stream, each packet's Test Sequence number is
compared with the previous packet. Only packets that match
the expected are considered in-sequence.
Network-layer Traffic Control Mechanisms
Packets that do not match the expected Test Sequence number
are counted as _not in-sequence_ or out-of-sequence. Every
packet that is received is either in-sequence or out-of-
sequence. Subtracting the in-sequence from the received
packets (for that stream) can derive the out-of-sequence
count.
Two types of events will prevent the in-sequence from
incrementing: packet loss and reordered packets.
Measurement units:
Packet count
See Also:
Stream
Test Sequence number
3.3.2 Out-of-order Packet
Definition:
A received packet with a Test Sequence number less that
expected.
Discussion:
As a stream of packets enter a DUT/SUT, they include a Stream
Test Sequence number indicating the order the packets were
sent to the DUT/SUT. On exiting the DUT/SUT, these packets
may arrive in a different order. Each packet that was re-
ordered is counted as an Out-of-order Packet.
Certain streaming protocol (such as TCP) require the packets
to be in a certain order. Packets outside this are dropped
by the streaming protocols even though there were properly
received by the IP layer. The type of reordering tolerated
by a streaming protocol varies from protocol to protocol, and
also by implementation.
Out-of-order Packet count is based on the worst case
streaming protocol. It allows for no reordering.
Packet loss does not affect the Out-of-order Packet count.
Only packets that were not received in the order that they
were transmitted.
Measurement units:
Packet count
See Also:
Stream
Test Sequence number
Network-layer Traffic Control Mechanisms
3.3.3 Duplicate Packet
Definition:
A received packet with a Test Sequence number matching a
previously received packet.
Discussion:
Measurement units:
Packet count
See Also:
Stream
Test Sequence number
3.3.4 Stream
Definition:
A group of packets tracked as a single entity by the traffic
receiver. A stream may share a common content such as type
(IP, UDP), packet size, or payload.
Discussion:
Streams are tracked by test sequence number or "unique
signature field" (RFC 2889). Streams define how individual
packet's statistics are grouped together to form an
intelligible summary.
Common stream groupings would be by egress interface,
destination address, source address, DSCP, or IP precedence.
A stream using test sequence numbers can track the ordering
of packets as they transverse the DUT/SUT.
Streams are not restricted to a pair of source and
destination interfaces as long as all packets are tracked as
a single entity. A mulitcast stream can be forward to
multiple destination interfaces.
Measurement units:
n/a
See Also:
Flow
MicroFlow [3]
Test sequence number
Network-layer Traffic Control Mechanisms
3.3.6 Test Sequence number
Definition:
A field in the IP payload portion of the packet that is used
to verify the order of the packets on the egress of the
DUT/SUT.
Discussion:
The traffic generator sets the test sequence number value and
the traffic receiver checks the value upon receipt of the
packet. The traffic generator changes the value on each
packet transmitted based on an algorithm agreed to by the
traffic receiver.
The traffic receiver keeps track of the sequence numbers on a
per stream basis. In addition to number of received packets,
the traffic receiver may also report number of in-sequence
packets, number of out-sequence packets, number of duplicate
packets, and number of reordered packets.
The recommended algorithm to use to change the sequence
number on sequential packets is an incrementing value.
Measurement units:
n/a
See Also:
Stream
3.4 Vectors
A vector is a group of packets all containing a specific DSCP A vector is a group of packets all containing a specific DSCP
or IP precedence value. Vectors are expressed as a pair of or IP precedence value. Vectors are expressed as a pair of
numbers. The first is being the particular diff-serv value. numbers. The first is being the particular diff-serv value.
The second is the metric expressed as a rate, loss The second is the metric expressed as a rate, loss
percentage, delay, or jitter. percentage, delay, or jitter.
3.2.1 Intended Vector 3.4.1 Intended Vector
Definition: Definition:
A vector describing the rate at which packets having a A vector describing the rate at which packets having a
specific code-point (or IP precedence) that an external specific code-point (or IP precedence) that an external
source attempts to transmit to a DUT/SUT. source attempts to transmit to a DUT/SUT.
Discussion: Discussion:
Intended loads across the different code-point classes Intended loads across the different code-point classes
determine the metrics associated with a specific code-point determine the metrics associated with a specific code-point
traffic class. traffic class.
Network-layer Traffic Control Mechanisms
Measurement Units: Measurement Units:
N-octets packets per second N-octets packets per second
Issues:
See Also: See Also:
Offered Vector Offered Vector
Expected Forwarding Vector Expected Forwarding Vector
Expected Loss Vector Expected Loss Vector
Expected Sequence Vector Expected Sequence Vector
Expected Delay Vector Expected Delay Vector
Expected Jitter Vector Expected Jitter Vector
Forwarding Vector Forwarding Vector
Loss Vector Loss Vector
3.2.2 Offered Vector 3.4.2 Offered Vector
Definition: Definition:
A vector describing the measured rate at which packets having A vector describing the measured rate at which packets having
a specific DSCP or IP precedence value are offered to the a specific DSCP or IP precedence value are offered to the
DUT/SUT. DUT/SUT.
Discussion: Discussion:
Offered loads across the different code-point classes, Offered loads across the different code-point classes,
constituting a code-point set, determine the metrics constituting a code-point set, determine the metrics
associated with a specific code-point traffic class. associated with a specific code-point traffic class.
Measurement Units: Measurement Units:
N-octets packets per second N-octets packets per second
Network-layer Traffic Control Mechanisms
Issues:
Packet size.
See Also: See Also:
Expected Forwarding Vector Expected Forwarding Vector
Expected Loss Vector Expected Loss Vector
Expected Sequence Vector Expected Sequence Vector
Expected Delay Vector Expected Delay Vector
Expected Jitter Vector Expected Jitter Vector
Forwarding Vector Forwarding Vector
Codepoint Set Codepoint Set
3.2.3 Expected Vectors 3.4.3 Expected Vectors
3.2.3.1 Expected Forwarding Vector 3.4.3.1 Expected Forwarding Vector
Definition: Definition:
A vector describing the expected output rate of packets A vector describing the expected output rate of packets
having a specific DSCP or IP precedence value. The value is having a specific DSCP or IP precedence value. The value is
dependent on the set of offered vectors and configuration of dependent on the set of offered vectors and configuration of
the DUT. the DUT.
Discussion: Discussion:
Network-layer Traffic Control Mechanisms
The DUT is configured in a certain way in order that service The DUT is configured in a certain way in order that service
differentiation occurs for behavior aggregates when a differentiation occurs for a particular DSCP or IP precedence
specific traffic mix consisting of multiple behavior value when a specific traffic mix consisting of multiple
aggregates is applied. This term attempts to capture the DSCPs or IP precedence values are applied. This term attempts
expected forwarding behavior, for which the device is to capture the expected forwarding behavior when subjected to
configured, when subjected to a certain offered load. a certain offered vectors.
The actual algorithms or mechanism, that the DUT uses to The actual algorithm or mechanism the DUT uses to achieve
achieve service differentiation, is not important in service differentiation is not important in describing the
describing the expected vector. expected forwarding vector.
Measurement units: Measurement units:
N-octet packets per second N-octet packets per second
Issues:
See Also: See Also:
Intended Vector Intended Vector
Offered Vector Offered Vector
Output Vectors Output Vectors
Expected Loss Vector Expected Loss Vector
Expected Sequence Vector Expected Sequence Vector
Expected Delay Vector Expected Delay Vector
Expected Jitter Vector Expected Jitter Vector
Network-layer Traffic Control Mechanisms
3.2.3.2 Expected Loss Vector 3.4.3.2 Expected Loss Vector
Definition: Definition:
A vector describing the percentage of packets, having a A vector describing the percentage of packets, having a
specific DSCP or IP precedence value, that should not be specific DSCP or IP precedence value, that should not be
forwarded. The value is dependent on the set of offered forwarded. The value is dependent on the set of offered
vectors and configuration of the DUT. vectors and configuration of the DUT.
Discussion: Discussion:
The DUT is configured in a certain way in order that service The DUT is configured in a certain way in order that service
differentiation occurs for behavior aggregates when a differentiation occurs for a particular DSCP or IP precedence
specific traffic mix consisting of multiple behavior value when a specific traffic mix consisting of multiple
aggregates is applied. This term attempts to capture the DSCPs or IP precedence values are applied. This term attempts
expected loss behavior, for which the device is configured, to capture the expected forwarding behavior when subjected to
when subjected to a certain offered load. a certain offered vectors.
The actual algorithms or mechanism, that the DUT uses to The actual algorithm or mechanism the DUT uses to achieve
achieve service differentiation, is not important in service differentiation is not important in describing the
describing the expected loss vector. expected loss vector.
Measurement Units: Measurement Units:
Percentage of intended packets that are expected to be Percentage of intended packets that are expected to be
dropped. dropped.
Issues:
See Also: See Also:
Intended Vector Intended Vector
Offered Vector Offered Vector
Expected Forwarding Vector Expected Forwarding Vector
Network-layer Traffic Control Mechanisms
Expected Sequence Vector Expected Sequence Vector
Expected Delay Vector Expected Delay Vector
Expected Jitter Vector Expected Jitter Vector
3.2.3.3 Expected Sequence Vector 3.2.3.3 Expected Sequence Vector
Definition: Definition:
A vector describing the expected sequencing of packets having A vector describing the expected in-sequence packets having a
a specific DSCP or IP precedence value. The value is specific DSCP or IP precedence value. The value is dependent
dependent on the set of offered vectors and configuration of on the set of offered vectors and configuration of the DUT.
the DUT.
Discussion: Discussion:
The DUT is configured in a certain way in order that service The DUT is configured in a certain way in order that service
differentiation occurs for behavior aggregates when a differentiation occurs for a particular DSCP or IP precedence
specific traffic mix consisting of multiple behavior value when a specific traffic mix consisting of multiple
aggregates is applied. This term attempts to capture the DSCPs or IP precedence values are applied. This term attempts
expected sequence behavior, for which the device is to capture the expected forwarding behavior when subjected to
configured, when subjected to a certain offered load. a certain offered vectors.
Network-layer Traffic Control Mechanisms
The actual algorithms or mechanism, that the DUT uses to The actual algorithm or mechanism the DUT uses to achieve
achieve service differentiation, is not important in service differentiation is not important in describing the
describing the expected vector. expected sequence vector.
Measurement Units: Measurement Units:
N-octet packets per second N-octet packets per second
Issues:
See Also: See Also:
Intended Vector Intended Vector
Offered Vector Offered Vector
Output Vectors Output Vectors
Expected Loss Vector Expected Loss Vector
Expected Forwarding Vector Expected Forwarding Vector
Expected Delay Vector Expected Delay Vector
Expected Jitter Vector Expected Jitter Vector
3.2.3.4 Expected Delay Vector 3.4.3.4 Expected Instantaneous Delay Vector
Definition: Definition:
A vector describing the expected delay for packets having a A vector describing the expected delay for packets having a
specific DSCP or IP precedence value. The value is dependent specific DSCP or IP precedence value. The value is dependent
on the set of offered vectors and configuration of the DUT. on the set of offered vectors and configuration of the DUT.
Discussion: Discussion:
The DUT is configured in a certain way in order that service The DUT is configured in a certain way in order that service
differentiation occurs for behavior aggregates when a differentiation occurs for a particular DSCP or IP precedence
specific traffic mix consisting of multiple behavior value when a specific traffic mix consisting of multiple
aggregates is applied. This term attempts to capture the DSCPs or IP precedence values are applied. This term attempts
expected delay behavior, for which the device is configured, to capture the expected forwarding behavior when subjected to
when subjected to a certain offered load. a certain offered vectors.
The actual algorithms or mechanism, that the DUT uses to Network-layer Traffic Control Mechanisms
achieve service differentiation, is not important in
describing the expected delay vector.
Measurement units: The actual algorithm or mechanism the DUT uses to achieve
service differentiation is not important in describing the
expected delay vector.
Measurement units:
Seconds. Seconds.
Issues: See Also:
Intended Vector
Offered Vector
Output Vectors
Expected Loss Vector
Expected Sequence Vector
Expected Forwarding Vector
Expected Jitter Vector
3.4.3.5 Expected Average Delay Vector
Definition:
A vector describing the expected average delay for packets
having a specific DSCP or IP precedence value. The value is
dependent on the set of offered vectors and configuration of
the DUT.
Discussion:
The DUT is configured in a certain way in order that service
differentiation occurs for a particular DSCP or IP precedence
value when a specific traffic mix consisting of multiple
DSCPs or IP precedence values are applied. This term attempts
to capture the expected forwarding behavior when subjected to
a certain offered vectors.
The actual algorithm or mechanism the DUT uses to achieve
service differentiation is not important in describing the
expected average delay vector.
Measurement units:
Seconds.
See Also: See Also:
Intended Vector Intended Vector
Offered Vector Offered Vector
Output Vectors Output Vectors
Expected Loss Vector Expected Loss Vector
Expected Sequence Vector Expected Sequence Vector
Expected Forwarding Vector Expected Forwarding Vector
Expected Jitter Vector Expected Jitter Vector
Network-layer Traffic Control Mechanisms Network-layer Traffic Control Mechanisms
3.2.3.5 Expected Jitter Vector 3.4.3.6 Expected Maximum Delay Vector
Definition: Definition:
A vector describing the expected variation in the delay of A vector describing the expected maximum delay for packets
packet arrival times for packets having specific DSCP or IP having a specific DSCP or IP precedence value. The value is
precedence value. The value is dependent on the set of dependent on the set of offered vectors and configuration of
offered vectors and configuration of the DUT. the DUT.
Discussion: Discussion:
Jitter is the absolute value of the difference between the The DUT is configured in a certain way in order that service
delay measurement of two packets belonging to the same differentiation occurs for a particular DSCP or IP precedence
stream. value when a specific traffic mix consisting of multiple
DSCPs or IP precedence values are applied. This term attempts
to capture the expected forwarding behavior when subjected to
a certain offered vectors.
The jitter between two consecutive packets in a stream is The actual algorithm or mechanism the DUT uses to achieve
reported as the "instantaneous jitter". Instantaneous jitter service differentiation is not important in describing the
can be expressed as |D(i) - D(i-1)| where D equals the delay expected maximum delay vector.
and i is the test sequence number. Packets lost are not
counted in the jitter measurement.
Average Jitter is the average of the instantaneous jitter Measurement units:
measured during the test duration. Seconds.
Peak-to-peak jitter is the maximum delay minus the minimum See Also:
delay of the packets forwarded by the DUT/SUT. Intended Vector
Offered Vector
Output Vectors
Expected Loss Vector
Expected Sequence Vector
Expected Forwarding Vector
Expected Jitter Vector
Measurement units: 3.4.3.7 Expected Minimum Delay Vector
Seconds (instantaneous)
Seconds P-P (peak to peak)
Seconds Avg (average)
Issues: Definition:
A vector describing the expected minimum delay for packets
having a specific DSCP or IP precedence value. The value is
dependent on the set of offered vectors and configuration of
the DUT.
Discussion:
The DUT is configured in a certain way in order that service
differentiation occurs for a particular DSCP or IP precedence
value when a specific traffic mix consisting of multiple
DSCPs or IP precedence values are applied. This term attempts
to capture the expected forwarding behavior when subjected to
a certain offered vectors.
The actual algorithm or mechanism the DUT uses to achieve
service differentiation is not important in describing the
expected minimum delay vector.
Network-layer Traffic Control Mechanisms
Measurement units:
Seconds.
See Also: See Also:
Intended Vector Intended Vector
Offered Vector Offered Vector
Output Vectors Output Vectors
Expected Loss Vector Expected Loss Vector
Expected Sequence Vector Expected Sequence Vector
Expected Delay Vector
Expected Forwarding Vector Expected Forwarding Vector
Expected Jitter Vector
3.2.4 Output Vectors 3.2.3.8 Expected Instantaneous Delay Variation Vector
Definition:
A vector describing the expected variation in the delay of
two consecutive packets' arrival times having a specific DSCP
or IP precedence value. The value is dependent on the set of
offered vectors and configuration of the DUT.
Discussion:
Instantaneous Delay Variation is the absolute value of the
difference between the delay measurement of two packets
belonging to the same stream.
The delay fluctuation between two consecutive packets in a
stream is reported as the "Instantaneous Delay Variation".
Instantaneous Delay Variation can be expressed as |D(i) -
D(i-1)| where D equals the delay and i is the test sequence
number. Packets lost are not counted in the measurement.
Forwarding Vector may contain several Instantaneous Delay
Variation Vectors. For n packets received in a Forwarding
Vector, there is n-1 several Instantaneous Delay Variation
Vectors.
Measurement units:
Seconds
See Also:
Delay
Offered Vector
Output Vectors
Expected Average Delay Variation Vector
Expected Peak-to-peak Delay Variation Vector
Stream
Network-layer Traffic Control Mechanisms Network-layer Traffic Control Mechanisms
3.2.4.1 Forwarding Vector 3.2.3.9 Expected Average Delay Variation Vector
Definition:
A vector describing the expected average variation in the
delay of packet arrival times for packets having specific
DSCP or IP precedence value. The value is dependent on the
set of offered vectors and configuration of the DUT.
Discussion:
Average Delay Variation is the average of all the
Instantaneous Delay Variation Vectors measured during the
test duration.
Measurement units:
Seconds
See Also:
Intended Vector
Offered Vector
Output Vectors
Expected Instantaneous Delay Variation Vector
Expected Peak-to-peak Delay Variation Vector
3.2.3.10 Expected Peak-to-peak Delay Variation Vector
Definition:
A vector describing the expected maximum variation in the
delay of packet arrival times for packets having specific
DSCP or IP precedence value. The value is dependent on the
set of offered vectors and configuration of the DUT.
Discussion:
Peak-to-peak Delay Variation Vector is the maximum delay
minus the minimum delay of the packets (in a vector)
forwarded by the DUT/SUT.
Peak-to-peak Delay Variation is not derived from the
Instantaneous Delay Variation Vector. Peak-to-peak Delay
Variation is based upon all the packets during the test
duration, not just two consecutive packets.
Measurement units:
Seconds
See Also:
Intended Vector
Offered Vector
Output Vectors
Expected Instantaneous Delay Variation Vector
Expected Average Delay Variation Vector
Network-layer Traffic Control Mechanisms
3.4.4 Output Vectors
3.4.4.1 Forwarding Vector
Definition: Definition:
The number of packets per second for all packets containing a The number of packets per second for all packets containing a
specific DSCP or IP precedence value that a device can be specific DSCP or IP precedence value that a device can be
observed to successfully transmit to the correct destination observed to successfully forward to the correct destination
interface in response to an offered vector. interface in response to an offered vector.
Discussion: Discussion:
Forwarding Vector is expressed as pair of numbers. Both the Forwarding Vector is expressed as pair of numbers. Both the
specific DSCP (or IP precedence) value AND the packets per specific DSCP (or IP precedence) value AND the packets per
second value combine to make a vector. second value combine to make a vector.
The Forwarding Vector represents packet rate based on its The Forwarding Vector represents packet rate based on its
specific DSCP (or IP precedence) value. It is not specific DSCP (or IP precedence) value. It is not
necessarily based on a stream or flow. The Forwarding Vector necessarily based on a stream or flow. The Forwarding Vector
may be expressed as per port of the DUT/SUT. However, it must may be expressed as per port of the DUT/SUT. However, it must
be consistent with the Expected Forwarding Vector. be consistent with the Expected Forwarding Vector.
Forwarding Vector is a per-hop measurement. The DUT/SUT may Forwarding Vector is a per-hop measurement. The DUT/SUT may
change the specific DSCP (or IP precedence) value for a change the specific DSCP (or IP precedence) value for a
multiple-hop measurement. multiple-hop measurement.
Measurement units: Measurement units:
N-octet packets per second N-octet packets per second
Issues:
See Also: See Also:
Intended Vector Intended Vector
Offered Vector Offered Vector
Expected Vectors Expected Vectors
Loss Vector Loss Vector
Sequence Vector Sequence Vector
Delay Vector Delay Vectors
Jitter Vector
3.2.4.2 Loss Vector 3.4.4.2 Loss Vector
Definition: Definition:
The percentage of packets containing specific DSCP or IP The percentage of packets containing specific DSCP or IP
precedence value that a DUT/SUT did not transmit to the precedence value that a DUT/SUT did not transmit to the
correct destination interface in response to an offered correct destination interface in response to an offered
vector. vector.
Discussion: Discussion:
Loss Vector is expressed as pair of numbers. Both the Loss Vector is expressed as pair of numbers. Both the
specific DSCP (or IP precedence) value AND the percentage specific DSCP (or IP precedence) value AND the percentage
skipping to change at page 12, line 20 skipping to change at page 21, line 20
port of the DUT/SUT. However, it must be consistent with the port of the DUT/SUT. However, it must be consistent with the
Expected Loss Vector Expected Loss Vector
Loss Vector is a per-hop measurement. The DUT/SUT may change Loss Vector is a per-hop measurement. The DUT/SUT may change
the specific DSCP or IP precedence value for a multiple-hop the specific DSCP or IP precedence value for a multiple-hop
measurement. measurement.
Measurement Units: Measurement Units:
Percentage of offered packets that are not forwarded. Percentage of offered packets that are not forwarded.
Issues:
See Also: See Also:
Intended Vector Intended Vector
Offered Vector Offered Vector
Expected Vectors Expected Vectors
Forwarding Vector Forwarding Vector
Sequence Vector Sequence Vector
Delay Vector Delay Vectors
Jitter Vector
3.2.4.3 Sequence Vector 3.4.4.3 Sequence Vector
Definition: Definition:
The number of packets per second for all packets containing a The number of packets per second for all packets containing a
specific DSCP or IP precedence value that a device can be specific DSCP or IP precedence value that a device can be
observed to transmit out of sequence to the correct observed to transmit in sequence to the correct destination
destination interface in response to an offered vector. interface in response to an offered vector.
Discussion: Discussion:
Sequence Vector is expressed as pair of numbers. Both the Sequence Vector is expressed as pair of numbers. Both the
specific DSCP (or IP precedence) value AND the packets per specific DSCP (or IP precedence) value AND the packets per
second value combine to make a vector. second value combine to make a vector.
The Sequence Vector represents packet rate based on its The Sequence Vector represents packet rate based on its
specific DSCP or IP precedence value. It is not necessarily specific DSCP or IP precedence value. It is not necessarily
based on a stream or flow. The Sequence Vector may be based on a stream or flow. The Sequence Vector may be
expressed as per port of the DUT/SUT. However, it must be expressed as per port of the DUT/SUT. However, it must be
skipping to change at page 13, line 5 skipping to change at page 21, line 56
Sequence Vector is a per-hop measurement. The DUT/SUT may Sequence Vector is a per-hop measurement. The DUT/SUT may
change the specific DSCP or IP precedence value for a change the specific DSCP or IP precedence value for a
multiple-hop measurement. multiple-hop measurement.
Measurement Units: Measurement Units:
N-octet packets per second N-octet packets per second
Issues: Issues:
See Also:
In-sequence Packet
Network-layer Traffic Control Mechanisms Network-layer Traffic Control Mechanisms
See Also:
Intended Vector Intended Vector
Offered Vector Offered Vector
Expected Vectors Expected Vectors
Loss Vector Loss Vector
Forwarding Vector Forwarding Vector
Delay Vector Delay Vectors
Jitter Vector
3.2.4.4 Delay Vector 3.4.4.4 Instantaneous Delay Vector
Definition: Definition:
The delay for packets containing specific DSCP or IP The delay for a packet containing specific DSCP or IP
precedence value that a device can be observed to precedence value that a device can be observed to
successfully transmit to the correct destination interface in successfully transmit to the correct destination interface in
response to an offered vector. response to an offered vector.
Discussion: Discussion:
Delay vector is expressed as pair of numbers. Both the Instantaneous Delay vector is expressed as pair of numbers.
specific DSCP (or IP precedence) value AND delay value Both the specific DSCP (or IP precedence) value AND delay
combine to make a vector. value combine to make a vector.
The Delay Vector represents delay on its specific DSCP or IP The Instantaneous Delay Vector represents delay on its
precedence value. It is not necessarily based on a stream or specific DSCP or IP precedence value. It is not necessarily
flow. The Delay vector may be expressed as per port of the based on a stream or flow. The Delay vector may be expressed
DUT/SUT. However, it must be consistent with the Expected as per port of the DUT/SUT. However, it must be consistent
Delay vector. with the Expected Delay vectors.
Delay vector is measured similarly regardless of the type of Instantaneous Delay Vector is a per-hop measurement. The
DUT/SUT. Latency [1] require some knowledge of whether the DUT/SUT may change the specific DSCP or IP precedence value
DUT/SUT is a "store and forward" or a "bit forwarding" for a multiple-hop measurement.
device. The fact that a DUT/SUT's technology has a lower
delay than another technology should be visible.
Delay Vector is a per-hop measurement. The DUT/SUT may Instantaneous Delay vector can be obtained at any offered
change the specific DSCP or IP precedence value for a load. Recommend at or below the channel capacity in the
multiple-hop measurement. absence of congestion. For congested delay, run the offered
load above the channel capacity.
Delay vector can be obtained at any offered load. Recommend Forwarding Vector may contain several Instantaneous Delay
at or below the channel capacity in the absence of Vectors. For every packet received in a Forwarding Vector,
congestion. For congested delay, run the offered load above there is a corresponding Instantaneous Delay Vector.
the channel capacity.
Measurement Units: Measurement Units:
seconds Seconds
Issues:
See Also: See Also:
Delay Delay
Network-layer Traffic Control Mechanisms
Intended Vector Intended Vector
Offered Vector Offered Vector
Expected Delay Vector Expected Delay Vectors
Loss Vector Average Delay Vector
Forwarding Vector Maximum Delay Vector
Jitter Vector Minimum Delay Vector
Network-layer Traffic Control Mechanisms
3.2.4.5 Jitter Vector 3.4.4.5 Average Delay Vector
Definition: Definition:
The variation in the delay for packets containing specific The average delay for packets containing specific DSCP or IP
DSCP or IP precedence value that a device can be observed to precedence value that a device can be observed to
successfully transmit to the correct destination interface in successfully transmit to the correct destination interface in
response to an offered vector. response to an offered vector.
Discussion: Discussion:
Jitter is the absolute value of the difference between the Average Delay vector is expressed as pair of numbers. Both
delay measurement of two packets belonging to the same the specific DSCP (or IP precedence) value AND delay value
stream.
Jitter vector is expressed as pair of numbers. Both the
specific DSCP (or IP precedence) value AND jitter value
combine to make a vector. combine to make a vector.
The jitter between two consecutive packets in a stream is The Delay Vector represents delay on its specific DSCP or IP
reported as the "instantaneous jitter". Instantaneous jitter precedence value. It is not necessarily based on a stream or
can be expressed as |D(i) - D(i-1)| where D equals the delay flow. The Delay vector may be expressed as per port of the
and i is the test sequence number. Packets lost are not DUT/SUT. However, it must be consistent with the Expected
counted in the jitter measurement. Delay vector.
Jitter vector is a per-hop measurement. The DUT/SUT may
change the specific DSCP or IP precedence value for a
multiple-hop measurement.
Average Jitter is the average of the instantaneous jitter The Average Delay Vector is computed by averaging all the
measured during the test duration. Instantaneous Delay Vectors for a given vector.
Peak-to-peak Jitter is the maximum delay minus the minimum Average Delay Vector is a per-hop measurement. The DUT/SUT
delay of the packets forwarded by the DUT/SUT. may change the specific DSCP or IP precedence value for a
multiple-hop measurement.
Measurement units: Average Delay vector can be obtained at any offered load.
Seconds (instantaneous) Recommend at or below the channel capacity in the absence of
Seconds P-P (peak to peak) congestion. For congested delay, run the offered load above
Seconds Avg (average) the channel capacity.
Issues: Measurement Units:
Seconds
See Also: See Also:
Delay
Intended Vector Intended Vector
Offered Vector Offered Vector
Expected Vectors Expected Delay Vectors
Network-layer Traffic Control Mechanisms Instantaneous Delay Vector
Maximum Delay Vector
Loss Vector Minimum Delay Vector
Sequence Vector
Delay Vector
Forwarding Vector
3.3 Measurement Terms
3.3.1 Channel Capacity 3.4.4.6 Maximum Delay Vector
Definition: Definition:
The maximum forwarding rate [2] at which none of the offered The maximum delay from all packets containing specific DSCP
packets are dropped by the DUT/SUT. or IP precedence value that a device can be observed to
successfully transmit to the correct destination interface in
Discussion: response to an offered vector.
Channel capacity measures the packet rate at the egress
interface(s) of the DUT/SUT. In contrast, throughput as
defined in RFC 1242 measures the packet rate is based on the
ingress interface(s) of the DUT/SUT.
Ingress-based measurements do not account for congestion of
the DUT/SUT. Channel capacity, as an egress measurement, does
take congestion into account.
Understanding channel capacity is a necessary precursor to
any measurement involving congestion. Throughput numbers can
be higher than channel capacity because of queueing.
This measurement differs from forwarding rate at maximum
offered load (FRMOL) [2] in that it is intolerant of loss.
Measurement units:
N-octet packets per second
Issues:
See Also:
Throughput [1]
Forwarding Rate at Maximum Offered Load [2]
3.3.2 Conforming
Definition: Network-layer Traffic Control Mechanisms
Packets which lie within specific rate, delay, or jitter
bounds.
Discussion: Discussion:
A DUT/SUT may be configured to allow a given traffic class to Maximum Delay vector is expressed as pair of numbers. Both
consume a given amount of bandwidth, or to fall within the specific DSCP (or IP precedence) value AND delay value
Network-layer Traffic Control Mechanisms combine to make a vector.
predefined delay or jitter boundaries. All packets that lie The Maximum Delay Vector represents delay on its specific
within specified bounds are then said to be conforming, DSCP or IP precedence value. It is not necessarily based on
whereas those outside the bounds are nonconforming. a stream or flow. The Maximum Delay vector may be expressed
as per port of the DUT/SUT. However, it must be consistent
with the Expected Delay vector.
Measurement units: Maximum Delay Vector is based upon the maximum Instantaneous
Delay Vector of all packets in a Forwarding Vector.
n/a Maximum Delay Vector is a per-hop measurement. The DUT/SUT
may change the specific DSCP or IP precedence value for a
multiple-hop measurement.
Issues: Measurement Units:
Seconds
See Also: See Also:
Expected Vector Delay
Forwarding Vector Intended Vector
Offered Vector Offered Vector
Nonconforming Expected Delay Vectors
Instantaneous Delay Vector
3.3.3 Nonconforming
Definition:
Packets that do not lie within specific rate, delay, or
jitter bounds.
Discussion:
A DUT/SUT may be configured to allow a given traffic class to
consume a given amount of bandwidth, or to fall within
predefined delay or jitter boundaries. All packets that do
not lie within these bounds are then said to be
nonconforming.
Measurement units:
n/a
Issues:
See Also:
Expected Vector
Forwarding Vector Forwarding Vector
Offered Vector Average Delay Vector
Conforming Minimum Delay Vector
3.3.4 Delay 3.4.4.7 Minimum Delay Vector
Definition: Definition:
The time interval starting when the last bit of the input IP The minimum delay from all packets containing specific DSCP
packets reaches the input port of the DUT/SUT and ending when or IP precedence value that a device can be observed to
the last bit of the output IP packets is seen on the output successfully transmit to the correct destination interface in
port of the DUT/SUT. response to an offered vector.
Discussion: Discussion:
Delay vector is expressed as pair of numbers. Both the
specific DSCP (or IP precedence) value AND delay value
combine to make a vector.
Network-layer Traffic Control Mechanisms The Minimum Delay Vector represents delay on its specific
DSCP or IP precedence value. It is not necessarily based on
Delay is measured the same regardless of the type of DUT/SUT. a stream or flow. The Minimum Delay vector may be expressed
Latency [1] require some knowledge of whether the DUT/SUT is as per port of the DUT/SUT. However, it must be consistent
a "store and forward" or a "bit forwarding" device. The fact with the Expected Delay vector.
that a DUT/SUT's technology has a lower delay than another
technology should be visible.
By specifying the metric to be inside the Internet protocol,
the tester is relieved from specifying the start/end for
every data link layer protocol that IP runs on. This avoids
determining if the start/end delimiter are included in the
frame. Also heterogeneous data link protocol can be used in
a test.
The measurement point at the end is closely simulates the way Network-layer Traffic Control Mechanisms
an internet datagram is processed. An internet datagram is
not passed up or down the stack unless it is complete.
Completion occurs once the last bit of the IP packet has been
received.
Delay can be run at any offered load. Recommend at or below Minimum Delay Vector is based upon the minimum Instantaneous
the channel capacity for non-congested delay. For congested Delay Vector of all packets in a Forwarding Vector.
delay, run the offered load above the channel capacity.
Measurement units: Minimum Delay Vector is a per-hop measurement. The DUT/SUT
may change the specific DSCP or IP precedence value for a
multiple-hop measurement.
Seconds. Minimum Delay vector can be obtained at any offered load.
Recommend at or below the channel capacity in the absence of
congestion. For congested delay, run the offered load above
the channel capacity.
Issues: Measurement Units:
Seconds
See Also: See Also:
Latency [1] Delay
Intended Vector
Offered Vector
Expected Delay Vectors
Instantaneous Delay Vector
Forwarding Vector
Average Delay Vector
Maximum Delay Vector
3.3.5 Flow 3.4.4.8 Instantaneous Delay Variation Vector
Definition: Definition:
A flow is a one or more of packets sharing a common intended The variation in the delay for two consecutive packets
pair of source and destination interfaces. containing specific DSCP or IP precedence value that a device
can be observed to successfully transmit to the correct
destination interface in response to an offered vector.
Discussion: Discussion:
Packets are grouped by the ingress and egress interfaces they Instantaneous Delay Variation is the absolute value of the
use on a given DUT/SUT. difference between the delay measurement of two packets
belonging to the same stream.
A flow can contain multiple source IP addresses and/or Jitter vector is expressed as pair of numbers. Both the
destination IP addresses. All packets in a flow must enter specific DSCP (or IP precedence) value AND jitter value
on the same ingress interface and exit on the same egress combine to make a vector.
interface, and have some common network layer content.
The delay fluctuation between two consecutive packets in a
stream is reported as the "Instantaneous Delay Variation".
Instantaneous Delay Variation can be expressed as |D(i) -
D(i-1)| where D equals the delay and i is the test sequence
number. Packets lost are not counted in the measurement.
Microflows [3] are a subset of flows. As defined in [3],
microflows require application-to-application measurement. In
contrast, flows use lower-layer classification criteria.
Since this document focuses on network-layer classification
Network-layer Traffic Control Mechanisms Network-layer Traffic Control Mechanisms
criteria, we concentrate here on the use of network-layer Instantaneous Delay Variation Vector is a per-hop
identifiers in describing a flow. Flow identifiers also may measurement. The DUT/SUT may change the specific DSCP or IP
reside at the data-link, transport, or application layers of precedence value for a multiple-hop measurement.
the ISO model. However, identifiers other than those at the
network layer are out of scope for this document.
A flow may contain a single code point/IP precedence value or Forwarding Vector may contain several Instantaneous Delay
may contain multiple values destined for a single egress Variation Vectors. For n packets received in a Forwarding
interface. This is determined by the test methodology. Vector, there is n-1 several Instantaneous Delay Variation
Vectors.
Measurement units: Measurement units:
Seconds
n/a
Issues:
See Also: See Also:
Microflow [3] Delay
Streams Forwarding Vector
Stream
Expected Vectors
Average Delay Variation Vector
Peak-to-peak Delay Variation Vector
3.3.6 Stream 3.4.4.9 Average Delay Variation Vector
Definition: Definition:
A group of packets tracked as a single entity by the traffic The average variation in the delay for packets containing
receiver. A stream may share a common content such as type specific DSCP or IP precedence value that a device can be
(IP, UDP), packet size, or payload. observed to successfully transmit to the correct destination
interface in response to an offered vector.
Discussion: Discussion:
Streams are tracked by Test sequence number or "unique Average Delay Variation is the average of all the
signature field" (RFC 2889). Streams define how individual Instantaneous Delay Variation Vectors measured during the
packet's statistics are grouped together to form an test duration.
intelligible summary.
Common stream groupings would be by egress interface, Average Delay Variation vector is expressed as pair of
destination address, source address, DSCP, or IP precedence. numbers. Both the specific DSCP (or IP precedence) value AND
A stream using Test sequence numbers can track the ordering jitter value combine to make a vector.
of packets as they transverse the DUT/SUT.
Streams are not restricted to a pair of source and Average Delay Variation vector is a per-hop measurement. The
destination interfaces as long as all packets are tracked as DUT/SUT may change the specific DSCP or IP precedence value
a single entity. A mulitcast stream can be forward to for a multiple-hop measurement.
multiple destination interfaces.
Measurement units: Measurement units:
Seconds
n/a
Issues:
Network-layer Traffic Control Mechanisms
See Also: See Also:
Flow Delay
MicroFlow [3] Forwarding Vector
Test sequence number Stream
Expected Vectors
Instantaneous Delay Variation Vector
Peak-to-peak Delay Variation Vector
Network-layer Traffic Control Mechanisms
3.3.7 Test Sequence number 3.4.4.10 Peak-to-peak Delay Variation Vector
Definition: Definition:
A field in the IP payload portion of the packet that is used The maximum possible variation in the delay for packets
to verify the order of the packets on the egress of the containing specific DSCP or IP precedence value that a device
DUT/SUT. can be observed to successfully transmit to the correct
destination interface in response to an offered vector.
Discussion: Discussion:
The traffic generator sets the Test sequence number value and Peak-to-peak Delay Variation Vector is the maximum delay
the traffic receiver checks the value upon receipt of the minus the minimum delay of the packets (in a vector)
packet. The traffic generator changes the value on each forwarded by the DUT/SUT.
packet transmitted based on an algorithm agreed to by the
traffic receiver.
The traffic receiver keeps track of the sequence numbers on a Delay Variation vector is expressed as pair of numbers. Both
per stream basis. In addition to number of received packets, the specific DSCP (or IP precedence) value AND jitter value
the traffic receiver may also report number of in-sequence combine to make a vector.
packets, number of out-sequence packets, number of duplicate
packets, and number of reordered packets.
The recommended algorithm to use to change the sequence Peak-to-peak Delay Variation is not derived from the
number on sequential packets is an incrementing value. Instantaneous Delay Variation Vector. Peak-to-peak Delay
Variation is based upon all the packets during the test
duration, not just two consecutive packets.
Measurement units: Measurement units:
Seconds
n/a
Issues:
See Also: See Also:
Delay
Forwarding Vector
Stream Stream
Expected Vectors
3.3.8 Undifferentiated Response Average Delay Variation Vector
Peak-to-peak Delay Variation Vector
Definition:
The vector(s) obtained when mechanisms used to support diff-
serv or IP precedence are disabled.
Discussion:
Enabling diff-serv or IP precedence mechanisms may impose
additional processing overhead for packets. This overhead may
degrade performance even when traffic belonging to only one
class, the best-effort class, is offered to the device.
Network-layer Traffic Control Mechanisms Network-layer Traffic Control Mechanisms
Measurements with "undifferentiated response" should be made
to establish a baseline.
The vector(s) obtained with DSCPs or IP precedence enabled
can be compared to the undifferentiated response to determine
the effect of differentiating traffic.
Measurement units:
n/a
4. Security Considerations 4. Security Considerations
Documents of this type do not directly affect the security of Documents of this type do not directly affect the security of
the Internet or of corporate networks as long as benchmarking the Internet or of corporate networks as long as benchmarking
is not performed on devices or systems connected to operating is not performed on devices or systems connected to
networks. production networks.
Packets with unintended and/or unauthorized DSCP or IP Packets with unintended and/or unauthorized DSCP or IP
precedence values may present security issues. Determining precedence values may present security issues. Determining
the security consequences of such packets is out of scope for the security consequences of such packets is out of scope for
this document. this document.
5. References 5. References
[1] Bradner, S., Editor, "Benchmarking Terminology for [1] Bradner, S., Editor, "Benchmarking Terminology for
Network Interconnection Devices", RFC 1242, July 1991. Network Interconnection Devices", RFC 1242, July 1991.
 End of changes. 

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