draft-ietf-diffserv-model-04.txt   draft-ietf-diffserv-model-05.txt 
Internet Engineering Task Force Y. Bernet Internet Engineering Task Force Y. Bernet
Diffserv Working Group Microsoft Diffserv Working Group Microsoft
INTERNET-DRAFT S. Blake INTERNET-DRAFT S. Blake
Expires January 2001 Ericsson Expires May 2001 Ericsson
draft-ietf-diffserv-model-04.txt D. Grossman draft-ietf-diffserv-model-05.txt D. Grossman
Motorola Motorola
A. Smith A. Smith
<editor> Allegro Networks
November 2000
An Informal Management Model for Diffserv Routers An Informal Management Model for Diffserv Routers
***** Preliminary Authors' Review DRAFT *****
Status of this Memo Status of this Memo
This document is an Internet-Draft and is in full conformance with all This document is an Internet-Draft and is in full conformance with all
provisions of Section 10 of RFC2026. Internet-Drafts are working provisions of Section 10 of RFC2026. Internet-Drafts are working
documents of the Internet Engineering Task Force (IETF), its areas, and documents of the Internet Engineering Task Force (IETF), its areas, and
its working groups. Note that other groups may also distribute working its working groups. Note that other groups may also distribute working
documents as Internet-Drafts. documents as Internet-Drafts.
Internet-Drafts are draft documents valid for a maximum of six months Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference material time. It is inappropriate to use Internet-Drafts as reference material
or to cite them other than as "work in progress." or to cite them other than as "work in progress."
The list of current Internet-Drafts can be accessed at The list of current Internet-Drafts can be accessed at
http://www.ietf.org/ietf/1id-abstracts.txt. The list of Internet-Draft http://www.ietf.org/ietf/1id-abstracts.txt. The list of Internet-Draft
Shadow Directories can be accessed at http://www.ietf.org/shadow.html. Shadow Directories can be accessed at http://www.ietf.org/shadow.html.
This document is a product of the IETF's Differentiated Services Working This document is a product of the IETF's Differentiated Services working
Group. Comments should be addressed to WG's mailing list at group. Comments should be addressed to WG's mailing list at
diffserv@ietf.org. The charter for Differentiated Services may be found diffserv@ietf.org. The charter for Differentiated Services may be found
at http://www.ietf.org/html.charters/diffserv-charter.html Copyright (C) at http://www.ietf.org/html.charters/diffserv-charter.html Copyright (C)
The Internet Society (2000). All Rights Reserved. The Internet Society (2000). All Rights Reserved.
Distribution of this memo is unlimited. Distribution of this memo is unlimited.
Abstract Abstract
This memo proposes an informal management model of Differentiated This document proposes an informal management model of Differentiated
Services (Diffserv) routers for use in their management and Services (Diffserv) routers for use in their management and
configuration. This model defines functional datapath elements (e.g. configuration. This model defines functional datapath elements (e.g.
classifiers, meters, actions (e.g. marking, absolute dropping, counting, classifiers, meters, actions (e.g. marking, absolute dropping, counting,
multiplexing), algorithmic droppers, queues and schedulers. It describes multiplexing), algorithmic droppers, queues and schedulers. It describes
possible configuration parameters for these elements and how they might possible configuration parameters for these elements and how they might
be interconnected to realize the range of traffic conditioning and per- be interconnected to realize the range of traffic conditioning and per-
hop behavior (PHB) functionalities described in the Diffserv hop behavior (PHB) functionalities described in the Diffserv
Architecture [DSARCH]. Architecture [DSARCH].
The model is intended to be abstract and capable of representing the The model is intended to be abstract and capable of representing the
configuration parameters important to Diffserv functionality for a configuration parameters important to Diffserv functionality for a
variety of specific router implementations. It is not intended as a variety of specific router implementations. It is not intended as a
guide to system implementation nor as a formal modelling description. guide to system implementation nor as a formal modelling description.
This model serves as the rationale for the design of an SNMP MIB [DSMIB] This model serves as the rationale for the design of an SNMP MIB [DSMIB]
and for other configuration interfaces (e.g. [DSPIB]) and, possibly, and for other configuration interfaces (e.g. other policy-management
more detailed formal models (e.g. [QOSDEVMOD]): these should all be protocols) and, possibly, more detailed formal models (e.g.
consistent with this model. [QOSDEVMOD]): these should all be consistent with this model.
1. Introduction 1. Introduction
Differentiated Services (Diffserv) [DSARCH] is a set of technologies Differentiated Services (Diffserv) [DSARCH] is a set of technologies
which allow network service providers to offer services with different which allow network service providers to offer services with different
kinds of network quality-of-service (QoS) objectives to different kinds of network quality-of-service (QoS) objectives to different
customers and their traffic streams. The premise of Diffserv networks is customers and their traffic streams. This document uses terminology
that routers within the core of the network handle packets in different defined in [DSARCH] and other work-in-progress from the IETF's Diffserv
traffic streams by forwarding them using different per-hop behaviors working group (some of these definitions are included here in Section 2
(PHBs). The PHB to be applied is indicated by a Diffserv codepoint for completeness).
(DSCP) in the IP header of each packet [DSFIELD]. Note that this
document uses the terminology defined in [DSARCH, DSTERMS] and in The premise of Diffserv networks is that routers within the core of the
Section 2. network handle packets in different traffic streams by forwarding them
using different per-hop behaviors (PHBs). The PHB to be applied is
indicated by a Diffserv codepoint (DSCP) in the IP header of each packet
[DSFIELD]. The DSCP markings are applied either by a trusted customer or
by the edge routers on entry to the Diffserv network.
The advantage of such a scheme is that many traffic streams can be The advantage of such a scheme is that many traffic streams can be
aggregated to one of a small number of behavior aggregates (BA) which aggregated to one of a small number of behavior aggregates (BA) which
are each forwarded using the same PHB at the router, thereby simplifying are each forwarded using the same PHB at the router, thereby simplifying
the processing and associated storage. In addition, there is no the processing and associated storage. In addition, there is no
signaling, other than what is carried in the DSCP of each packet, and no signaling, other than what is carried in the DSCP of each packet, and no
other related processing that is required in the core of the Diffserv other related processing that is required in the core of the Diffserv
network since QoS is invoked on a packet-by- packet basis. network since QoS is invoked on a packet-by- packet basis.
The Diffserv architecture enables a variety of possible services which The Diffserv architecture enables a variety of possible services which
could be deployed in a network. These services are reflected to could be deployed in a network. These services are reflected to
customers at the edges of the Diffserv network in the form of a Service customers at the edges of the Diffserv network in the form of a Service
Level Specification (SLS) [DSTERMS]. The ability to provide these Level Specification (SLS - see section 2). The ability to provide these
services depends on the availability of cohesive management and services depends on the availability of cohesive management and
configuration tools that can be used to provision and monitor a set of configuration tools that can be used to provision and monitor a set of
Diffserv routers in a coordinated manner. To facilitate the development Diffserv routers in a coordinated manner. To facilitate the development
of such configuration and management tools it is helpful to define a of such configuration and management tools it is helpful to define a
conceptual model of a Diffserv router that abstracts away implementation conceptual model of a Diffserv router that abstracts away implementation
details of particular Diffserv routers from the parameters of interest details of particular Diffserv routers from the parameters of interest
for configuration and management. The purpose of this memo is to define for configuration and management. The purpose of this document is to
such a model. define such a model.
The basic forwarding functionality of a Diffserv router is defined in The basic forwarding functionality of a Diffserv router is defined in
other specifications; e.g., [DSARCH, DSFIELD, AF-PHB, EF-PHB]. other specifications; e.g., [DSARCH, DSFIELD, AF-PHB, EF-PHB].
This document is not intended in any way to constrain or to dictate the This document is not intended in any way to constrain or to dictate the
implementation alternatives of Diffserv routers. It is expected that implementation alternatives of Diffserv routers. It is expected that
router implementers will demonstrate a great deal of variability in router implementers will demonstrate a great deal of variability in
their implementations. To the extent that implementers are able to model their implementations. To the extent that implementers are able to model
their implementations using the abstractions described in this memo, their implementations using the abstractions described in this document,
configuration and management tools will more readily be able to configuration and management tools will more readily be able to
configure and manage networks incorporating Diffserv routers of assorted configure and manage networks incorporating Diffserv routers of assorted
origins. origins.
o Section 3 starts by describing the basic high-level blocks of a o Section 3 starts by describing the basic high-level blocks of a
Diffserv router. It explains the concepts used in the model, Diffserv router. It explains the concepts used in the model,
including the hierarchical management model for these blocks which including the hierarchical management model for these blocks which
uses low-level functional datapath elements such as Classifiers, uses low-level functional datapath elements such as Classifiers,
Actions, Queues. Actions, Queues.
skipping to change at page 3, line 39 skipping to change at page 3, line 43
o Section 7 discusses the basic queueing elements of Algorithmic o Section 7 discusses the basic queueing elements of Algorithmic
Droppers, Queues and Schedulers and their functional behaviors Droppers, Queues and Schedulers and their functional behaviors
(e.g. traffic shaping). (e.g. traffic shaping).
o Section 8 shows how the low-level elements can be combined to build o Section 8 shows how the low-level elements can be combined to build
modules called Traffic Conditioning Blocks (TCBs) which are useful modules called Traffic Conditioning Blocks (TCBs) which are useful
for management purposes. for management purposes.
o Section 9 discusses security concerns. o Section 9 discusses security concerns.
o Appendix A contains a brief discussion of the token bucket and
leaky bucket algorithms used in this model and some of the
practical effects of the use of token buckets within the Diffserv
architecture.
2. Glossary 2. Glossary
This memo uses terminology which is defined in [DSARCH] and in This document uses terminology which is defined in [DSARCH]. There is
[DSTERMS]. Some of the terms defined there are defined again here in also current work-in-progress on this terminology in the IETF and some
order to provide additional detail, along with some new terms specific of the definitions provided here are taken from that work. Some of the
to this document. terms from these other references are defined again here in order to
provide additional detail, along with some new terms specific to this
document.
Absolute A functional datapath element which simply discards all Absolute A functional datapath element which simply discards all
Dropper packets arriving at its input. Dropper packets arriving at its input.
Algorithmic A functional datapath element which selectively discards Algorithmic A functional datapath element which selectively discards
Dropper packets that arrive at its input, based on a discarding Dropper packets that arrive at its input, based on a discarding
algorithm. It has one data input and one output. algorithm. It has one data input and one output.
Classifier A functional datapath element which consists of filters Classifier A functional datapath element which consists of filters
which select matching and non-matching packets and that select matching and non-matching packets. Based
forwards them along a particular datapath within the on this selection, packets are forwarded along the
router. A classifier splits a single incoming traffic appropriate datapath within the router. A classifier,
stream into multiple outgoing ones. therefore, splits a single incoming traffic stream into
multiple outgoing streams.
Counter A functional datapath element which updates a packet Counter A functional datapath element which updates a packet
counter and also an octet counter for every counter and also an octet counter for every
packet that passes through it. Used for collecting packet that passes through it.
statistics.
Datapath A conceptual path taken by packets with particular
characteristics through a Diffserv router. Decisions
as to the path taken by a packet are made by functional
datapath elements such as Classifiers and Meters.
Filter A set of wildcard, prefix, masked, range and/or exact Filter A set of wildcard, prefix, masked, range and/or exact
match conditions on the content of a packet's match conditions on the content of a packet's
headers or other data, and/or on implicit or derived headers or other data, and/or on implicit or derived
attributes associated with the packet. A filter is attributes associated with the packet. A filter is
said to match only if each condition is satisfied. said to match only if each condition is satisfied.
Functional A basic building block of the conceptual router. Functional A basic building block of the conceptual router.
Datapath Typical elements are Classifiers, Meters, Actions, Datapath Typical elements are Classifiers, Meters, Actions,
Element Algorithmic Droppers, Queues and Schedulers. Element Algorithmic Droppers, Queues and Schedulers.
Multiplexer A multiplexor. Multiplexer A multiplexor.
(Mux) (Mux)
Multiplexor A functional datapath element that merges multiple Multiplexor A functional datapath element that merges multiple
(Mux) traffic streams (datapaths) into a single traffic (Mux) traffic streams (datapaths) into a single traffic
stream (datapath). stream (datapath).
Non-work- A property of a scheduling algorithm such that it Non-work- A property of a scheduling algorithm such that it
conserving services packets no sooner than a scheduled departure conserving services packets no sooner than a scheduled departure
time, even if this means leaving packets in a FIFO time, even if this means leaving packets queued
while the link is idle. while the output (e.g. a network link or connection
to the next element) is idle.
Policing The process of comparing the arrival of data packets Policing The process of comparing the arrival of data packets
against a temporal profile and forwarding, delaying against a temporal profile and forwarding, delaying
or dropping them so as to make the output stream or dropping them so as to make the output stream
conformant to the profile. Policing is modelled conformant to the profile.
here as the combination of either a meter or a
scheduler with either an absolute dropper or an
algorithmic dropper.
Queueing A combination of functional datapath elements Queueing A combination of functional datapath elements
Block that modulates the transmission of packets belonging Block that modulates the transmission of packets belonging
to a traffic streams and determines their to a traffic streams and determines their
ordering, possibly storing them temporarily or ordering, possibly storing them temporarily or
discarding them. discarding them.
Scheduling An algorithm which determines which queue of a set Scheduling An algorithm which determines which queue of a set
algorithm of queues to service next. This may be based on the algorithm of queues to service next. This may be based on the
relative priority of the queues, on a weighted fair relative priority of the queues, on a weighted fair
bandwidth sharing policy or some other policy. Such bandwidth sharing policy or some other policy. Such
an algorithm may be either work-conserving or non- an algorithm may be either work-conserving or non-
work-conserving. work-conserving.
Service-Level A set of parameters and their values which together
Specification define the service offered to a traffic stream by a
(SLS) Diffserv domain.
Shaping The process of delaying packets within a traffic stream Shaping The process of delaying packets within a traffic stream
to cause it to conform to some defined temporal profile. to cause it to conform to some defined temporal profile.
Shaping can be implemented using a queue serviced by a Shaping can be implemented using a queue serviced by a
non-work-conserving scheduling algorithm. non-work-conserving scheduling algorithm.
Traffic A logical datapath entity consisting of a number of Traffic A logical datapath entity consisting of a number of
Conditioning functional datapath elements interconnected in Conditioning functional datapath elements interconnected in
Block (TCB) such a way as to perform a specific set of traffic Block (TCB) such a way as to perform a specific set of traffic
conditioning functions on an incoming traffic stream. conditioning functions on an incoming traffic stream.
A TCB can be thought of as an entity with one A TCB can be thought of as an entity with one
input and one or more outputs and a set of control input and one or more outputs and a set of control
parameters. parameters.
Traffic A set of parameters and their values which together
Conditioning specify a set of classfier rules and a traffic profile.
Specification A TCS is an integral element of a SLS.
(TCS)
Work- A property of a scheduling algorithm such that it Work- A property of a scheduling algorithm such that it
conserving services a packet, if one is available, at every conserving services a packet, if one is available, at every
transmission opportunity." transmission opportunity.
3. Conceptual Model 3. Conceptual Model
This section introduces a block diagram of a Diffserv router and This section introduces a block diagram of a Diffserv router and
describes the various components illustrated in Figure 1. Note that a describes the various components illustrated in Figure 1. Note that a
Diffserv core router is assumed to include only a subset of these Diffserv core router is likely to require only a subset of these
components: the model presented here is intended to cover the case of components: the model presented here is intended to cover the case of
both Diffserv edge and core routers. both Diffserv edge and core routers.
3.1. Elements of a Diffserv Router 3.1. Components of a Diffserv Router
The conceptual model includes abstract definitions for the following: The conceptual model includes abstract definitions for the following:
o Traffic Classification elements. o Traffic Classification elements.
o Metering functions. o Metering functions.
o Actions of Marking, Absolute Dropping, Counting and o Actions of Marking, Absolute Dropping, Counting and
Multiplexing. Multiplexing.
skipping to change at page 6, line 19 skipping to change at page 6, line 41
into higher-level blocks known as Traffic Conditioning Blocks into higher-level blocks known as Traffic Conditioning Blocks
(TCBs). (TCBs).
The components and combinations of components described in this document The components and combinations of components described in this document
form building blocks that need to be manageable by Diffserv form building blocks that need to be manageable by Diffserv
configuration and management tools. One of the goals of this document is configuration and management tools. One of the goals of this document is
to show how a model of a Diffserv device can be built using these to show how a model of a Diffserv device can be built using these
component blocks. This model is in the form of a connected directed component blocks. This model is in the form of a connected directed
acyclic graph (DAG) of functional datapath elements that describes the acyclic graph (DAG) of functional datapath elements that describes the
traffic conditioning and queueing behaviors that any particular packet traffic conditioning and queueing behaviors that any particular packet
will experience when forwarded to the Diffserv router. will experience when forwarded to the Diffserv router. Figure 1
illustrates the major functional blocks of a Diffserv router.
Figure 1 illustrates the major functional blocks of a Diffserv router:
3.1.1. Datapath 3.1.1. Datapath
An ingress interface, routing core and egress interface are illustrated An ingress interface, routing core and egress interface are illustrated
at the center of the diagram. In actual router implementations, there at the center of the diagram. In actual router implementations, there
may be an arbitrary number of ingress and egress interfaces may be an arbitrary number of ingress and egress interfaces
interconnected by the routing core. The routing core element serves as interconnected by the routing core. The routing core element serves as
an abstraction of a router's normal routing and switching functionality.
The routing core moves packets between interfaces according to policies
outside the scope of Diffserv. The actual queueing delay and packet loss
behavior of a specific router's switching fabric/backplane is not
modeled by the routing core; these should be modeled using the
functional datapath elements described later. The routing core should be
thought of as an infinite bandwidth, zero- delay backplane connecting
ingress and egress interfaces.
The components of interest on the ingress/egress interfaces are the
functional datapath elements (e.g. classifiers, queueing elements) that
support Diffserv traffic conditioning and per-hop behaviors [DSARCH].
These are the fundamental components comprising a Diffserv router and
will be the focal point of our conceptual model.
3.1.2. Configuration and Management Interface
Diffserv operating parameters are monitored and provisioned through this
interface. Monitored parameters include statistics regarding traffic
carried at various Diffserv service levels. These statistics may be
important for accounting purposes and/or for tracking compliance to
Traffic Conditioning Specifications (TCSs) [DSTERMS] negotiated with
+---------------+ +---------------+
| Diffserv | | Diffserv |
Mgmt | configuration | Mgmt | configuration |
<----+-->| & management |------------------+ <----+-->| & management |------------------+
SNMP,| | interface | | SNMP,| | interface | |
COPS | +---------------+ | COPS | +---------------+ |
etc. | | | etc. | | |
| | | | | |
| v v | v v
| +-------------+ +-------------+ | +-------------+ +-------------+
skipping to change at page 7, line 31 skipping to change at page 7, line 31
| | | | | |
| | | | | |
| +------------+ | | +------------+ |
+-->| QOS agent | | +-->| QOS agent | |
-------->| (optional) |---------------------+ -------->| (optional) |---------------------+
QOS | (e.g. RSVP)| QOS | (e.g. RSVP)|
cntl +------------+ cntl +------------+
msgs msgs
Figure 1: Diffserv Router Major Functional Blocks Figure 1: Diffserv Router Major Functional Blocks
customers. Provisioned parameters are primarily classification rules, TC an abstraction of a router's normal routing and switching functionality.
and PHB configuration parameters. The network administrator interacts The routing core moves packets between interfaces according to policies
with the Diffserv configuration and management interface via one or more outside the scope of Diffserv (note: it is possible that such policies
for output-interface selection might involve use of packet fields such
as the DSCP but this is outside the scope of this model). The actual
queueing delay and packet loss behavior of a specific router's switching
fabric/backplane is not modeled by the routing core; these should be
modeled using the functional datapath elements described later. The
routing core of this model can be thought of as an infinite bandwidth,
zero-delay backplane connecting interfaces - properties like the
behaviour of the core when overloaded need to be reflected back into the
queueing elements that are modelled around it e.g. when too much traffic
is directed across the core at an egress interface, the excess must
either be dropped or queued somewhere: the elements performing these
functions must be modelled on one of the interfaces involved.
The components of interest at the ingress to and egress from interfaces
are the functional datapath elements (e.g. Classifiers, Queueing
elements) that support Diffserv traffic conditioning and per-hop
behaviors [DSARCH]. These are the fundamental components comprising a
Diffserv router and are the focal point of this model.
3.1.2. Configuration and Management Interface
Diffserv operating parameters are monitored and provisioned through this
interface. Monitored parameters include statistics regarding traffic
carried at various Diffserv service levels. These statistics may be
important for accounting purposes and/or for tracking compliance to
Traffic Conditioning Specifications (TCSs) negotiated with customers.
Provisioned parameters are primarily the TCS parameters for Classifiers
and Meters and the associated PHB configuration parameters for Actions
and Queueing elements. The network administrator interacts with the
Diffserv configuration and management interface via one or more
management protocols, such as SNMP or COPS, or through other router management protocols, such as SNMP or COPS, or through other router
configuration tools such as serial terminal or telnet consoles. configuration tools such as serial terminal or telnet consoles.
Specific policy rules and goals governing the Diffserv behaviour of a Specific policy rules and goals governing the Diffserv behaviour of a
router are presumed to be installed by policy management mechanisms. router are presumed to be installed by policy management mechanisms.
However, Diffserv routers are subject to implementation decisions which However, Diffserv routers are always subject to implementation limits
form a meta- policy that scopes the kinds of policies which can be which scope the kinds of policies which can be successfully implemented
successfully implemented by the router. External reporting of such by the router. External reporting of such implementation capabilities is
implementation capabilities are considered out of scope for this considered out of scope for this document.
document.
3.1.3. Optional QoS Agent Module 3.1.3. Optional QoS Agent Module
Diffserv routers may snoop or participate in either per-microflow or Diffserv routers may snoop or participate in either per-microflow or
per-flow-aggregate signaling of QoS requirements [E2E] e.g. using the per-flow-aggregate signaling of QoS requirements [E2E] e.g. using the
RSVP protocol. Snooping of RSVP messages may be used, for example, to RSVP protocol. Snooping of RSVP messages may be used, for example, to
learn how to classify traffic without actually participating as a RSVP learn how to classify traffic without actually participating as a RSVP
protocol peer. Diffserv routers may reject or admit RSVP reservation protocol peer. Diffserv routers may reject or admit RSVP reservation
requests to provide a means of admission control to Diffserv-based requests to provide a means of admission control to Diffserv-based
services or they may use these requests to trigger provisioning changes services or they may use these requests to trigger provisioning changes
for a flow-aggregation in the Diffserv network. A flow-aggregation in for a flow-aggregation in the Diffserv network. A flow-aggregation in
this context might be equivalent to a Diffserv BA or it may be more this context might be equivalent to a Diffserv BA or it may be more
fine-grained, relying on a MF classifier [DSARCH]. Note that the fine-grained, relying on a MF classifier [DSARCH]. Note that the
conceptual model of such a router implements the Integrated Services conceptual model of such a router implements the Integrated Services
Model as described in [INTSERV], applying the control plane controls to Model as described in [INTSERV], applying the control plane controls to
the data classified and conditioned in the data plane, as desribed in the data classified and conditioned in the data plane, as desribed in
skipping to change at page 8, line 23 skipping to change at page 9, line 4
Model as described in [INTSERV], applying the control plane controls to Model as described in [INTSERV], applying the control plane controls to
the data classified and conditioned in the data plane, as desribed in the data classified and conditioned in the data plane, as desribed in
[E2E]. [E2E].
Note that a QoS Agent component of a Diffserv router, if present, might Note that a QoS Agent component of a Diffserv router, if present, might
be active only in the control plane and not in the data plane. In this be active only in the control plane and not in the data plane. In this
scenario, RSVP could be used merely to signal reservation state without scenario, RSVP could be used merely to signal reservation state without
installing any actual reservations in the data plane of the Diffserv installing any actual reservations in the data plane of the Diffserv
router: the data plane could still act purely on Diffserv DSCPs and router: the data plane could still act purely on Diffserv DSCPs and
provide PHBs for handling data traffic without the normal per-microflow provide PHBs for handling data traffic without the normal per-microflow
handling expected to support some Intserv services. handling expected to support some Intserv services.
3.2. Hierarchical Model of Diffserv Components 3.2. Diffserv Functions at Ingress and Egress
This document focuses on the Diffserv-specific components of the router. This document focuses on the Diffserv-specific components of the router.
Figure 2 shows a high-level view of ingress and egress interfaces of a Figure 2 shows a high-level view of ingress and egress interfaces of a
router. The diagram illustrates two Diffserv router interfaces, each router. The diagram illustrates two Diffserv router interfaces, each
having an ingress and an egress component. It shows classification, having a set of ingress and a set of egress elements. It shows
metering, action and queueing functions which might be instantiated on classification, metering, action and queueing functions which might be
each interface's ingress and egress component. instantiated at each interface's ingress and egress.
In principle, if one were to construct a network entirely out of two- In principle, if one were to construct a network entirely out of two-
port routers (in appropriate places connected by LANs or similar media), port routers (connected by LANs or similar media), then it might be
then it would be necessary for each router to perform four QoS control necessary for each router to perform four QoS control functions in the
functions in the datapath on traffic in each direction: datapath on traffic in each direction:
- Classify each message according to some set of rules, possibly just - Classify each message according to some set of rules, possibly just
a "match everything" rule. a "match everything" rule.
- If necessary, determine whether the data stream the message is part - If necessary, determine whether the data stream the message is part
of is within or outside its rate by metering the stream. of is within or outside its rate by metering the stream.
- Perform a set of resulting actions, including applying a drop - Perform a set of resulting actions, including applying a drop
policy appropriate to the classification and queue in question and policy appropriate to the classification and queue in question and
perhaps additionally marking the traffic with a Differentiated perhaps additionally marking the traffic with a Differentiated
Services Code Point (DSCP) as defined in [DSCP]. Services Code Point (DSCP) [DSFIELD].
Interface A Interface B Interface A Interface B
+-------------+ +---------+ +-------------+ +-------------+ +---------+ +-------------+
| ingress i/f | | | | egress i/f | | ingress: | | | | egress: |
| classify, | | | | classify, | | classify, | | | | classify, |
--->| meter, |---->| |---->| meter, |---> --->| meter, |---->| |---->| meter, |--->
| action, | | | | action, | | action, | | | | action, |
| queueing | | | | queueing | | queueing | | routing | | queueing |
+-------------+ | routing | +-------------+ +-------------+ | core | +-------------+
| core | | egress: | | | | ingress: |
+-------------+ | | +-------------+
| egress i/f | | | | ingress i/f |
| classify, | | | | classify, | | classify, | | | | classify, |
<---| meter, |<----| |<----| meter, |<--- <---| meter, |<----| |<----| meter, |<---
| action, | | | | action, | | action, | | | | action, |
| queueing | +---------+ | queueing | | queueing | +---------+ | queueing |
+-------------+ +-------------+ +-------------+ +-------------+
Figure 2. Traffic Conditioning and Queueing Elements Figure 2. Traffic Conditioning and Queueing Elements
- Enqueue the traffic for output in the appropriate queue, which may - Enqueue the traffic for output in the appropriate queue. The
either shape the traffic or simply forward it with some minimum scheduling of output from this queue may lead to shaping of the
rate or maximum latency. traffic or may simply cause it to be forwarded with some minimum
rate or maximum latency assurance.
If the network is now built out of N-port routers, the expected behavior If the network is now built out of N-port routers, the expected behavior
of the network should be identical. Therefore, this model must provide of the network should be identical. Therefore, this model must provide
for essentially the same set of functions on the ingress as on the for essentially the same set of functions at the ingress as on the
egress port of the router. Some interfaces will be "edge" interfaces and egress of a router's interfaces. The one point of difference in the
some will be "interior" to the Differentiated Services domain. The one model between ingress and the egress is that all traffic at the egress
point of difference between an ingress and an egress interface is that of an interface is queued, while traffic at the ingress to an interface
all traffic on an egress interface is queued, while traffic on an is likely to be queued only for shaping purposes, if at all. Therefore,
ingress interface will typically be queued only for shaping purposes, if equivalent functional datapath elements may be modelled at both the
at all. Therefore, equivalent functional datapath elements may be ingress to and egress from an interface.
modelled on both the ingress and egress components of an interface.
Note that it is not mandatory that each of these functional datapath Note that it is not mandatory that each of these functional datapath
elements be implemented on both ingress and egress components; equally, elements be implemented at both ingress and egress; equally, the model
the model allows that multiple sets of these elements may be placed in allows that multiple sets of these elements may be placed in series
series and/or in parallel at ingress or at egress. The arrangement of and/or in parallel at ingress or at egress. The arrangement of elements
elements is dependent on the service requirements on a particular is dependent on the service requirements on a particular interface on a
interface on a particular router. By modelling these elements on both particular router. By modelling these elements at both ingress and
ingress and egress components, it is not implied that they must be egress, it is not implied that they must be implemented in this way in a
implemented in this way in a specific router. For example, a router may specific router. For example, a router may implement all shaping and PHB
implement all shaping and PHB queueing on the interface egress component queueing at the interface egress or may instead implement it only at the
or may instead implement it only on the ingress component. Furthermore, ingress. Furthermore, the classification needed to map a packet to an
the classification needed to map a packet to an egress component queue egress queue (if present) need not be implemented at the egress but
(if present) need not be implemented on the egress component but instead instead might be implemented at the ingress, with the packet passed
may be implemented on the ingress component, with the packet passed
through the routing core with in-band control information to allow for through the routing core with in-band control information to allow for
egress queue selection. egress queue selection.
Specifically, some interfaces will be at the outer "edge" and some will
be towards the "core" of the Diffserv domain. It is to be expected (from
the general principles guiding the motivation of Diffserv) that "edge"
interfaces, or at least the routers that contain them, will contain more
complexity and require more configuration than those in the core.
3.3. Shaping and Policing
Diffserv nodes may apply shaping, policing and/or marking to traffic
streams that exceed the bounds of their TCS in order to prevent one
traffic stream from seizing more than its share of resources from a
Diffserv network. In this model, Shaping, sometimes considered as a TC
action, is treated as a function of queueing elements - see section 7.
Algorithmic Dropping techniques (e.g. RED) are similarly treated since
these often are closely associated with queues. Policing is modelled as
either a concatenation of a Meter with an Absolute Dropper or as a
concatenation of an Algorithmic Dropper with a Scheduler. These elements
will discard packets which exceed the TCS.
3.4. Hierarchical View of the Model
>From a device-level configuration management perspective, the following >From a device-level configuration management perspective, the following
hierarchy exists: hierarchy exists:
At the lowest level considered here, are individual functional At the lowest level considered here, are individual functional
datapath elements, each with their own configuration parameters and datapath elements, each with their own configuration parameters and
management counters and flags. management counters and flags.
At the next level, the network administrator manages groupings of At the next level, the network administrator manages groupings of
these functional datapath elements interconnected in a DAG. These these functional datapath elements interconnected in a DAG. These
functional datapath elements are organized in self-contained TCBs functional datapath elements are organized in self-contained TCBs
which are used to implement some desired network policy (see which are used to implement some desired network policy (see
Section 8). One or more TCBs may be instantiated on each ingress or Section 8). One or more TCBs may be instantiated at each
egress component; they may be connected in series and/or in interface's ingress or egress; they may be connected in series
parallel configurations on the multiple outputs of a preceding TCB. and/or in parallel configurations on the multiple outputs of a
A TCB can be thought of as a "black box" with one input and one or preceding TCB. A TCB can be thought of as a "black box" with one
more outputs (in the data path). Each interface (ingress or egress) input and one or more outputs (in the data path). Each interface
may have a different TCB configuration. may have a different TCB configuration and each direction (ingress
or egress) may too.
At the topmost level considered here, the network administrator At the topmost level considered here, the network administrator
manages interfaces. Each interface consists of an ingress component manages interfaces. Each interface has ingress and egress
and an egress component. Each component may contain one or more functionality, with each of these expressed as one or more TCBs.
TCBs. This level of the hierarchy is what was illustrated in Figure 2.
Further levels may be built on top of this hierarchy, in particular ones
for aiding in the repetitive configuration tasks likely for routers with
many interfaces: some such "template" tools for Diffserv routers are
outside the scope of this model but are under study by other working
groups within IETF.
4. Classifiers 4. Classifiers
4.1. Definition 4.1. Definition
Classification is performed by a classifier element. Classifiers are 1:N Classification is performed by a classifier element. Classifiers are 1:N
(fan-out) devices: they take a single traffic stream as input and (fan-out) devices: they take a single traffic stream as input and
generate N logically separate traffic streams as output. Classifiers are generate N logically separate traffic streams as output. Classifiers are
parameterized by filters and output streams. Packets from the input parameterized by filters and output streams. Packets from the input
stream are sorted into various output streams by filters which match the stream are sorted into various output streams by filters which match the
skipping to change at page 10, line 45 skipping to change at page 12, line 4
Classification is performed by a classifier element. Classifiers are 1:N Classification is performed by a classifier element. Classifiers are 1:N
(fan-out) devices: they take a single traffic stream as input and (fan-out) devices: they take a single traffic stream as input and
generate N logically separate traffic streams as output. Classifiers are generate N logically separate traffic streams as output. Classifiers are
parameterized by filters and output streams. Packets from the input parameterized by filters and output streams. Packets from the input
stream are sorted into various output streams by filters which match the stream are sorted into various output streams by filters which match the
contents of the packet or possibly match other attributes associated contents of the packet or possibly match other attributes associated
with the packet. Various types of classifiers using different filters with the packet. Various types of classifiers using different filters
are described in the following sections. Figure 3 illustrates a are described in the following sections. Figure 3 illustrates a
classifier, where the outputs connect to succeeding functional datapath classifier, where the outputs connect to succeeding functional datapath
elements. elements.
The simplest possible Classifier element is one that matches all packets The simplest possible Classifier element is one that matches all packets
that are applied at its input. In this case, the Classifier element is that are applied at its input. In this case, the Classifier element is
just a no-op and may be omitted. just a no-op and may be omitted.
unclassified classified Note that we allow a Multiplexor (see Section 6.5) before the Classifier
traffic traffic to allow input from multiple traffic streams. For example, if traffic
+------------+ streams originating from multiple ingress interfaces feed through a
| |--> match Filter1 --> OutputA single Classifier then the interface number could be one of the packet
------->| classifier |--> match Filter2 --> OutputB classification keys used by the Classifier. This optimization may be
| |--> no match --> OutputC important for scalability in the management plane. Classifiers may also
+------------+ be cascaded in sequence to perform more complex lookup operations whilst
still maintaining such scalability.
Figure 3. An Example Classifier
Note that we allow a multiplexor (see Section 6.5) before the classifier
to allow input from multiple traffic streams. For example, if multiple
ingress sub-interfaces feed through a single classifier then the
interface number can be considered by the classifier as a packet
attribute and be included in the packet's classification key. This
optimization may be important for scalability in the management plane.
Another example of a packet attribute could be an integer representing Another example of a packet attribute could be an integer representing
the BGP community string associated with the packet's best-matching the BGP community string associated with the packet's best-matching
route. route. Other contextual information may also be used by a Classifier
e.g. knowledge that a particular interface faces a Diffserv domain or a
legacy IPTOS domain [DSARCH] could be used when determining whether a
DSCP is present or not.
The following classifier separates traffic into one of three output The following classifier separates traffic into one of three output
streams based on three filters: streams based on three filters:
Filter Matched Output Stream Filter Matched Output Stream
-------------- --------------- -------------- ---------------
Filter1 A Filter1 A
Filter2 B Filter2 B
Filter3 (no match) C no match C
Where Filters1 and Filter2 are defined to be the following BA filters Where Filters1 and Filter2 are defined to be the following BA filters
([DSARCH], Section 4.2.1 ): ([DSARCH], Section 4.2.1 ):
unclassified classified
traffic traffic
+------------+
| |--> match Filter1 --> OutputA
------->| classifier |--> match Filter2 --> OutputB
| |--> no match --> OutputC
+------------+
Figure 3. An Example Classifier
Filter DSCP Filter DSCP
------ ------ ------ ------
1 101010 1 101010
2 111111 2 111111
3 ****** (wildcard) 3 ****** (wildcard)
4.1.1. Filters 4.1.1. Filters
A filter consists of a set of conditions on the component values of a A filter consists of a set of conditions on the component values of a
packet's classification key (the header values, contents, and attributes packet's classification key (the header values, contents, and attributes
skipping to change at page 12, line 19 skipping to change at page 13, line 31
In general there are a set of possible component conditions including In general there are a set of possible component conditions including
exact, prefix, range, masked and wildcard matches. Note that ranges can exact, prefix, range, masked and wildcard matches. Note that ranges can
be represented (with less efficiency) as a set of prefixes and that be represented (with less efficiency) as a set of prefixes and that
prefix matches are just a special case of both masked and range matches. prefix matches are just a special case of both masked and range matches.
In the case of a MF classifier [DSARCH], the classification key consists In the case of a MF classifier [DSARCH], the classification key consists
of a number of packet header fields. The filter may specify a different of a number of packet header fields. The filter may specify a different
condition for each key component, as illustrated in the example below condition for each key component, as illustrated in the example below
for a IPv4/TCP classifier: for a IPv4/TCP classifier:
Filter IP Src Addr IP Dest Addr TCP SrcPort TCP DestPort
------ ------------- ------------- ----------- ------------
Filter4 172.31.8.1/32 172.31.3.X/24 X 5003
In this example, the fourth octet of the destination IPv4 address and In this example, the fourth octet of the destination IPv4 address and
the source TCP port are wildcard or "don't cares". the source TCP port are wildcard or "don't care".
MF classification of fragmented packets is impossible if the filter uses MF classification of fragmented packets is impossible if the filter uses
transport-layer port numbers e.g. TCP port numbers. MTU-discovery is transport-layer port numbers e.g. TCP port numbers. MTU-discovery is
therefore a prerequisite for proper operation of a Diffserv network that therefore a prerequisite for proper operation of a Diffserv network that
uses such classifiers. uses such classifiers.
4.1.2. Overlapping Filters 4.1.2. Overlapping Filters
Note that it is easy to define sets of overlapping filters in a Note that it is easy to define sets of overlapping filters in a
classifier. For example: classifier. For example:
Filter IP Src Addr IP Dest Addr TCP SrcPort TCP DestPort
------ ------------- ------------- ----------- ------------
Filter4 172.31.8.1/32 172.31.3.X/24 X 5003
Filter5: Filter5:
Type: Masked-DSCP Type: Masked-DSCP
Value: 111000 Value: 111000
Mask: 111000 Mask: 111000
Filter6: Filter6:
Type: Masked-DSCP Type: Masked-DSCP
Value: 000111 (binary) Value: 000111 (binary)
Mask: 000111 (binary) Mask: 000111 (binary)
skipping to change at page 15, line 37 skipping to change at page 17, line 8
Filter14: -- priority 4 or 5 Filter14: -- priority 4 or 5
Type: Ieee8021pPriority Type: Ieee8021pPriority
Value: 100 (binary) Value: 100 (binary)
Mask: 110 (binary) Mask: 110 (binary)
Filter15: -- VLAN 2304 Filter15: -- VLAN 2304
Type: Ieee8021QVlan Type: Ieee8021QVlan
Value: 100100000000 (binary) Value: 100100000000 (binary)
Mask: 111111111111 (binary) Mask: 111111111111 (binary)
Such classifiers may be subject of other standards or may be enterprise- Such classifiers may be the subject of other standards or may be
specific but are not discussed further here. proprietary to a router vendor but they are not discussed further here.
5. Meters 5. Meters
Metering is defined in [DSARCH]. Diffserv network providers may choose Metering is defined in [DSARCH]. Diffserv network providers may choose
to offer services to customers based on a temporal (i.e., rate) profile to offer services to customers based on a temporal (i.e., rate) profile
within which the customer submits traffic for the service. In this within which the customer submits traffic for the service. In this
event, a meter might be used to trigger real-time traffic conditioning event, a meter might be used to trigger real-time traffic conditioning
actions (e.g., marking) by routing a non-conforming packet through an actions (e.g., marking) by routing a non-conforming packet through an
appropriate next-stage action element. Alternatively, by counting appropriate next-stage action element. Alternatively, by counting
conforming and/or non-conforming traffic, it might also be used for conforming and/or non-conforming traffic using a Counter element
collecting data for out-of-band management functions such as billing downstream of the Meter, it might also be used to help in collecting
applications. data for out-of-band management functions such as billing applications.
Meters are logically 1:N (fan-out) devices (although a multiplexor can Meters are logically 1:N (fan-out) devices (although a multiplexor can
be used in front of a meter). Meters are parameterized by a temporal be used in front of a meter). Meters are parameterized by a temporal
profile and by conformance levels, each of which is associated with a profile and by conformance levels, each of which is associated with a
meter's output. Each output can be connected to another functional meter's output. Each output can be connected to another functional
element. element.
Note that this model of a meter differs slightly from that described in Note that this model of a meter differs slightly from that described in
[DSARCH]. In that description the meter is not a datapath element but is [DSARCH]. In that description the meter is not a datapath element but is
instead used to monitor the traffic stream and send control signals to instead used to monitor the traffic stream and send control signals to
action elements to dynamically modulate their behavior based on the action elements to dynamically modulate their behavior based on the
conformance of the packet. Figure 4 illustrates a meter with 3 levels of conformance of the packet. Figure 4 illustrates a meter with 3 levels of
conformance. conformance.
In some Diffserv examples, three levels of conformance are discussed in
terms of colors, with green representing conforming, yellow representing
partially conforming and red representing non-conforming [AF-PHB]. These
different conformance levels may be used to trigger different queueing,
marking or dropping treatment later on in the processing. Other example
meters use a binary notion of conformance; in the general case N levels
of conformance can be supported. In general there is no constraint on
the type of functional datapath element following a meter output, but
care must be taken not to inadvertently configure a datapath that
results in packet reordering that is not consistent with PHB
requirements.
A meter, according to this model, measures the rate at which packets
making up a stream of traffic pass it, compares the rate to some set of
thresholds and produces some number (two or more) potential results: a
given packet is said to "conform" to the meter if, at the time that the
packet is being looked at, the stream appears to be within the meter's
limit rate.
unmetered metered unmetered metered
traffic traffic traffic traffic
+---------+ +---------+
| |--------> conformance A | |--------> conformance A
--------->| meter |--------> conformance B --------->| meter |--------> conformance B
| |--------> conformance C | |--------> conformance C
+---------+ +---------+
Figure 4. A Generic Meter Figure 4. A Generic Meter
5.1. Token-Bucket Model In some Diffserv examples e.g. [AF-PHB], three levels of conformance are
discussed in terms of colors, with green representing conforming, yellow
The concept of conformance to a meter bears comment. The concept applied representing partially conforming and red representing non-conforming.
in several rate-control architectures, including ATM, Frame Relay, These different conformance levels may be used to trigger different
Integrated Services and Differentiated Services, is variously described queueing, marking or dropping treatment later on in the processing.
as a "leaky bucket" or a "token bucket". Both token buckets and leaky Other example meters use a binary notion of conformance; in the general
buckets are, by definition, theoretical relationships between some case N levels of conformance can be supported. In general there is no
defined burst_size, rate and interval: constraint on the type of functional datapath element following a meter
output, but care must be taken not to inadvertently configure a datapath
rate = burst_size/interval that results in packet reordering that is not consistent with the
requirements of the relevant PHB specification.
Thus, a token bucket or leaky bucket might specify an information rate
of 1.2 Mbps with a burst size of 1500 bytes. In this case, the token
rate is 1,200,000 bits per second, the token burst is 12,000 bits and
the token interval is 10 milliseconds. The specification says that
conforming traffic will in the worst case come in 100 bursts per second
of 1500 bytes and at a rate not to exceed 1.2 Mbps.
A leaky bucket algorithm is primarily used for traffic shaping (handled
under Queues and Schedulers in this model). Traffic theoretically
departs from a device at a rate of one bit every so many time units (in
the example, one bit every 0.83 microseconds) but, in fact, departs in
multi-bit units (packets) at a rate approximating that. In the example,
it might send one 1500 byte packet every 10 ms, or one 500 byte packet
every 3.3 ms. It is also possible to build multi-rate leaky buckets, in
which traffic departs from the switch at varying rates depending on
recent activity or inactivity.
Implementations generally seek as constant a transmission rate as
achievable. In theory, a 10 Mbps shaped transmission stream from an
algorithmic implementation and a stream which is running at 10 Mbps
because its bottleneck link has been a 10 Mbps Ethernet link should be
indistinguishable. Depending on configuration, the approximation to
theoretical smoothness may vary by moving as much as an MTU from one
token interval to another. Traffic may also be jostled by traffic
competing for the same transmission resources.
A token bucket measures the arrival rate of traffic from another system,
which may have originally been shaped using a leaky bucket shaper or its
equivalent, and determines whether it (still) conforms to the
specification. Multi-rate token buckets (token buckets with both a peak
and a mean rate, and sometimes more rates) are commonly used, such as
described in [SRTCM] and [TRTCM]. In this case, absolute smoothness is
not expected, but conformance to one or more of the specified rates is
expected.
Simplistically, a data stream is said to conform to a simple token
bucket parameterised by a {rate, burst_size} if the system receives in
any time interval, t, at most, an amount of data not exceeding (rate *
t) + burst_size.
For the multi-rate token bucket case, the data stream is said to conform
if, for each of the rates, the stream conforms to the token-bucket
profile appropriate for traffic of that class. For example, received
traffic that arrives pre-classified as one of the "excess" rates (e.g.
AF12 or AF13 traffic for a device implementing the AF1x PHB) is only
compared to the relevant "excess" token bucket profile.
When used as a leaky bucket shaper, this definition interacts with clock
granularity in ways one might not expect. A leaky bucket releases a
packet only when all of its bits would have been allowed: it does not
borrow from future capacity. If the clock is very fine grain, on the
order of the bit rate or faster, this is not an issue. But if the clock
is relatively slow (and millisecond or multi-millisecond clocks are not
unusual in networking equipment), this can introduce jitter to the
shaped stream.
The fact that data is organized into variable length packets introduces A meter, according to this model, measures the rate at which packets
some uncertainty in this conformance decision. When used in a Scheduler, making up a stream of traffic pass it, compares the rate to some set of
a leaky bucket releases a packet only when all of its bits would have thresholds and produces some number of potential results (two or more):
been allowed: it does not borrow from future capacity. Ideally, when a given packet is said to be "conformant" to a level of the meter if, at
used in a Meter, a token bucket accepts a packet only if all of its bits the time that the packet is being examined, the stream appears to be
would have been accepted and does not borrow excess capacity required within the rate limit for the profile associated with that level. A
from future capacity. This is consistent with [SRTCM] and [TRTCM]. In fuller discussion of conformance to meter profiles (and the associated
real-world deployment, where MTUs are often larger than the burst size requirements that this places on the schedulers upstream) is provided in
offered by a link-layer network service provider and TCP is more Appendix A.
commonly ACK-paced than shaped using a leaky bucket, the loose model
offers a solution to the problems that arise. For a more detailed look
at the practical issues, see Appendix A.
5.2. Examples 5.1. Examples
The following are some examples of possible meters. The following are some examples of possible meters.
5.2.1. Average Rate Meter 5.1.1. Average Rate Meter
An example of a very simple meter is an average rate meter. This type of An example of a very simple meter is an average rate meter. This type of
meter measures the average rate at which packets are submitted to it meter measures the average rate at which packets are submitted to it
over a specified averaging time. over a specified averaging time.
An average rate profile may take the following form: An average rate profile may take the following form:
Meter1: Meter1:
Type: AverageRate Type: AverageRate
Profile: Profile1 Profile: Profile1
ConformingOutput: Queue1 ConformingOutput: Queue1
NonConformingOutput: Counter1 NonConformingOutput: Counter1
Profile1: Profile1:
Type: AverageRate Type: AverageRate
AverageRate: 120 kbps AverageRate: 120 kbps
Delta: 100 msec Delta: 100 msec
skipping to change at page 19, line 15 skipping to change at page 18, line 50
Type: AverageRate Type: AverageRate
Profile: Profile1 Profile: Profile1
ConformingOutput: Queue1 ConformingOutput: Queue1
NonConformingOutput: Counter1 NonConformingOutput: Counter1
Profile1: Profile1:
Type: AverageRate Type: AverageRate
AverageRate: 120 kbps AverageRate: 120 kbps
Delta: 100 msec Delta: 100 msec
A meter measuring against this profile would continually maintain a A Meter measuring against this profile would continually maintain a
count that indicates the total number of packets arriving between time T count that indicates the total number and/or cumulative byte-count of
(now) and time T - 100 msecs. So long as an arriving packet does not
push the count over 12 kbits in the last 100 msec then the packet would
be deemed conforming. Any packet that pushes the count over 12 kbits
would be deemed non-conforming. Thus, this meter deems packets to
correspond to one of two conformance levels: conforming or non-
conforming and sends them on for the appropriate subsequent treatment.
5.2.2. Exponential Weighted Moving Average (EWMA) Meter packets arriving between time T (now) and time T - 100 msecs. So long as
an arriving packet does not push the count over 12 kbits in the last 100
msec then the packet would be deemed conforming. Any packet that pushes
the count over 12 kbits would be deemed non-conforming. Thus, this Meter
deems packets to correspond to one of two conformance levels: conforming
or non-conforming and sends them on for the appropriate subsequent
treatment.
The EWMA form of meter is easy to implement in hardware and can be 5.1.2. Exponential Weighted Moving Average (EWMA) Meter
The EWMA form of Meter is easy to implement in hardware and can be
parameterized as follows: parameterized as follows:
avg_rate(t) = (1 - Gain) * avg_rate(t') + Gain * rate(t) avg_rate(t) = (1 - Gain) * avg_rate(t') + Gain * rate(t)
t = t' + Delta t = t' + Delta
For a packet arriving at time t: For a packet arriving at time t:
if (avg_rate(t) > AverageRate) if (avg_rate(t) > AverageRate)
non-conforming non-conforming
else else
conforming conforming
"Gain" controls the time constant (e.g. frequency response) of what is "Gain" controls the time constant (e.g. frequency response) of what is
essentially a simple IIR low-pass filter. "rate(t)" measures the number essentially a simple IIR low-pass filter. "rate(t)" measures the number
of incoming bytes in a small fixed sampling interval, Delta. Any packet of incoming bytes in a small fixed sampling interval, Delta. Any packet
that arrives and pushes the average rate over a predefined rate that arrives and pushes the average rate over a predefined rate
AverageRate is deemed non-conforming. An EWMA meter profile might look AverageRate is deemed non-conforming. An EWMA Meter profile might look
something like the following: something like the following:
Meter2: Meter2:
Type: ExpWeightedMovingAvg Type: ExpWeightedMovingAvg
Profile: Profile2 Profile: Profile2
ConformingOutput: Queue1 ConformingOutput: Queue1
NonConformingOutput: AbsoluteDropper1 NonConformingOutput: AbsoluteDropper1
Profile2: Profile2:
Type: ExpWeightedMovingAvg Type: ExpWeightedMovingAvg
AverageRate: 25 kbps AverageRate: 25 kbps
Delta: 10 usec Delta: 10 usec
Gain: 1/16 Gain: 1/16
5.2.3. Two-Parameter Token Bucket Meter 5.1.3. Two-Parameter Token Bucket Meter
A more sophisticated meter might measure conformance to a token bucket A more sophisticated Meter might measure loose conformance to a token
(TB) profile. A TB profile generally has two parameters, an average bucket (TB) profile (see above and Appendix A for discussions of loose
token rate and a burst size. TB meters compare the arrival rate of and strict conformance to a token bucket). A TB profile generally has
packets to the average rate specified by the TB profile. Logically,
tokens accumulate in a bucket at the average rate, up to a maximum two parameters, an average token rate and a burst size. TB Meters
credit which is the burst size. Packets of length L bytes are considered compare the arrival rate of packets to the average rate specified by the
conforming if any tokens are available in the bucket at the time of TB profile. Logically, tokens accumulate in a bucket at the average
packet arrival: up to L bytes may then be borrowed from future token rate, up to a maximum credit which is the burst size. Packets of length
allocations. Packets are allowed to exceed the average rate in bursts up L bytes are considered conforming if any tokens are available in the
to the burst size. Packets which arrive to find a bucket with no tokens bucket at the time of packet arrival: up to L bytes may then be borrowed
in it are deemed non-conforming. A two-parameter TB meter has exactly from future token allocations. Packets are allowed to exceed the average
two possible conformance levels (conforming, non-conforming). TB rate in bursts up to the burst size. Packets which arrive to find a
implementation details are discussed in Appendix A. Note that this is a bucket with no tokens in it are deemed non-conforming. A two-parameter
"lenient" meter that allows some borrowing, as discussed above. TB meter has exactly two possible conformance levels (conforming, non-
conforming). Note that "strict" conformance meters are also useful -
see e.g. [SRTCM] and [TRTCM].
A two-parameter TB meter might appear as follows: A two-parameter TB meter might appear as follows:
Meter3: Meter3:
Type: SimpleTokenBucket Type: SimpleTokenBucket
Profile: Profile3 Profile: Profile3
ConformingOutput: Queue1 ConformingOutput: Queue1
NonConformingOutput: AbsoluteDropper1 NonConformingOutput: AbsoluteDropper1
Profile3: Profile3:
Type: SimpleTokenBucket Type: SimpleTokenBucket
AverageRate: 200 kbps AverageRate: 200 kbps
BurstSize: 100 kbytes BurstSize: 100 kbytes
ConformanceType: loose
5.2.4. Multi-Stage Token Bucket Meter 5.1.4. Multi-Stage Token Bucket Meter
More complicated TB meters might define two burst sizes and three More complicated TB meters might define multiple burst sizes and more
conformance levels. Packets found to exceed the larger burst size are conformance levels. Packets found to exceed the larger burst size are
deemed non-conforming. Packets found to exceed the smaller burst size deemed non-conforming. Packets found to exceed the smaller burst size
are deemed partially conforming. Packets exceeding neither are deemed are deemed partially conforming. Packets exceeding neither are deemed
conforming. Token bucket meters designed for Diffserv networks are conforming. Token bucket meters designed for Diffserv networks are
described in more detail in [SRTCM, TRTCM, GTC]; in some of these described in more detail in [SRTCM, TRTCM, GTC]; in some of these
references three levels of conformance are discussed in terms of colors, references, three levels of conformance are discussed in terms of colors
with green representing conforming, yellow representing partially with green representing conforming, yellow representing partially
conforming and red representing non- conforming. Often these multi- conforming and red representing non-conforming. Note that these
multiple-conformance-level meters can sometimes be implemented using an
conformance level meters can be implemented using an appropriate appropriate sequence of multiple two-parameter TB meters.
configuration of multiple two- parameter TB meters.
A profile for a multi-stage TB meter with three levels of conformance A profile for a multi-stage TB meter with three levels of conformance
might look as follows: might look as follows:
Meter4: Meter4:
Type: TwoRateTokenBucket Type: TwoRateTokenBucket
ProfileA: Profile4 ProfileA: Profile4
ConformingOutputA: Queue1 ConformingOutputA: Queue1
ProfileB: Profile5 ProfileB: Profile5
ConformingOutputB: Marker1 ConformingOutputB: Marker1
skipping to change at page 21, line 29 skipping to change at page 21, line 19
Profile4: Profile4:
Type: SimpleTokenBucket Type: SimpleTokenBucket
AverageRate: 100 kbps AverageRate: 100 kbps
BurstSize: 20 kbytes BurstSize: 20 kbytes
Profile5: Profile5:
Type: SimpleTokenBucket Type: SimpleTokenBucket
AverageRate: 100 kbps AverageRate: 100 kbps
BurstSize: 100 kbytes BurstSize: 100 kbytes
5.2.5. Null Meter 5.1.5. Null Meter
A null meter has only one output: always conforming, and no associated A null meter has only one output: always conforming, and no associated
temporal profile. Such a meter is useful to define in the event that the temporal profile. Such a meter is useful to define in the event that the
configuration or management interface does not have the flexibility to configuration or management interface does not have the flexibility to
omit a meter in a datapath segment. omit a meter in a datapath segment.
Meter5: Meter5:
Type: NullMeter Type: NullMeter
Output: Queue1 Output: Queue1
skipping to change at page 22, line 14 skipping to change at page 22, line 5
- Multiplexing - Multiplexing
- Counting - Counting
- Null action - do nothing - Null action - do nothing
The corresponding action elements are described in the following The corresponding action elements are described in the following
sections. sections.
Diffserv nodes may apply shaping, policing and/or marking to traffic 6.1. DSCP Marker
streams that exceed the bounds of their TCS in order to prevent a
traffic stream from seizing more than its share of resources from a
Diffserv network. Shaping, sometimes considered as a TC action, is
treated as a part of the queueing module in this model, as is the use of
Algorithmic Dropping techniques - see section 7. Policing is modelled
as either a concatenation of a Meter with an Absolute Dropper or as a
concatenation of an Algorithmic Dropper with a Scheduler. These elements
will discard packets which exceed the TCS. Marking is performed by a
Marker Action, which (in this context) alters the DSCP, and thus the
PHB, of the packet to give it a lower-grade treatment at subsequent
Diffserv nodes.
6.1. Marker
Markers are 1:1 elements which set a codepoint (e.g. the DSCP in an IP DSCP Markers are 1:1 elements which set a codepoint (e.g. the DSCP in an
header). Markers may also act on unmarked packets (e.g. those submitted IP header). DSCP Markers may also act on unmarked packets (e.g. those
with DSCP of zero) or may re-mark previously marked packets. In submitted with DSCP of zero) or may re-mark previously marked packets.
particular, the model supports the application of marking based on a In particular, the model supports the application of marking based on a
preceding classifier match. The mark set in a packet will determine its preceding classifier match. The mark set in a packet will determine its
subsequent treatment in downstream nodes of a network and possibly also subsequent PHB treatment in downstream nodes of a network and possibly
in subsequent processing stages within this router. also in subsequent processing stages within this router.
DSCP Markers for Diffserv are normally parameterized by a single DSCP Markers for Diffserv are normally parameterized by a single
parameter: the 6-bit DSCP to be marked in the packet header. parameter: the 6-bit DSCP to be marked in the packet header.
Marker1: Marker1:
Type: DSCPMarker Type: DSCPMarker
Mark: 010010 Mark: 010010
6.2. Absolute Dropper 6.2. Absolute Dropper
Absolute Droppers simply discard packets. There are no parameters for Absolute Droppers simply discard packets. There are no parameters for
these droppers. Because this Algorithmic Dropper is a terminating point these droppers. Because this Absolute Dropper is a terminating point of
of the datapath and has no outputs, it is probably desirable to forward the datapath and has no outputs it is probably desirable to forward the
the packet through a Counter Action first for instrumentation purposes. packet through a Counter Action first for instrumentation purposes.
AbsoluteDropper1: AbsoluteDropper1:
Type: AbsoluteDropper Type: AbsoluteDropper
Absolute Droppers are not the only elements than can cause a packet to Absolute Droppers are not the only elements than can cause a packet to
be discarded: another element is an Algorithmic Dropper element (see be discarded: another element is an Algorithmic Dropper element (see
Section 7.1.3). However, since this element's behavior is closely tied Section 7.1.3). However, since this element's behavior is closely tied
the state of one or more queues, we choose to distinguish it as a the state of one or more queues, we choose to distinguish it as a
separate functional datapath element. separate functional datapath element.
6.3. Multiplexor 6.3. Multiplexor
skipping to change at page 23, line 31 skipping to change at page 23, line 12
Type: Multiplexor Type: Multiplexor
Output: Queue2 Output: Queue2
6.4. Counter 6.4. Counter
One passive action is to account for the fact that a data packet was One passive action is to account for the fact that a data packet was
processed. The statistics that result might be used later for customer processed. The statistics that result might be used later for customer
billing, service verification or network engineering purposes. Counters billing, service verification or network engineering purposes. Counters
are 1:1 functional datapath elements which update a counter by L and a are 1:1 functional datapath elements which update a counter by L and a
packet counter by 1 every time a L-byte sized packet passes through packet counter by 1 every time a L-byte sized packet passes through
them. Counters can be used to count packets about to be be dropped by an them. Counters can be used to count packets about to be dropped by an
Absolute Dropper or to count packets arriving at or departing from some Absolute Dropper or to count packets arriving at or departing from some
other functional element. other functional element.
Counter1: Counter1:
Type: Counter Type: Counter
Output: Queue1 Output: Queue1
6.5. Null Action 6.5. Null Action
A null action has one input and one output. The element performs no A null action has one input and one output. The element performs no
skipping to change at page 24, line 25 skipping to change at page 24, line 4
performed), as a feedback control signal to reactive control protocols performed), as a feedback control signal to reactive control protocols
such as TCP, because a meter exceeds a configured profile (i.e. such as TCP, because a meter exceeds a configured profile (i.e.
policing). policing).
The queueing elements in this model represent a logical abstraction of a The queueing elements in this model represent a logical abstraction of a
queueing system, which is used to configure PHB-related parameters. The queueing system, which is used to configure PHB-related parameters. The
model can be used to represent a broad variety of possible model can be used to represent a broad variety of possible
implementations. However, it need not necessarily map one-to-one with implementations. However, it need not necessarily map one-to-one with
physical queueing systems in a specific router implementation. physical queueing systems in a specific router implementation.
Implementors should map the configurable parameters of the Implementors should map the configurable parameters of the
implementation's queueing systems to these queueing element parameters implementation's queueing systems to these queueing element parameters
as appropriate to achieve equivalent behaviors. as appropriate to achieve equivalent behaviors.
7.1. Queueing Model 7.1. Queueing Model
Queueing is a function a which lends itself to innovation. It must be Queueing is a function which lends itself to innovation. It must be
modelled to allow a broad range of possible implementations to be modelled to allow a broad range of possible implementations to be
represented using common structures and parameters. This model uses represented using common structures and parameters. This model uses
functional decomposition as a tool to permit the needed lattitude. functional decomposition as a tool to permit the needed lattitude.
Queueing systems perform three distinct, but related, functions: they Queueing systems perform three distinct, but related, functions: they
store packets, they modulate the departure of packets belonging to store packets, they modulate the departure of packets belonging to
various traffic streams and they selectively discard packets. This model various traffic streams and they selectively discard packets. This model
decomposes the queueing block into the component elements that perform decomposes queueing into the component elements that perform each of
each of these functions: Queues, Schedulers and Algorithmic Droppers, these functions: Queues, Schedulers and Algorithmic Droppers,
respectively. These elements may be connected together as part of a TCB respectively. These elements may be connected together as part of a
containing one or more Queues, zero or more Algorithmic Droppers and one TCB, as described in section 8.
or more Schedulers.
The remainder of this section discusses FIFO Queues: typically, the The remainder of this section discusses FIFO Queues: typically, the
Queue element of this model will be implemented as a FIFO data Queue element of this model will be implemented as a FIFO data
structure. However, this does not preclude implementations which are not structure. However, this does not preclude implementations which are not
strictly FIFO, in that they also support operations that remove or strictly FIFO, in that they also support operations that remove or
examine packets (e.g., for use by discarders) other than at the head or examine packets (e.g., for use by discarders) other than at the head or
tail. However, such operations MUST NOT have the effect of reordering tail. However, such operations MUST NOT have the effect of reordering
packets belonging to the same microflow. packets belonging to the same microflow.
Note that the term FIFO has multiple different common usages: it is Note that the term FIFO has multiple different common usages: it is
sometimes taken to mean, among other things, a data structure that sometimes taken to mean, among other things, a data structure that
permits items to be removed only in the order in which they were permits items to be removed only in the order in which they were
inserted or a service discipline which is non- reordering. inserted or a service discipline which is non- reordering.
7.1.1. FIFO Queue 7.1.1. FIFO Queue
In this model, a FIFO Queue element is a data structure which at any In this model, a FIFO Queue element is a data structure which at any
time may contain zero or more packets. It may have one or more time may contain zero or more packets. It may have one or more
thresholds associated with it. A FIFO has one or more inputs and exactly thresholds associated with it. A FIFO has one or more inputs and exactly
one output. It must support an enqueue operation to add a packet to the one output. It must support an enqueue operation to add a packet to the
tail of the queue, and a dequeue operation to remove a packet from the tail of the queue and a dequeue operation to remove a packet from the
head of the queue. Packets must be dequeued in the order in which they head of the queue. Packets must be dequeued in the order in which they
were enqueued. A FIFO has a current depth, which indicates the number of were enqueued. A FIFO has a current depth, which indicates the number of
packets that it contains at a particular time. FIFOs in this model are packets and/or bytes that it contains at a particular time. FIFOs in
modelled without inherent limits on their depth - obviously this does this model are modelled without inherent limits on their depth -
not reflect the reality of implementations: FIFO size limits are obviously this does not reflect the reality of implementations: FIFO
modelled here by an algorithmic dropper associated with the FIFO, size limits are modelled here by an algorithmic dropper associated with
typically at its input. It is quite likely that, every FIFO will be the FIFO, typically at its input. It is quite likely that every FIFO
preceded by an algorithmic dropper. One exception might be the case will be preceded by an algorithmic dropper. One exception might be the
where the packet stream has already been policed to a profile that can case where the packet stream has already been policed to a profile that
never exceed the scheduler bandwidth available at the FIFO's output -
this would not need an algorithmic dropper at the input to the FIFO. can never exceed the scheduler bandwidth available at the FIFO's output
- this would not need an algorithmic dropper at the input to the FIFO.
this would not need an algorithmic dropper at the input to the FIFO.
This representation of a FIFO allows for one common type of depth limit, This representation of a FIFO allows for one common type of depth limit,
one that results from a FIFO supplied from a limited pool of buffers, one that results from a FIFO supplied from a limited pool of buffers,
shared between multiple FIFOs. shared between multiple FIFOs.
In an implementation, packets are presumably stored in one or more In an implementation, packets are presumably stored in one or more
buffers. Buffers are allocated from one or more free buffer pools. If buffers. Buffers are allocated from one or more free buffer pools. If
there are multiple instances of a FIFO, their packet buffers may or may there are multiple instances of a FIFO, their packet buffers may or may
not be allocated out of the same free buffer pool. Free buffer pools may not be allocated out of the same free buffer pool. Free buffer pools may
also have one or more threshold associated with them, which may affect also have one or more thresholds associated with them, which may affect
discarding and/or scheduling. Other than this, buffering mechanisms are discarding and/or scheduling. Other than this, buffering mechanisms are
implementation specific and not part of this model. implementation specific and not part of this model.
A FIFO might be represented using the following parameters: A FIFO might be represented using the following parameters:
Queue1: Queue1:
Type: FIFO Type: FIFO
Output: Scheduler1 Output: Scheduler1
Note that a FIFO must provide triggers and/or current state information Note that a FIFO must provide triggers and/or current state information
skipping to change at page 26, line 4 skipping to change at page 25, line 31
Queue1: Queue1:
Type: FIFO Type: FIFO
Output: Scheduler1 Output: Scheduler1
Note that a FIFO must provide triggers and/or current state information Note that a FIFO must provide triggers and/or current state information
to other elements upstream and downstream from it: in particular, it is to other elements upstream and downstream from it: in particular, it is
likely that the current depth will need to be used by Algorithmic likely that the current depth will need to be used by Algorithmic
Dropper elements placed before or after the FIFO. It will also likely Dropper elements placed before or after the FIFO. It will also likely
need to provide an implicit "I have packets for you" signal to need to provide an implicit "I have packets for you" signal to
downstream Scheduler elements. downstream Scheduler elements.
7.1.2. Scheduler 7.1.2. Scheduler
A scheduler is an element which gates the departure of each packet that A scheduler is an element which gates the departure of each packet that
arrives at one of its inputs, based on a service discipline. It has one arrives at one of its inputs, based on a service discipline. It has one
or more input and exactly one output. Each input has an upstream element or more inputs and exactly one output. Each input has an upstream
to which it is connected, and a set of parameters that affects the element to which it is connected, and a set of parameters that affects
scheduling of packets received at that input. the scheduling of packets received at that input.
The service discipline (also known as a scheduling algorithm) is an The service discipline (also known as a scheduling algorithm) is an
algorithm which might take any of the following as its input(s): algorithm which might take any of the following as its input(s):
a) static parameters such as relative priority associated with each of a) static parameters such as relative priority associated with each of
the scheduler's inputs. the scheduler's inputs.
b) absolute token bucket parameters for maximum or minimum rates b) absolute token bucket parameters for maximum or minimum rates
associated with each of the scheduler's inputs. associated with each of the scheduler's inputs.
c) parameters, such as packet length or DSCP, associated with the c) parameters, such as packet length or DSCP, associated with the
packet currently present at its input. packet currently present at its input.
d) absolute time and/or local state. d) absolute time and/or local state.
Possible service disciplines fall into a number of categories, including Possible service disciplines fall into a number of categories, including
(but not limited to) first come, first served (FCFS), strict priority, (but not limited to) first come, first served (FCFS), strict priority,
weighted fair bandwidth sharing (e.g., WFQ, WRR, etc.), rate-limited weighted fair bandwidth sharing (e.g. WFQ), rate-limited strict priority
strict priority and rate-based. Service disciplines can be further and rate-based. Service disciplines can be further distinguished by
distinguished by whether they are work-conserving or non-work-conserving whether they are work-conserving or non-work-conserving (see Glossary).
(see Glossary). Non-work-conserving schedulers can be used to shape Non-work-conserving schedulers can be used to shape traffic streams to
traffic streams to match some profile by delaying packets that might be match some profile by delaying packets that might be deemed non-
deemed non-conforming by some downstream node: a packet is delayed until conforming by some downstream node: a packet is delayed until such time
such time as it would conform to a downstream meter using the same as it would conform to a downstream meter using the same profile.
profile.
[DSARCH] defines PHBs without specifying required scheduling algorithms. [DSARCH] defines PHBs without specifying required scheduling algorithms.
However, PHBs such as the class selectors [DSFIELD], EF [EF-PHB] and AF However, PHBs such as the class selectors [DSFIELD], EF [EF-PHB] and AF
[AF-PHB] have descriptions or configuration parameters which strongly [AF-PHB] have descriptions or configuration parameters which strongly
suggest the sort of scheduling discipline needed to implement them. This suggest the sort of scheduling discipline needed to implement them. This
memo discusses a minimal set of queue parameters to enable realization document discusses a minimal set of queue parameters to enable
of these per- hop behaviors. It does not attempt to specify an all- realization of these PHBs. It does not attempt to specify an all-
embracing set of parameters to cover all possible implementation models. embracing set of parameters to cover all possible implementation models.
A mimimal set includes: A mimimal set includes:
a) a minimum service rate profile which allows rate guarantees for a) a minimum service rate profile which allows rate guarantees for
each traffic stream as required by EF and AF without specifying the each traffic stream as required by EF and AF without specifying the
details of how excess bandwidth between these traffic streams is details of how excess bandwidth between these traffic streams is
shared. Additional parameters to control this behavior should be shared. Additional parameters to control this behavior should be
made available, but are dependent on the particular scheduling made available, but are dependent on the particular scheduling
algorithm implemented. algorithm implemented.
b) a service priority, used only after the minimum rate profiles of b) a service priority, used only after the minimum rate profiles of
all inputs have been satisfied, to decide how to allocate any all inputs have been satisfied, to decide how to allocate any
remaining bandwidth. remaining bandwidth.
c) a maximum service rate profile, for use only with a non-work- c) a maximum service rate profile, for use only with a non-work-
conserving service discipline. conserving service discipline.
For an implementation of the EF PHB using a strict priority scheduling Any one of these profiles is composed, for the purposes of this model,
algorithm that assumes that the aggregate EF rate has been appropriately of both a rate (in suitable units of bits, bytes or larger chunks in
bounded to avoid starvation, the minimum rate profile would be reported some unit of time) and a burst size, as discussed further in Appendix A.
as zero and the maximum service rate would be reported as line rate.
Such an implementation, with multiple priority classes, could also be By way of example, for an implementation of the EF PHB using a strict
used for the Diffserv class selectors [DSFIELD]. priority scheduling algorithm that assumes that the aggregate EF rate
has been appropriately bounded by upstream policing to avoid starvation
of other BAs, the service rate profiles are not used: the minimum
service rate profile would be defaulted to zero and the maximum service
rate profile would effectively be the "line rate". Such an
implementation, with multiple priority classes, could also be used for
the Diffserv class selectors [DSFIELD].
Alternatively, setting the service priority values for each input to the Alternatively, setting the service priority values for each input to the
scheduler to the same value enables the scheduler to satisfy the minimum scheduler to the same value enables the scheduler to satisfy the minimum
service rates for each input, so long as the sum of all minimum service service rates for each input, so long as the sum of all minimum service
rates is less than or equal to the line rate. rates is less than or equal to the line rate.
For example, a non-work-conserving scheduler, allocating spare bandwidth For example, a non-work-conserving scheduler, allocating spare bandwidth
equally between all its inputs, might be represented using the following equally between all its inputs, might be represented using the following
parameters: parameters:
skipping to change at page 28, line 23 skipping to change at page 28, line 4
Input3: Input3:
MaxRateProfile: WorkConserving MaxRateProfile: WorkConserving
MinRateProfile: none MinRateProfile: none
Priority: 3 Priority: 3
7.1.3. Algorithmic Dropper 7.1.3. Algorithmic Dropper
An Algorithmic Dropper is an element which selectively discards packets An Algorithmic Dropper is an element which selectively discards packets
that arrive at its input, based on a discarding algorithm. It has one that arrive at its input, based on a discarding algorithm. It has one
data input and one output.
In this model (but not necessarily in a real implementation), a packet data input and one output. In this model (but not necessarily in a real
enters the dropper at its input and either its buffer is returned to a implementation), a packet enters the dropper at its input and either its
free buffer pool or the packet exits the dropper at the output. buffer is returned to a free buffer pool or the packet exits the dropper
Alternatively, an Algorithmic Dropper may invoke operations on a FIFO at the output.
which selectively removes a packet, then return its buffer to the free
buffer pool, based on a discarding algorithm. In this case, the Alternatively, an Algorithmic Dropper can be thought of as invoking
operation ould be modelled as a side-effect on the FIFO upon which it operations on a FIFO which selectively remove a packet and return its
operated, rather than as having a discrete input and output. These two buffer to the free buffer pool based on a discarding algorithm. In this
treatments are equivalent and we choose the former here. case, the operation could be modelled as being a side-effect on the FIFO
upon which it operated, rather than as having a discrete input and
output. This treatment is equivalent and we choose the one described in
the previous paragraph for this model.
The Algorithmic Dropper is modelled as having a single input. It is The Algorithmic Dropper is modelled as having a single input. It is
possible that packets which were classified differently by a Classifier possible that packets which were classified differently by a Classifier
in this TCB will end up passing through the same dropper. The dropper's in this TCB will end up passing through the same dropper. The dropper's
algorithm may need to apply different calculations based on algorithm may need to apply different calculations based on
characteristics of the incoming packet e.g. its DSCP. So there is a characteristics of the incoming packet e.g. its DSCP. So there is a
need, in implementations of this model, to be able to relate information need, in implementations of this model, to be able to relate information
about which classifier element was matched by a packet from a Classifier about which classifier element was matched by a packet from a Classifier
to an Algorithmic Dropper. In the rare cases where this is required, to an Algorithmic Dropper. In the rare cases where this is required,
the chosen model is to insert another Classifier element at this point the chosen model is to insert another Classifier element at this point
in the flow and for it to feed into multiple Algorithmic Dropper in the flow and for it to feed into multiple Algorithmic Dropper
elements, each one implementing a drop calculation that is independent elements, each one implementing a drop calculation that is independent
of any classification keys of the packet: this will likely require the of any classification keys of the packet: this will likely require the
creation of a new TCB to contain the Classifier and the Algorithmic creation of a new TCB to contain the Classifier and the Algorithmic
Dropper elements. Dropper elements.
There are many other formulations of a model that could represent this NOTE: There are many other formulations of a model that could
linkage, other than the one described above: one formulation would have represent this linkage that are different to the one described
above: one formulation would have been to have a pointer from one
of the drop probability calculation algorithms inside the dropper
to the original Classifier element that selects this algorithm.
Another way would have been to have multiple "inputs" to the
Algorithmic Dropper element fed from the preceding elements,
leading eventually back to the Classifier elements that matched the
packet. Yet another formulation might have been for the Classifier
to (logically) include some sort of "classification identifier"
along with the packet along its path, for use by any subsequent
element. And yet another could have been to include a classifier
inside the dropper, in order for it to pick out the drop algorithm
to be applied. These other approaches could be used by
implementations but were deemed to be less clear than the approach
taken here.
been to have a pointer from one of the drop probability calculation An Algorithmic Dropper, illustrated in Figure 5, has one or more
algorithms inside the dropper to the original Classifier element that triggers that cause it to make a decision whether or not to drop one (or
selects this algorithm. Another way would have been to have multiple
"inputs" to the Algorithmic Dropper element fed from the preceding
elements, leading eventually back to the Classifier elements that
matched the packet. Yet another formulation might have been for the
Classifier to (logically) include some sort of "classification
identifier" along with the packet along its path, for use by any
subsequent element. And yet another could have been to include a
classifier inside the dropper, in order for it to pick out the drop
algorithm to be applied. All of these other approaches were deemed to be
more clumsy or less useful than the approach taken here.
An Algorithmic Dropper, shown in Figure 5, has one or more triggers that possibly more than one) packet. A trigger may be internal (the arrival
cause it to make a decision whether or not to drop one (or possibly more of a packet at the input to the dropper) or it may be external
than one) packet. A trigger may be internal (the arrival of a packet at (resulting from one or more state changes at another element, such as a
the input to the dropper) or it may be external (resulting from one or FIFO depth crossing a threshold or a scheduling event). It is likely
more state changes at another element, such as a FIFO depth exceeding a that an instantaneous FIFO depth will need to be smoothed over some
threshold or a scheduling event). It is likely that an instantaneous averaging interval. Some dropping algorithms may require several trigger
FIFO depth will need to be smoothed over some averaging interval. Some inputs feeding back from events elsewhere in the system e.g. depth-
dropping algorithms may require several trigger inputs feeding back from smoothing functions that calculate averages over more than one time
events elsewhere in the system e.g. smoothing functions that calculate interval. Smoothing functions are outside the scope of this document
averages over more than one time interval. Smoothing functions are and are not modelled here, we merely indicate where they might be added
outside the scope of this document and are not modelled here, we merely in the model.
indicate where they might be added in the model.
A trigger may be a boolean combination of events (e.g. a FIFO depth A trigger may be a boolean combination of events (e.g. a FIFO depth
exceeding a threshold OR a buffer pool depth falling below a threshold). exceeding a threshold OR a buffer pool depth falling below a threshold).
The dropping algorithm makes a decision on whether to forward or to The dropping algorithm makes a decision on whether to forward or to
discard a packet. It takes as its parameters some set of dynamic discard a packet and, if discarding, whether to discard it from the
parameters (e.g. averaged or instantaneous FIFO depth) and some set of head, tail or other part of the associated queue. It takes as its
static parameters (e.g. thresholds) and possibly parameters associated parameters some set of dynamic parameters e.g. smoothed or instantaneous
with the packet (e.g. its PHB, as determined by a classifier, which will
determine on which of the droppers inputs the packet arrives). It may
also have internal state and is likely to keep counters regarding the
dropped packets (there is no appropriate place here to include a Counter
Action element).
RED, RED-on-In-and-Out (RIO) and Drop-on-threshold are examples of
dropping algorithms. Tail-dropping and head-dropping are effected by the
location of the dropper relative to the FIFO.
Note that, although an Algorithmic Dropper may require knowledge of data
fields in a packet, as discovered by a Classifier in the same TCB, it
may not modify the packet (i.e. it is not a marker).
+--------------------------------------+ +--------------------------------------+
| +------------+ +-----------+ |Algorithmic | +------------+ +-----------+ |Algorithmic
| | smoothing | n |trigger & | |Dropper | | smoothing | n |trigger & | |Dropper
| | function(s)|---/--->|discard | | | | function(s)|---/--->|discard | |
| | (optional) | |calc. | | | | (optional) | |calc. | |
| +------------+ +-----------+ | | +------------+ +-----------+ |
| ^ TailDrop| |HeadDrop | | ^ TailDrop| |HeadDrop |
+------------|-------------|-|---------+ +------------|-------------|-|---------+
| | | | | |
+---|-------------+ | +---|-------------+ |
| | | | | |
v |Depth v v |Depth v
Input ----------------------+ to Scheduler Input ----------------------+ Output
-----------------------------> |x|x|x|x|x|x|x|-------------------> -----------------------------> |x|x|x|x|x|x|x|------------------->
----------------------+ ----------------------+
FIFO | FIFO |
| |
| | | | | |
| v | bit-bucket | v | bit-bucket
+---+ +---+
Figure 5. Algorithmic Dropper + Queue Figure 5. Algorithmic Dropper + Queue
An Algorithmic Dropper which uses a RIO algorithm might be represented FIFO depth, some set of static parameters e.g. thresholds, and possibly
using the following parameters: other parameters associated with the packet. It may also have internal
state and is likely to keep counters regarding the dropped packets
(there is no appropriate place here to include a Counter Action
element). Note that, although an Algorithmic Dropper may require
knowledge of data fields in a packet, as discovered by a Classifier in
the same TCB, it may not modify the packet (i.e. it is not a marker).
AlgorithmicDropper1: RED, RED-on-In-and-Out (RIO) and Drop-on-threshold are examples of
dropping algorithms. Tail-dropping and head-dropping are effected by the
location of the dropper relative to the FIFO.
For example, a dropper using a RIO algorithm might be represented using
2 Algorithmic Droppers with the following parameters:
AlgorithmicDropper1: (for in-profile traffic)
Type: AlgorithmicDropper Type: AlgorithmicDropper
Discipline: RIO Discipline: RED, discard from tail
Trigger: Internal Trigger: Internal
Output: Fifo1 Output: Fifo1
InputA: (in profile)
MinThresh: Fifo1.Depth > 20 kbyte MinThresh: Fifo1.Depth > 20 kbyte
MaxThresh: Fifo1.Depth > 30 kbyte MaxThresh: Fifo1.Depth > 30 kbyte
SampleWeight .002
MaxDropProb 1%
InputB: (out of profile) AlgorithmicDropper2: (for out-of-profile traffic)
Type: AlgorithmicDropper
Discipline: RED, discard from tail
Trigger: Internal
Output: Fifo1
MinThresh: Fifo1.Depth > 10 kbyte MinThresh: Fifo1.Depth > 10 kbyte
MaxThresh: Fifo1.Depth > 20 kbyte MaxThresh: Fifo1.Depth > 20 kbyte
SampleWeight .002 SampleWeight .002
MaxDropProb 1% MaxDropProb 2%
Another form of Algorithmic Dropper, a threshold-dropper, might be Another form of Algorithmic Dropper, a threshold-dropper, might be
represented using the following parameters: represented using the following parameters:
AlgorithmicDropper2: AlgorithmicDropper3:
Type: AlgorithmicDropper Type: AlgorithmicDropper
Discipline: Drop-on-threshold Discipline: Drop-on-threshold, discard from tail
Trigger: Fifo2.Depth > 20 kbyte Trigger: Fifo2.Depth > 20 kbyte
Output: Fifo1 Output: Fifo1
Yet another Algorithmic Dropper which drops all out-of-profile packets
whenever the FIFO threshold exceeds a certain depth (this Algorithmic
Dropper is not part of the larger TCB example) might be represented with
the following parameters:
AlgorithmicDropper3:
Type: AlgorithmicDropper2Input
Discipline: Drop-out-packets-on-threshold
Output: Fifo3
InputA: (in profile)
Trigger: none
InputB: (out of profile)
Trigger: Fifo3.Depth > 100 kbyte
7.1.4. Constructing queueing blocks from the elements
A queueing block is constructed by concatenation of these functional
datapath elements. Elements of the same type may appear more than once
in a queueing block, either in parallel or in series. Typically, a
queueing block will have relatively many elements in parallel and few in
series. Iteration and recursion are not supported constructs in this
grammar. A queueing block must have at least one Queue, zero or more
Algorithmic Droppers and at least one Scheduler. The following inter-
connections are allowed:
1) The input of a Queue may be the input of the queueing block or it
may be connected to the output of an Algorithmic Dropper or to an
output of a Scheduler.
2) Each input of a Scheduler may be connected to the output of a
Queue, to the output of an Algorithmic Dropper or to the output of
another Scheduler.
3) The input of an Algorithmic Dropper must be the input of the
queueing block.
4) The output of the queueing block may be the output of a Queue, an
Algorithmic Dropper or a Scheduler.
Note, in particular, that Schedulers may operate in series such that a
packet at the head of a Queue feeding the concatenated Schedulers is
serviced only after all of the scheduling criteria are met. For example,
a Queue which carries EF traffic streams may be served first by a non-
work-conserving Scheduler to shape the stream to a maximum rate, then by
a work-conserving Scheduler to mix EF traffic streams with other traffic
streams. Alternatively, there might be a Queue and/or a dropper between
the two Schedulers.
7.2. Sharing load among traffic streams using queueing 7.2. Sharing load among traffic streams using queueing
Queues are used, in Differentiated Services, for a number of purposes. Queues are used, in Differentiated Services, for a number of purposes.
In essence, they are simply places to store traffic until it is In essence, they are simply places to store traffic until it is
transmitted. However, when several queues are used together in a transmitted. However, when several queues are used together in a
queueing system, they can also achieve effects beyond that for given queueing system, they can also achieve effects beyond that for given
traffic streams. They can be used to limit variation in delay or impose traffic streams. They can be used to limit variation in delay or impose
a maximum rate (shaping), to permit several streams to share a link in a a maximum rate (shaping), to permit several streams to share a link in a
semi-predictable fashion (load sharing), or to move variation in delay semi-predictable fashion (load sharing), or to move variation in delay
from some streams to other streams. from some streams to other streams.
Traffic shaping is often used to condition traffic such that packets Traffic shaping is often used to condition traffic such that packets
arriving in a burst will be "smoothed" and deemed conforming by arriving in a burst will be "smoothed" and deemed conforming by
subsequent downstream meters in this or other nodes. Shaping may also be subsequent downstream meters in this or other nodes. In [DSARCH] a
used to isolate certain traffic streams from the effects of other shaper is described as a queueing element controlled by a meter which
traffic streams of the same BA. defines its temporal profile. However, this representation of a shaper
differs substantially from typical shaper implementations.
In [DSARCH] a shaper is described as a queueing element controlled by a
meter which defines its temporal profile. However, this representation
of a shaper differs substantially from typical shaper implementations.
In this conceptual model, a shaper is realized by using a non-work- In the model described here, a shaper is realized by using a non-work-
conserving Scheduler. Some implementations may elect to have queues conserving Scheduler. Some implementations may elect to have queues
whose sole purpose is shaping, while others may integrate the shaping whose sole purpose is shaping, while others may integrate the shaping
function with other buffering, discarding and scheduling associated with function with other buffering, discarding and scheduling associated with
access to a resource. Shapers operate by delaying the departure of access to a resource. Shapers operate by delaying the departure of
packets that would be deemed non-conforming by a meter configured to the packets that would be deemed non-conforming by a meter configured to the
shaper's maximum service rate profile. The packet is scheduled to depart shaper's maximum service rate profile. The packet is scheduled to depart
no sooner than such time that it would become conforming. no sooner than such time that it would become conforming.
7.2.1. Load Sharing 7.2.1. Load Sharing
Load sharing is the traditional use of queues. It was theoretically Load sharing is the traditional use of queues. It was theoretically
explored in a paper by Floyd [FLOYD] in 1993, but has been in use in explored in a paper by Floyd [FLOYD] in 1993, but has been in use in
communications systems since the 1970's. communications systems since the 1970's.
[DSARCH] discusses load sharing as dividing an interface among traffic [DSARCH] discusses load sharing as dividing an interface among traffic
classes predictably or applying a minimum rate to each of a set of classes predictably or applying a minimum rate to each of a set of
traffic classes, which might be measured as an absolute lower bound on traffic classes, which might be measured as an absolute lower bound on
the rate a traffic stream achieves, or a fraction of the rate an the rate a traffic stream achieves or a fraction of the rate an
interface offers. It is generally implemented as some form of weighted interface offers. It is generally implemented as some form of weighted
round robin among a set of FIFO queues or WFQ system. This has queueing algorithm among a set of FIFO queues i.e. a WFQ scheme. This
interesting side-effects. has interesting side-effects.
A key effect sought is to ensure that the mean rate the traffic in a A key effect sought is to ensure that the mean rate the traffic in a
stream experiences is never lower than some threshold when there is at stream experiences is never lower than some threshold when there is at
least that much traffic to send. When there is less traffic than this, least that much traffic to send. When there is less traffic than this,
the queue tends to be starved for traffic, meaning that the queuing the queue tends to be starved of traffic, meaning that the queuing
system will not delay its traffic by very much. When there is system will not delay its traffic by very much. When there is
significantly more and the queue fills, packets in this class will be significantly more traffic and the queue starts filling, packets in this
delayed significantly more than traffic in other classes that are under-
using their available capacity. This form of queuing system therefore class will be delayed significantly more than traffic in other classes
tends to move delay and variation in delay from under-used classes of that are under-using their available capacity. This form of queuing
traffic to heavier users, as well as managing the rates of the traffic system therefore tends to move delay and variation in delay from under-
streams. used classes of traffic to heavier users, as well as managing the rates
of the traffic streams.
A side-effect of a WRR or WFQ implementation is that between any two A side-effect of a WRR or WFQ implementation is that between any two
packets in a given traffic class, the scheduler may emit one or more packets in a given traffic class, the scheduler may emit one or more
packets from each of the other classes in the queuing system. In cases packets from each of the other classes in the queuing system. In cases
where average behavior is in view, this is perfectly acceptable. In where average behavior is in view, this is perfectly acceptable. In
cases where traffic is very intolerant of jitter and there are a number cases where traffic is very intolerant of jitter and there are a number
of competing classes, this may have undesirable consequences. of competing classes, this may have undesirable consequences.
7.2.2. Traffic Priority 7.2.2. Traffic Priority
Traffic Prioritization is a special case of load sharing, wherein a Traffic Prioritization is a special case of load sharing, wherein a
certain traffic class is deemed so jitter-intolerant that if it has certain traffic class is deemed so jitter-intolerant that if it has
traffic present, that traffic must be sent at the earliest possible traffic present, that traffic must be sent at the earliest possible
time. By extension, several priorities might be defined, such that time. By extension, several priorities might be defined, such that
traffic in each of several classes is given preferential service over traffic in each of several classes is given preferential service over
any traffic of a lower class. It is the obvious implementation of IP any traffic of a lower class. It is the obvious implementation of IP
Precedence as described in [RFC 791], of 802.1p traffic classes Precedence as described in [RFC 791], of 802.1p traffic classes [802.1D]
[802.1D], and other similar technologies. and other similar technologies.
Priority is often abused in real networks; people tend to think that Priority is often abused in real networks; people tend to think that
traffic which has a business priority deserves this treatment, and talk traffic which has a high business priority deserves this treatment and
more about the business imperatives than the actual application talk more about the business imperatives than the actual application
requirements. This can have severe consequences; networks have been requirements. This can have severe consequences; networks have been
configured which placed business-critical traffic at a higher priority configured which placed business-critical traffic at a higher priority
than routing traffic, resulting in congestive collapse of the networks. than routing-protocol traffic, resulting in collapse of the network's
However, it has a legitimate use in services like EF, where it is management or control systems. However, it may have a legitimate use for
absolutely known, due to policing, that a traffic stream does not abuse services based on an Expedited Forwarding (EF) PHB, where it is
its rate, and the application is indeed jitter-intolerant enough to absolutely sure, thanks to policing at all possible traffic entry
points, that a traffic stream does not abuse its rate and that the
merit this type of handling. application is indeed jitter-intolerant enough to merit this type of
handling. Note that, even in cases with well-policed ingress points,
there is still the possibility of unexpected traffic loops within an un-
policed core part of the network causing such collapse.
8. Traffic Conditioning Blocks (TCBs) 8. Traffic Conditioning Blocks (TCBs)
The Classifier, Meter, Action, Algorithmic Dropper, Queue and Scheduler The Classifier, Meter, Action, Algorithmic Dropper, Queue and Scheduler
functional datapath elements described above can be combined into functional datapath elements described above can be combined into
Traffic Conditioning Blocks (TCBs). A TCB is an abstraction of a set of Traffic Conditioning Blocks (TCBs). A TCB is an abstraction of a set of
functional datapath elements that may be used to facilitate the functional datapath elements that may be used to facilitate the
definition of specific traffic conditioning functionality e.g. it might definition of specific traffic conditioning functionality e.g. it might
be likened to a template which can be replicated multiple times for be likened to a template which can be replicated multiple times for
different traffic streams or different customers. It has no likely different traffic streams or different customers. It has no likely
physical representation in the implementation of the data path: it is physical representation in the implementation of the data path: it is
invented purely as an abstraction for use by management tools. invented purely as an abstraction for use by management tools.
This model describes the configuration and management of a Diffserv This model describes the configuration and management of a Diffserv
interface in terms of a TCB that contains, by definition, zero or more interface in terms of a TCB that contains, by definition, zero or more
Classifier, Meter, Action, Algorithmic Dropper, Queue and Scheduler Classifier, Meter, Action, Algorithmic Dropper, Queue and Scheduler
elements. These elements are arranged arbitrarily according to the elements. These elements are arranged arbitrarily according to the
policy being expressed, but always in the order here. Traffic may be policy being expressed, but always in the order here. Traffic may be
skipping to change at page 34, line 28 skipping to change at page 33, line 19
This model describes the configuration and management of a Diffserv This model describes the configuration and management of a Diffserv
interface in terms of a TCB that contains, by definition, zero or more interface in terms of a TCB that contains, by definition, zero or more
Classifier, Meter, Action, Algorithmic Dropper, Queue and Scheduler Classifier, Meter, Action, Algorithmic Dropper, Queue and Scheduler
elements. These elements are arranged arbitrarily according to the elements. These elements are arranged arbitrarily according to the
policy being expressed, but always in the order here. Traffic may be policy being expressed, but always in the order here. Traffic may be
classified; classified traffic may be metered; each stream of traffic classified; classified traffic may be metered; each stream of traffic
identified by a combination of classifiers and meters may have some set identified by a combination of classifiers and meters may have some set
of actions performed on it, followed by drop algorithms; packets of the of actions performed on it, followed by drop algorithms; packets of the
traffic stream may ultimately be stored into a queue and then be traffic stream may ultimately be stored into a queue and then be
scheduled out to the next TCB or physical interface. It is possible to scheduled out to the next TCB or physical interface. It is permissible
omit elements or include null elements of any type, or to concatenate to omit elements or include null elements of any type, or to concatenate
multiple functional datapath elements of the same type. multiple functional datapath elements of the same type.
When the Diffserv treatment for a given packet needs to have those When the Diffserv treatment for a given packet needs to have such
building blocks repeated, this is performed by cascading multiple TCBs: building blocks repeated, this is performed by cascading multiple TCBs:
an output of one TCB may drive the input of a succeeding one. For an output of one TCB may drive the input of a succeeding one. For
example, consider the case where traffic of a set of classes is shaped example, consider the case where traffic of a set of classes is shaped
to a set of rates, but the total output rate of the group of classes to a set of rates, but the total output rate of the group of classes
must also be limited to a rate. One might imagine a set of network news must also be limited to a rate. One might imagine a set of network news
feeds, each with a certain maximum rate, and a policy that their feeds, each with a certain maximum rate, and a policy that their
aggregate may not exceed some figure. This may be simply accomplished by aggregate may not exceed some figure. This may be simply accomplished by
cascading two TCBs. The first classifies the traffic into its separate cascading two TCBs. The first classifies the traffic into its separate
feeds and queues each feed separately. The feeds (or a subset of them) feeds and queues each feed separately. The feeds (or a subset of them)
are now fed into a second TCB, which places all input (these news feeds) are now fed into a second TCB, which places all input (these news feeds)
into a single queue with a certain maximum rate. In implementation, one into a single queue with a certain maximum rate. In implementation, one
could imagine this as the several literal queues, a CBQ or WFQ system could imagine this as the several literal queues, a CBQ or WFQ system
with an appropriate (and complex) weighting scheme, or a number of other with an appropriate (and complex) weighting scheme, or a number of other
approaches. But they would have the same externally measurable effect on approaches. But they would have the same externally measurable effect on
the traffic as if they had been literally implemented with separate the traffic as if they had been literally implemented with separate
TCBs. TCBs.
8.1. TCB
A generalised TCB might consist of the following stages: A generalised TCB might consist of the following stages:
- Classification stage - Classification stage
- Metering stage - Metering stage
- Action stage - Action stage (involving Markers, Absolute Droppers,
- Algorithmic Dropping stage Counters and Multiplexors)
- Queueing stage - Queueing stage (involving Algorithmic Droppers, Queues
- Scheduling stage and Schedulers)
where each stage may consist of a set of parallel datapaths consisting where each stage may consist of a set of parallel datapaths consisting
of pipelined elements. of pipelined elements.
A Classifier or a Meter is typically a 1:N element, an Action, A Classifier or a Meter is typically a 1:N element, an Action,
Algorithmic Dropper or Queue is typically a 1:1 element and a Scheduler Algorithmic Dropper or Queue is typically a 1:1 element and a Scheduler
is a N:1 element. A complete TCB should, however, result in a 1:1 or 1:N is a N:1 element. A complete TCB should, however, result in a 1:1 or 1:N
abstract element. Note that the fan-in or fan-out of an element is not abstract element. Note that the fan-in or fan-out of an element is not
an important defining characteristic of this taxonomy. an important defining characteristic of this taxonomy.
8.1. An Example TCB 8.1.1. Building blocks for Queueing
Some particular rules are applied to the ordering of elements within a
Queueing stage within a TCB: elements of the same type may appear more
than once, either in parallel or in series. Typically, a queueing stage
will have relatively many elements in parallel and few in series.
Iteration and recursion are not supported constructs (the elements are
arranged in an acyclic graph). The following inter-connections of
elements are allowed:
1) The input of a Queue may be the input of the queueing block or it
may be connected to the output of an Algorithmic Dropper or to an
output of a Scheduler.
2) Each input of a Scheduler may be connected to the output of a
Queue, to the output of an Algorithmic Dropper or to the output of
another Scheduler.
3) The input of an Algorithmic Dropper must be the first element of
the queueing stage, the output of another Algorithmic Dropper.
4) The output of the queueing block may be the output of a Queue, an
Algorithmic Dropper or a Scheduler.
Note, in particular, that Schedulers may operate in series such that a
packet at the head of a Queue feeding the concatenated Schedulers is
serviced only after all of the scheduling criteria are met. For example,
a Queue which carries EF traffic streams may be served first by a non-
work-conserving Scheduler to shape the stream to a maximum rate, then by
a work-conserving Scheduler to mix EF traffic streams with other traffic
streams. Alternatively, there might be a Queue and/or a dropper between
the two Schedulers.
8.2. An Example TCB
A SLS is presumed to have been negotiated between the customer and the A SLS is presumed to have been negotiated between the customer and the
provider which specifies the handling of the customer's traffic by the provider which specifies the handling of the customer's traffic, as
provider's network. The agreement might be of the following form: defined by a TCS) by the provider's network. The agreement might be of
the following form:
DSCP PHB Profile Treatment DSCP PHB Profile Treatment
---- --- ------- ---------------------- ---- --- ------- ----------------------
001001 EF Profile4 Discard non-conforming. 001001 EF Profile4 Discard non-conforming.
001100 AF11 Profile5 Shape to profile, tail-drop when full. 001100 AF11 Profile5 Shape to profile, tail-drop when full.
001101 AF21 Profile3 Re-mark non-conforming to DSCP 001000, 001101 AF21 Profile3 Re-mark non-conforming to DSCP 001000,
tail-drop when full. tail-drop when full.
other BE none Apply RED-like dropping. other BE none Apply RED-like dropping.
This SLS specifies that the customer may submit packets marked for DSCP This SLS specifies that the customer may submit packets marked for DSCP
001001 which will get EF treatment so long as they remain conforming to 001001 which will get EF treatment so long as they remain conforming to
Profile1 and will be discarded if they exceed this profile. The Profile4 and will be discarded if they exceed this profile. The
discarded packets are counted in this example, perhaps for use by the discarded packets are counted in this example, perhaps for use by the
provider's sales department in convincing the customer to buy a larger provider's sales department in convincing the customer to buy a larger
SLS. Packets marked for DSCP 001100 will be shaped to Profile2 before SLS. Packets marked for DSCP 001100 will be shaped to Profile5 before
forwarding. Packets marked for DSCP 001101 will be metered to Profile3 forwarding. Packets marked for DSCP 001101 will be metered to Profile3
with non-conforming packets "downgraded" by being re-marked with a DSCP with non-conforming packets "downgraded" by being re-marked with a DSCP
of 001000. It is implicit in this agreement that conforming packets are of 001000. It is implicit in this agreement that conforming packets are
given the PHB originally indicated by the packets' DSCP field. given the PHB originally indicated by the packets' DSCP field.
Figures 6 and 7 illustrates a TCB that might be used to handle this SLS Figures 6 and 7 illustrates a TCB that might be used to handle this SLS
at an ingress interface at the customer/provider boundary. at an ingress interface at the customer/provider boundary.
The Classification stage of this example consists of a single BA The Classification stage of this example consists of a single BA
classifier. The BA classifier is used to separate traffic based on the classifier. The BA classifier is used to separate traffic based on the
Diffserv service level requested by the customer (as indicated by the Diffserv service level requested by the customer (as indicated by the
DSCP in each submitted packet's IP header). We illustrate three DSCP
filter values: A, B and C. The 'X' in the BA classifier is a wildcard
filter that matches every packet not otherwise matched.
The path for DSCP 001100 proceeds directly to Dropper1 whilst the paths
for DSCP 001001 and 001101 include a metering stage. All other traffic
is passed directly on to Dropper3. There is a separate meter for each
set of packets corresponding to classifier outputs A and C. Each meter
uses a specific profile, as specified in the TCS, for the corresponding
Diffserv service level. The meters in this example each indicate one of
two conformance levels: conforming or non-conforming.
Following the Metering stage is an Action stage in some of the branches.
Packets submitted for DSCP 001001 (Classifier output A) that are deemed
non-conforming by Meter1 are counted and discarded while packets that
are conforming are passed on to Queue1. Packets submitted for DSCP
001101 (Classifier output C) that are deemed non-conforming by Meter2
are re-marked and then both conforming and non-conforming packets are
multiplexed together before being passed on to Dropper2/Queue3.
+-----+ +-----+
| A|---------------------------> to Queue1 | A|---------------------------> to Queue1
+->| | +->| |
| | B|--+ +-----+ +-----+ | | B|--+ +-----+ +-----+
| +-----+ | | | | | | +-----+ | | | | |
| Meter1 +->| |--->| | | Meter1 +->| |--->| |
| | | | | | | | | |
| +-----+ +-----+ | +-----+ +-----+
| Counter1 Absolute | Counter1 Absolute
submitted +-----+ | Dropper1 submitted +-----+ | Dropper1
traffic | A|-----+ traffic | A|-----+
--------->| B|----------------------------------------> to Dropper1 --------->| B|--------------------------------------> to AlgDropper1
| C|-----+ | C|-----+
| X|--+ | | X|--+ |
+-----+ | | +-----+ +-----+ +-----+ | | +-----+ +-----+
Classifier1| | | A|--------------->|A | Classifier1| | | A|--------------->|A |
(BA) | +->| | | |--> to Dropper2 (BA) | +->| | | |--> to AlgDrop2
| | B|--+ +-----+ +->|B | | | B|--+ +-----+ +->|B |
| +-----+ | | | | +-----+ | +-----+ | | | | +-----+
| Meter2 +->| |-+ Mux1 | Meter2 +->| |-+ Mux1
| | | | | |
| +-----+ | +-----+
| Marker1 | Marker1
+-------------------------------------> to Dropper3 +-----------------------------------> to AlgDropper3
Figure 6: An Example Traffic Conditioning Block (Part 1) Figure 6: An Example Traffic Conditioning Block (Part 1)
DSCP in each submitted packet's IP header). We illustrate three DSCP
filter values: A, B and C. The 'X' in the BA classifier is a wildcard
filter that matches every packet not otherwise matched.
The path for DSCP 001100 proceeds directly to Dropper1 whilst the paths
for DSCP 001001 and 001101 include a metering stage. All other traffic
is passed directly on to Dropper3. There is a separate meter for each
set of packets corresponding to classifier outputs A and C. Each meter
uses a specific profile, as specified in the TCS, for the corresponding
Diffserv service level. The meters in this example each indicate one of
two conformance levels: conforming or non-conforming.
Following the Metering stage is an Action stage in some of the branches.
Packets submitted for DSCP 001001 (Classifier output A) that are deemed
non- conforming by Meter1 are counted and discarded while packets that
are conforming are passed on to Queue1. Packets submitted for DSCP
001101 (Classifier output C) that are deemed non-conforming by Meter2
are re-marked and then both conforming and non-conforming packets are
multiplexed together before being passed on to Dropper2/Queue3.
The Algorithmic Dropping, Queueing and Scheduling stages are realised as The Algorithmic Dropping, Queueing and Scheduling stages are realised as
follows, illustrated in figure 7. Note that the figure does not show any follows, illustrated in figure 7. Note that the figure does not show any
of the implicit control linkages between elements that allow e.g. an of the implicit control linkages between elements that allow e.g. an
Algorithmic Dropper to sense the current state of a succeeding Queue. Algorithmic Dropper to sense the current state of a succeeding Queue.
Conforming DSCP 001001 packets from Meter1 are passed directly to Conforming DSCP 001001 packets from Meter1 are passed directly to
Queue1: there is no way, with a configuration of the following Scheduler Queue1: there is no way, with configuration of the following Scheduler
that patches the metering, for these packets to overflow the depth of to match the metering, for these packets to overflow the depth of Queue1
Queue1 so there is never a requirement for dropping at this point. so there is no requirement for dropping at this point. Packets marked
Packets marked for DSCP 001100 must be passed through a tail-dropper, for DSCP 001100 must be passed through a tail-dropper, AlgDropper1,
Dropper1, which serves to limit the depth of the following queue, which serves to limit the depth of the following queue, Queue2: packets
Queue2: packets that arrive to a full queue will be discarded - this is that arrive to a full queue will be discarded. This is likely to be an
likely to be an error case: the customer is obviously not sticking to error case: the customer is obviously not sticking to its agreed
its agreed profile. Similarly, all packets from the original DSCP profile. Similarly, all packets from the original DSCP 001101 stream
001101 stream (some may have been re-marked by this stage) are passed to (some may have been re-marked by this stage) are passed to AlgDropper2
Dropper2 and Queue3. Packets marked for all other DSCPs are passed to and Queue3. Packets marked for all other DSCPs are passed to
Dropper3 which is a RED-like algorithmic dropper: based on feedback of AlgDropper3 which is a RED-like Algorithmic Dropper: based on feedback
the current depth of Queue4, this dropper is likely to discard enough of the current depth of Queue4, this dropper is supposed to discard
packets from its input stream to keep the queue depth under control. enough packets from its input stream to keep the queue depth under
control.
These four Queue elements are then serviced by a Scheduler element These four Queue elements are then serviced by a Scheduler element
Scheduler1: tis must be configured to give each of its inputs an Scheduler1: this must be configured to give each of its inputs an
appropriate priority and/or bandwidth share. Inputs A and C are given appropriate priority and/or bandwidth share. Inputs A and C are given
guarantees of bandwidth, as appropriate for the contracted profiles. guarantees of bandwidth, as appropriate for the contracted profiles.
Input B is given a limit on the bandwidth it can use i.e. a non-work- Input B is given a limit on the bandwidth it can use i.e. a non-work-
conserving discipline in order to achieve the desired shaping of this conserving discipline in order to achieve the desired shaping of this
stream. Input D is given no limits or guarantees but a lower priority stream. Input D is given no limits or guarantees but a lower priority
than the other queues, appropriate for its best-effort status. Traffic than the other queues, appropriate for its best-effort status. Traffic
then exits the Scheduler in a single orderly stream. then exits the Scheduler in a single orderly stream.
The interconnections of the TCB elements illustrated in Figures 6 and 7 The interconnections of the TCB elements illustrated in Figures 6 and 7
can be represented as follows: can be represented textually as follows:
TCB1: TCB1:
Classifier1: Classifier1:
FilterA: Meter1 FilterA: Meter1
FilterB: Dropper1
FilterC: Meter2
Default: Dropper3
Meter1:
Type: AverageRate
Profile: Profile1
ConformingOutput: Queue1
NonConformingOutput: Counter1
Counter1:
from Meter1 +-----+ from Meter1 +-----+
------------------------------->| |----+ ------------------------------->| |----+
| | | | | |
+-----+ | +-----+ |
Queue1 | Queue1 |
| +-----+ | +-----+
from Classifier1 +-----+ +-----+ +->|A | from Classifier1 +-----+ +-----+ +->|A |
---------------->| |------->| |------>|B |-------> ---------------->| |------->| |------>|B |------->
| | | | +--->|C | exiting | | | | +--->|C | exiting
skipping to change at page 38, line 30 skipping to change at page 38, line 4
+-----+ +-----+ | +-----+ +-----+ |
AlgDropper2 Queue3 | AlgDropper2 Queue3 |
| |
from Classifier1 +-----+ +-----+ | from Classifier1 +-----+ +-----+ |
---------------->| |------->| |----+ ---------------->| |------->| |----+
| | | | | | | |
+-----+ +-----+ +-----+ +-----+
AlgDropper3 Queue4 AlgDropper3 Queue4
Figure 7: An Example Traffic Conditioning Block (Part 2) Figure 7: An Example Traffic Conditioning Block (Part 2)
FilterB: Dropper1
FilterC: Meter2
Default: Dropper3
Meter1:
Type: AverageRate
Profile: Profile4
ConformingOutput: Queue1
NonConformingOutput: Counter1
Counter1:
Output: AbsoluteDropper1 Output: AbsoluteDropper1
Meter2: Meter2:
Type: AverageRate Type: AverageRate
Profile: Profile3 Profile: Profile3
ConformingOutput: Mux1.InputA ConformingOutput: Mux1.InputA
NonConformingOutput: Marker1 NonConformingOutput: Marker1
Marker1: Marker1:
Type: DSCPMarker Type: DSCPMarker
skipping to change at page 40, line 10 skipping to change at page 39, line 39
Priority: 40 Priority: 40
InputC: InputC:
MaxRateProfile: none MaxRateProfile: none
MinRateProfile: Profile3 MinRateProfile: Profile3
Priority: 20 Priority: 20
InputD: InputD:
MaxRateProfile: none MaxRateProfile: none
MinRateProfile: none MinRateProfile: none
Priority: 10 Priority: 10
8.2. An Example TCB to Support Multiple Customers 8.3. An Example TCB to Support Multiple Customers
The TCB described above can be installed on an ingress interface to The TCB described above can be installed on an ingress interface to
implement a provider/customer TCS if the interface is dedicated to the implement a provider/customer TCS if the interface is dedicated to the
customer. However, if a single interface is shared between multiple customer. However, if a single interface is shared between multiple
customers, then the TCB above will not suffice, since it does not customers, then the TCB above will not suffice, since it does not
differentiate among traffic from different customers. Its classification differentiate among traffic from different customers. Its classification
stage uses only BA classifiers. stage uses only BA classifiers.
The configuration is readily modified to support the case of multiple The configuration is readily modified to support the case of multiple
customers per interface, as follows. First, a TCB is defined for each customers per interface, as follows. First, a TCB is defined for each
skipping to change at page 40, line 21 skipping to change at page 40, line 4
The TCB described above can be installed on an ingress interface to The TCB described above can be installed on an ingress interface to
implement a provider/customer TCS if the interface is dedicated to the implement a provider/customer TCS if the interface is dedicated to the
customer. However, if a single interface is shared between multiple customer. However, if a single interface is shared between multiple
customers, then the TCB above will not suffice, since it does not customers, then the TCB above will not suffice, since it does not
differentiate among traffic from different customers. Its classification differentiate among traffic from different customers. Its classification
stage uses only BA classifiers. stage uses only BA classifiers.
The configuration is readily modified to support the case of multiple The configuration is readily modified to support the case of multiple
customers per interface, as follows. First, a TCB is defined for each customers per interface, as follows. First, a TCB is defined for each
customer to reflect the TCS with that customer: TCB1, defined above is customer to reflect the TCS with that customer: TCB1, defined above is
the TCB for customer 1 and elements are created for TCB2 and for TCB3 the TCB for customer 1. Similar elements are created for TCB2 and for
which reflect the agreements with customers 2 and 3 respectively. These TCB3 which reflect the agreements with customers 2 and 3 respectively.
3 TCBs may or may not share the same elements and parameters. These 3 TCBs may or may not contain similar elements and parameters.
Finally, a classifier is added to the front end to separate the traffic Finally, a classifier is added to the front end to separate the traffic
from the three different customers. This forms a new TCB, TCB4, which is from the three different customers. This forms a new TCB, TCB4, which is
illustrated in Figure 8. illustrated in Figure 8.
A representation of this multi-customer TCB might be: A representation of this multi-customer TCB might be:
TCB4: TCB4:
Classifier4: Classifier4:
submitted +-----+
traffic | A|--------> TCB1
--------->| B|--------> TCB2
| C|--------> TCB3
| X|------+ +-----+
+-----+ +-->| |
Classifier4 +-----+
AbsoluteDrop4
Figure 8: An Example of a Multi-Customer TCB
Filter1: to TCB1 Filter1: to TCB1
Filter2: to TCB2 Filter2: to TCB2
Filter3: to TCB3 Filter3: to TCB3
No Match: AbsoluteDropper4 No Match: AbsoluteDropper4
AbsoluteDropper4:
Type: AbsoluteDropper
TCB1: TCB1:
(as defined above) (as defined above)
TCB2: TCB2:
(similar to TCB1, perhaps with different (similar to TCB1, perhaps with different
elements or numeric parameters) elements or numeric parameters)
TCB3: TCB3:
(similar to TCB1, perhaps with different (similar to TCB1, perhaps with different
elements or numeric parameters) elements or numeric parameters)
submitted +-----+
traffic | A|--------> TCB1
--------->| B|--------> TCB2
| C|--------> TCB3
| X|------+ +-----+
+-----+ +-->| |
Classifier4 +-----+
AbsoluteDrop4
Figure 8: An Example of a Multi-Customer TCB
and the filters, based on each customer's source MAC address, could be and the filters, based on each customer's source MAC address, could be
defined as follows: defined as follows:
Filter1: Filter1:
Type: MacAddress Type: MacAddress
SrcValue: 01-02-03-04-05-06 (source MAC address of customer 1) SrcValue: 01-02-03-04-05-06 (source MAC address of customer 1)
SrcMask: FF-FF-FF-FF-FF-FF SrcMask: FF-FF-FF-FF-FF-FF
DestValue: 00-00-00-00-00-00 DestValue: 00-00-00-00-00-00
DestMask: 00-00-00-00-00-00 DestMask: 00-00-00-00-00-00
skipping to change at page 41, line 47 skipping to change at page 41, line 32
In this example, Classifier4 separates traffic submitted from different In this example, Classifier4 separates traffic submitted from different
customers based on the source MAC address in submitted packets. Those customers based on the source MAC address in submitted packets. Those
packets with recognized source MAC addresses are passed to the TCB packets with recognized source MAC addresses are passed to the TCB
implementing the TCS with the corresponding customer. Those packets with implementing the TCS with the corresponding customer. Those packets with
unrecognized source MAC addresses are passed to a dropper. unrecognized source MAC addresses are passed to a dropper.
TCB4 has a Classifier stage and an Action element stage performing TCB4 has a Classifier stage and an Action element stage performing
dropping of all unmatched traffic. dropping of all unmatched traffic.
8.3. TCBs Supporting Microflow-based Services 8.4. TCBs Supporting Microflow-based Services
The TCB illustrated above describes a configuration that might be The TCB illustrated above describes a configuration that might be
suitable for enforcing a SLS at a router's ingress. It assumes that the suitable for enforcing a SLS at a router's ingress. It assumes that the
customer marks its own traffic for the appropriate service level. It customer marks its own traffic for the appropriate service level. It
then limits the rate of aggregate traffic submitted at each service then limits the rate of aggregate traffic submitted at each service
level, thereby protecting the resources of the Diffserv network. It does level, thereby protecting the resources of the Diffserv network. It does
not provide any isolation between the customer's individual microflows. not provide any isolation between the customer's individual microflows.
A more complex example might be a TCB configuration that offers A more complex example might be a TCB configuration that offers
additional functionality to the customer. It recognizes individual additional functionality to the customer. It recognizes individual
customer microflows and marks each one independently. It also isolates customer microflows and marks each one independently. It also isolates
the customer's individual microflows from each other in order to prevent the customer's individual microflows from each other in order to prevent
a single microflow from seizing an unfair share of the resources a single microflow from seizing an unfair share of the resources
skipping to change at page 42, line 21 skipping to change at page 42, line 4
additional functionality to the customer. It recognizes individual additional functionality to the customer. It recognizes individual
customer microflows and marks each one independently. It also isolates customer microflows and marks each one independently. It also isolates
the customer's individual microflows from each other in order to prevent the customer's individual microflows from each other in order to prevent
a single microflow from seizing an unfair share of the resources a single microflow from seizing an unfair share of the resources
available to the customer at a certain service level. This is available to the customer at a certain service level. This is
illustrated in Figure 9. illustrated in Figure 9.
Suppose that the customer has an SLS which specifices 2 service levels, Suppose that the customer has an SLS which specifices 2 service levels,
to be identifed to the provider by DSCP A and DSCP B. Traffic is first to be identifed to the provider by DSCP A and DSCP B. Traffic is first
directed to a MF classifier which classifies traffic based on directed to a MF classifier which classifies traffic based on
miscellaneous classification criteria, to a granularity sufficient to
identify individual customer microflows. Each microflow can then be
+-----+ +-----+ +-----+ +-----+
Classifier1 | | | |---------------+ Classifier1 | | | |---------------+
(MF) +->| |-->| | +-----+ | (MF) +->| |-->| | +-----+ |
+-----+ | | | | |---->| | | +-----+ | | | | |---->| | |
| A|------ +-----+ +-----+ +-----+ | | A|------ +-----+ +-----+ +-----+ |
--->| B|-----+ Marker1 Meter1 Absolute | --->| B|-----+ Marker1 Meter1 Absolute |
| C|---+ | Dropper1 | +-----+ | C|---+ | Dropper1 | +-----+
| X|-+ | | +-----+ +-----+ +-->|A | | X|-+ | | +-----+ +-----+ +-->|A |
+-----+ | | | | | | |------------------>|B |---> +-----+ | | | | | | |------------------>|B |--->
| | +->| |-->| | +-----+ +-->|C | to TCB2 | | +->| |-->| | +-----+ +-->|C | to TCB2
skipping to change at page 43, line 5 skipping to change at page 42, line 29
| | | | | |---------------+ | | | | | |---------------+
| |--->| |-->| | +-----+ | |--->| |-->| | +-----+
| | | | |---->| | | | | | |---->| |
| +-----+ +-----+ +-----+ | +-----+ +-----+ +-----+
| Marker3 Meter3 Absolute | Marker3 Meter3 Absolute
| Dropper3 | Dropper3
V etc. V etc.
Figure 9: An Example of a Marking and Traffic Isolation TCB Figure 9: An Example of a Marking and Traffic Isolation TCB
miscellaneous classification criteria, to a granularity sufficient to
identify individual customer microflows. Each microflow can then be
marked for a specific DSCP The metering elements limit the contribution marked for a specific DSCP The metering elements limit the contribution
of each of the customer's microflows to the service level for which it of each of the customer's microflows to the service level for which it
was marked. Packets exceeding the allowable limit for the microflow are was marked. Packets exceeding the allowable limit for the microflow are
dropped. dropped.
This TCB could be formally specified as follows: This TCB could be formally specified as follows:
TCB1: TCB1:
Classifier1: (MF) Classifier1: (MF)
FilterA: Marker1 FilterA: Marker1
skipping to change at page 44, line 34 skipping to change at page 44, line 19
FilterB: Meter6 FilterB: Meter6
Meter5: Meter5:
ConformingOutput: Queue1 ConformingOutput: Queue1
NonConformingOutput: AbsoluteDropper4 NonConformingOutput: AbsoluteDropper4
Meter6: Meter6:
ConformingOutput: Queue2 ConformingOutput: Queue2
NonConformingOutput: AbsoluteDropper5 NonConformingOutput: AbsoluteDropper5
8.4. Cascaded TCBs 8.5. Cascaded TCBs
Nothing in this model prevents more complex scenarios in which one Nothing in this model prevents more complex scenarios in which one
microflow TCB precedes another (e.g. for TCBs implementing separate TCSs microflow TCB precedes another (e.g. for TCBs implementing separate TCSs
for the source and for a set of destinations). for the source and for a set of destinations).
9. Security Considerations 9. Security Considerations
Security vulnerabilities of Diffserv network operation are discussed in Security vulnerabilities of Diffserv network operation are discussed in
[DSARCH]. This document describes an abstract functional model of [DSARCH]. This document describes an abstract functional model of
Diffserv router elements. Certain denial-of-service attacks such as Diffserv router elements. Certain denial-of-service attacks such as
skipping to change at page 45, line 20 skipping to change at page 45, line 30
within the profile; the definition of the leaky-bucket scheduler is within the profile; the definition of the leaky-bucket scheduler is
conservative in that a packet is to be transmitted only if the whole conservative in that a packet is to be transmitted only if the whole
packet fits within the profile. This difference may be exploited by a packet fits within the profile. This difference may be exploited by a
malicious scheduler either to obtain QoS treatment for more octets than malicious scheduler either to obtain QoS treatment for more octets than
allowed in the TCS or to disrupt (perhaps only slightly) the QoS allowed in the TCS or to disrupt (perhaps only slightly) the QoS
guarantees promised to other traffic streams. guarantees promised to other traffic streams.
10. Acknowledgments 10. Acknowledgments
Concepts, terminology, and text have been borrowed liberally from Concepts, terminology, and text have been borrowed liberally from
[POLTERM], [DSMIB] and [DSPIB]. We wish to thank the authors of those [POLTERM], as well as from other IETF work on MIBs and policy-
documents: Fred Baker, Michael Fine, Keith McCloghrie, John Seligson, management. We wish to thank the authors of some of those documents:
Kwok Chan and Scott Hahn for their contributions. Fred Baker, Michael Fine, Keith McCloghrie, John Seligson, Kwok Chan,
Scott Hahn and Andrea Westerinen for their contributions.
This document has benefitted from the comments and suggestions of This document has benefitted from the comments and suggestions of
several participants of the Diffserv working group. several participants of the Diffserv working group, particularly John
Strassner and Walter Weiss.
11. References 11. References
[AF-PHB] [AF-PHB]
J. Heinanen, F. Baker, W. Weiss, and J. Wroclawski, "Assured J. Heinanen, F. Baker, W. Weiss, and J. Wroclawski, "Assured
Forwarding PHB Group", RFC 2597, June 1999. Forwarding PHB Group", RFC 2597, June 1999.
[DSARCH] [DSARCH]
M. Carlson, W. Weiss, S. Blake, Z. Wang, D. Black, and E. Davies, M. Carlson, W. Weiss, S. Blake, Z. Wang, D. Black, and E. Davies,
"An Architecture for Differentiated Services", RFC 2475, December "An Architecture for Differentiated Services", RFC 2475, December
1998 1998
[DSFIELD] [DSFIELD]
K. Nichols, S. Blake, F. Baker, and D. Black, "Definition of the K. Nichols, S. Blake, F. Baker, and D. Black, "Definition of the
Differentiated Services Field (DS Field) in the IPv4 and IPv6 Differentiated Services Field (DS Field) in the IPv4 and IPv6
Headers", RFC 2474, December 1998. Headers", RFC 2474, December 1998.
[DSMIB] [DSMIB]
F. Baker, A. Smith, K. Chan, "Differentiated Services MIB", F. Baker, A. Smith, K. Chan, "Differentiated Services MIB",
Internet Draft <http://www.ietf.org/internet-drafts/draft-ietf- Internet Draft <http://www.ietf.org/internet-drafts/draft-ietf-
diffserv-mib-04.txt>, July 2000. diffserv-mib-05.txt>, November 2000.
[DSPIB]
M. Fine, K. McCloghrie, J. Seligson, K. Chan, S. Hahn, and A.
Smith, "Quality of Service Policy Information Base", Internet Draft
<draft-ietf-diffserv-pib-00.txt>, March 2000.
[DSTERMS]
D. Grossman, "New Terminology for Diffserv", Internet Draft <draft-
ietf-diffserv-new-terms-02.txt>, November 1999.
[E2E] [E2E]
Y. Bernet, R. Yavatkar, P. Ford, F. Baker, L. Zhang, M. Speer, K. Y. Bernet, R. Yavatkar, P. Ford, F. Baker, L. Zhang, M. Speer, K.
Nichols, R. Braden, B. Davie, J. Wroclawski, and E. Felstaine, Nichols, R. Braden, B. Davie, J. Wroclawski, and E. Felstaine,
"Integrated Services Operation over Diffserv Networks", Internet "Integrated Services Operation over Diffserv Networks", Internet
Draft <http://www.ietf.org/internet-drafts/draft-ietf-issll- Draft <http://www.ietf.org/internet-drafts/draft-ietf-issll-
diffserv-rsvp-04.txt>, March 2000. diffserv-rsvp-04.txt>, March 2000.
[EF-PHB] [EF-PHB]
V. Jacobson, K. Nichols, and K. Poduri, "An Expedited Forwarding V. Jacobson, K. Nichols, and K. Poduri, "An Expedited Forwarding
skipping to change at page 46, line 36 skipping to change at page 46, line 39
[GTC] [GTC]
L. Lin, J. Lo, and F. Ou, "A Generic Traffic Conditioner", Internet L. Lin, J. Lo, and F. Ou, "A Generic Traffic Conditioner", Internet
Draft <http://www.ietf.org/internet-drafts/draft-lin-diffserv- Draft <http://www.ietf.org/internet-drafts/draft-lin-diffserv-
gtc-01.txt>, August 1999. gtc-01.txt>, August 1999.
[INTSERV] [INTSERV]
R. Braden, D. Clark and S. Shenker, "Integrated Services in the R. Braden, D. Clark and S. Shenker, "Integrated Services in the
Internet Architecture: an Overview" RFC 1633, June 1994. Internet Architecture: an Overview" RFC 1633, June 1994.
[POLTERM] [POLTERM]
F. Reichmeyer, D. Grossman, J. Strassner, M. Condell, "A Common A. Westerinen et al., "Policy Terminology", Internet Draft
Terminology for Policy Management", Internet Draft <http://www.ietf.org/internet-drafts/draft-ietf-policy-
<http://www.ietf.org/internet-drafts/draft-reichmeyer-polterm-
[QOSDEVMOD] [QOSDEVMOD]
J. Strassner, W. Weiss, D. Durham, A. Westerinen, "Information J. Strassner, A. Westerinen, B. Moore, "Information Model for
Model for Describing Network Device QoS Mechanisms", Internet Draft Describing Network Device QoS Mechanisms", Internet Draft
<http://www.ietf.org/internet-drafts/draft-ietf-policy-qos-device- <http://www.ietf.org/internet-drafts/draft-ietf-policy-qos-device-
[QUEUEMGMT] [QUEUEMGMT]
B. Braden et al., "Recommendations on Queue Management and B. Braden et al., "Recommendations on Queue Management and
Congestion Avoidance in the Internet", RFC 2309, April 1998. Congestion Avoidance in the Internet", RFC 2309, April 1998.
[SRTCM] [SRTCM]
J. Heinanen, and R. Guerin, "A Single Rate Three Color Marker", RFC J. Heinanen, and R. Guerin, "A Single Rate Three Color Marker", RFC
2697, September 1999. 2697, September 1999.
skipping to change at page 47, line 26 skipping to change at page 47, line 26
511-522. <ftp://ftp.ee.lbl.gov/papers/vic-mm95.ps.Z> 511-522. <ftp://ftp.ee.lbl.gov/papers/vic-mm95.ps.Z>
[802.1D] [802.1D]
"Information technology - Telecommunications and information "Information technology - Telecommunications and information
exchange between systems - Local and metropolitan area networks - exchange between systems - Local and metropolitan area networks -
Common specifications - Part 3: Media Access Control (MAC) Bridges: Common specifications - Part 3: Media Access Control (MAC) Bridges:
Revision. This is a revision of ISO/IEC 10038: 1993, 802.1j-1992 Revision. This is a revision of ISO/IEC 10038: 1993, 802.1j-1992
and 802.6k-1992. It incorporates P802.11c, P802.1p and P802.12e.", and 802.6k-1992. It incorporates P802.11c, P802.1p and P802.12e.",
ISO/IEC 15802-3: 1998. ISO/IEC 15802-3: 1998.
12. Appendix A. Simple Token Bucket Discussion and Definition 12. Appendix A. Discussion of Token Buckets and Leaky Buckets
The concept used for rate-control in several architectures, including
ATM, Frame Relay, Integrated Services and Differentiated Services,
consists of "leaky buckets" and/or "token buckets". Both of these are,
by definition, theoretical relationships between some defined
burst_size, rate and interval:
rate = burst_size/interval
Thus, a token bucket or leaky bucket might specify an information rate
of 1.2 Mbps with a burst size of 1500 bytes. In this case, the token
rate is 1,200,000 bits per second, the token burst is 12,000 bits and
the token interval is 10 milliseconds. The specification says that
conforming traffic will in the worst case come in 100 bursts per second
of 1500 bytes and at an average rate exceeding 1.2 Mbps.
A.1 Leaky Buckets
A leaky bucket algorithm is primarily used for shaping traffic as it
leaves an interface onto the network (handled under Queues and
Schedulers in this model). Traffic theoretically departs from an
interface at a rate of one bit every so many time units (in the example,
one bit every 0.83 microseconds) but, in fact, departs in multi-bit
units (packets) at a rate approximating the theoretical, as measured
over a longer interval. In the example, it might send one 1500 byte
packet every 10 ms or perhaps one 500 byte packet every 3.3 ms. It is
also possible to build multi-rate leaky buckets in which traffic departs
from the interface at varying rates depending on recent activity or
inactivity.
Implementations generally seek as constant a transmission rate as
achievable. In theory, a 10 Mbps shaped transmission stream from an
algorithmic implementation and a stream which is running at 10 Mbps
because its bottleneck link has been a 10 Mbps Ethernet link should be
indistinguishable. Depending on configuration, the approximation to
theoretical smoothness may vary by moving as much as an MTU from one
token interval to another. Traffic may also be jostled by other traffic
competing for the same transmission resources.
A.2 Token Buckets
A token bucket, on the other hand, measures the arrival rate of traffic
from another device. This traffic may originally have been shaped using
a leaky bucket shaper or its equivalent. The token bucket determines
whether the traffic (still) conforms to the specification. Multi-rate
token buckets (e.g. token buckets with both a peak rate and a mean rate,
and sometimes more) are commonly used, such as described in [SRTCM] and
[TRTCM]. In this case, absolute smoothness is not expected, but
conformance to one or more of the specified rates is.
Simplistically, a data stream is said to conform to a simple token
bucket parameterised by a {rate, burst_size} if the system receives in
any time interval, t, at most, an amount of data not exceeding (rate *
t) + burst_size.
For the multi-rate token bucket case, the data stream is said to conform
if, for each of the rates, the stream conforms to the token-bucket
profile appropriate for traffic of that class. For example, received
traffic that arrives pre-classified as one of the "excess" rates (e.g.
AF12 or AF13 traffic for a device implementing the AF1x PHB) is only
compared to the relevant "excess" token bucket profile.
A.3 Some consequences
When used as a leaky bucket shaper, the above definition interacts with
clock granularity in ways one might not expect. A leaky bucket releases
a packet only when all of its bits would have been allowed: it does not
borrow from future capacity. If the clock is very fine grain, on the
order of the bit rate or faster, this is not an issue. But if the clock
is relatively slow (and millisecond or multi-millisecond clocks are not
unusual in networking equipment), this can introduce jitter to the
shaped stream.
The fact that data is organized into variable length packets introduces
some uncertainty in the conformance decision made by a downstream Meter
that is attempting to determine conformance to a traffic profile.
Theoretically, in this case, a token bucket accepts a packet only if all
of its bits would have been accepted and does not borrow the required
excess capacity from future capacity - this is referred to as a "strict"
token bucket. This is consistent with [SRTCM] and [TRTCM]. In real-
world deployment, however, where MTUs are often larger than the burst
size offered by a link-layer network service provider and TCP is more
commonly ACK-paced than shaped using a leaky bucket, a "loose" or
"lenient" token bucket definition that would accept a packet if any of
its bits were within a profile offers a solution to the practical
problems that may arise from use of a strict meter.
Internet Protocol (IP) packets are of variable-length but theoretical Internet Protocol (IP) packets are of variable-length but theoretical
token buckets operate using fixed-length time intervals or pieces of token buckets operate using fixed-length time intervals or pieces of
data. This leaves an implementor of a token bucket scheme with a data. This leaves an implementor of a token bucket scheme with a
dilemma. When the amount of bandwidth tokens, TB, left in the token dilemma. When the amount of bandwidth tokens, TB, left in the token
bucket is positive but less than the size of the packet being operated bucket is positive but less than the size of the packet being operated
on, one of three things can be done: on, one of three things can be done:
(1) The whole size of the packet can be substracted from the bucket, (1) The whole size of the packet can be substracted from the bucket,
leaving it negative, remembering that the token bucket size must leaving it negative, remembering that the token bucket size must
skipping to change at page 49, line 10 skipping to change at page 50, line 50
between the end of that token interval and the actual arrival of between the end of that token interval and the actual arrival of
the packet are lost. As a result, natural jitter in the network the packet are lost. As a result, natural jitter in the network
conspires against the algorithm to reduce the actual acceptance conspires against the algorithm to reduce the actual acceptance
rate. Overcoming this error requires the maximum token bucket rate. Overcoming this error requires the maximum token bucket
size to be significantly greater than the MTU. size to be significantly greater than the MTU.
(3) Third, operationally, a strict token bucket is reasonable for (3) Third, operationally, a strict token bucket is reasonable for
traffic which has been shaped by a leaky bucket shaper or a traffic which has been shaped by a leaky bucket shaper or a
serial line. However, traffic in the Internet is rarely shaped in serial line. However, traffic in the Internet is rarely shaped in
that way. TCP applies no shaping to its traffic, but rather that way. TCP applies no shaping to its traffic, but rather
depends on longer-range Ack Clocking behavior to help it depends on longer-range ACK-clocking behavior to help it
approximate a certain rate and explicitly sends traffic bursts approximate a certain rate and explicitly sends traffic bursts
during slow start, retransmission and fast recovery. Video-on-IP during slow start, retransmission and fast recovery. Video-on-IP
implementations such as [VIC] may have a leaky bucket shaper implementations such as [VIC] may have a leaky bucket shaper
available to them, but often do not, and simply enqueue the available to them, but often do not, and simply enqueue the
output of their codec for transmission on the appropriate output of their codec for transmission on the appropriate
interface. As a result, in each of these cases, a strict shaper interface. As a result, in each of these cases, a strict shaper
may reject traffic in the short term (single token interval) may reject traffic in the short term (single token interval)
which it would have accepted if it had a longer time in view and which it would have accepted if it had a longer time in view and
which it needs to accept for the application to work properly. To which it needs to accept for the application to work properly. To
work around this, the token interval must approximate or exceed work around this, the token interval must approximate or exceed
the RTT of the session or sessions in question and the burst size the RTT of the session or sessions in question and the burst size
must accommodate the largest burst that the originator might must accommodate the largest burst that the originator might
send. send.
A.4 Mathematics
The behavior defined in [SRTCM] and [TRTCM] is not mandatory for The behavior defined in [SRTCM] and [TRTCM] is not mandatory for
compliance, but we give here a mathematical definition of two- parameter compliance, but we give here a mathematical definition of two- parameter
token bucket operation which is consistent with those documents and token bucket operation which is consistent with those documents and
which can be used to define a shaping profile. which can be used to define a shaping profile.
Define a token bucket with bucket size BS, token accumulation rate R and Define a token bucket with bucket size BS, token accumulation rate R and
instantaneous token occupancy T(t). Assume that T(0) = BS. instantaneous token occupancy T(t). Assume that T(0) = BS.
Then after an arbitrary interval with no packet arrivals, T(t) will not Then after an arbitrary interval with no packet arrivals, T(t) will not
change since the bucket is already full of tokens. Assume a packet of change since the bucket is already full of tokens. Assume a packet of
skipping to change at page 50, line 13 skipping to change at page 52, line 7
the packet does not conform and T(t) = T(t-). the packet does not conform and T(t) = T(t-).
This function can be used to define a shaping profile. If a packet of This function can be used to define a shaping profile. If a packet of
size C arrives at time t, it will be eligible for transmission at time size C arrives at time t, it will be eligible for transmission at time
te given as follows (we still assume C <= BS): te given as follows (we still assume C <= BS):
te = max { t, t" } te = max { t, t" }
where t" = (C - T(t') + t'*R)/R, T(t") = C, the time when C credits have where t" = (C - T(t') + t'*R)/R, T(t") = C, the time when C credits have
accumulated in the bucket, and when the packet would conform if the accumulated in the bucket, and when the packet would conform if the
token bucket were a meter. te != t" only if t != t". token bucket were a meter. te != t" only if t > t".
13. Authors' Addresses 13. Authors' Addresses
Yoram Bernet Yoram Bernet
Microsoft Microsoft
One Microsoft Way One Microsoft Way
Redmond, WA 98052 Redmond, WA 98052
Phone: +1 425 936 9568 Phone: +1 425 936 9568
E-mail: yoramb@microsoft.com E-mail: yoramb@microsoft.com
Andrew Smith
FAX: +1 414 345 1827
E-mail: ah_smith@pacbell.net
Steven Blake Steven Blake
Ericsson Ericsson
920 Main Campus Drive, Suite 500 920 Main Campus Drive, Suite 500
Raleigh, NC 27606 Raleigh, NC 27606
Phone: +1 919 472 9913 Phone: +1 919 472 9913
E-mail: slblake@torrentnet.com E-mail: slblake@torrentnet.com
Daniel Grossman Daniel Grossman
Motorola Inc. Motorola Inc.
20 Cabot Blvd. 20 Cabot Blvd.
Mansfield, MA 02048 Mansfield, MA 02048
Phone: +1 508 261 5312 Phone: +1 508 261 5312
E-mail: dan@dma.isg.mot.com E-mail: dan@dma.isg.mot.com
Andrew Smith (editor)
Allegro Networks
6399 San Ignacio Ave.
San Jose, CA 95119
FAX: +1 415 345 1827
E-mail: andrew@allegronetworks.com
Table of Contents Table of Contents
1 Introduction .................................................... 2 1 Introduction .................................................... 2
2 Glossary ........................................................ 3 2 Glossary ........................................................ 4
3 Conceptual Model ................................................ 5 3 Conceptual Model ................................................ 6
3.1 Elements of a Diffserv Router ................................. 5 3.1 Components of a Diffserv Router ............................... 6
3.1.1 Datapath .................................................... 6 3.1.1 Datapath .................................................... 6
3.1.2 Configuration and Management Interface ...................... 6 3.1.2 Configuration and Management Interface ...................... 8
3.1.3 Optional QoS Agent Module ................................... 8
3.2 Diffserv Functions at Ingress and Egress ...................... 9
3.1.3 Optional QoS Agent Module ................................... 7 3.3 Shaping and Policing .......................................... 10
3.2 Hierarchical Model of Diffserv Components ..................... 8 3.4 Hierarchical View of the Model ................................ 11
4 Classifiers ..................................................... 10 4 Classifiers ..................................................... 11
4.1 Definition .................................................... 10 4.1 Definition .................................................... 11
4.1.1 Filters ..................................................... 11 4.1.1 Filters ..................................................... 13
4.1.2 Overlapping Filters ......................................... 12 4.1.2 Overlapping Filters ......................................... 13
4.2 Examples ...................................................... 13 4.2 Examples ...................................................... 15
4.2.1 Behaviour Aggregate (BA) Classifier ......................... 13 4.2.1 Behaviour Aggregate (BA) Classifier ......................... 15
4.2.2 Multi-Field (MF) Classifier ................................. 14 4.2.2 Multi-Field (MF) Classifier ................................. 15
4.2.3 Free-form Classifier ........................................ 14 4.2.3 Free-form Classifier ........................................ 16
4.2.4 Other Possible Classifiers .................................. 15 4.2.4 Other Possible Classifiers .................................. 16
5 Meters .......................................................... 15 5 Meters .......................................................... 17
5.1 Token-Bucket Model ............................................ 17 5.1 Examples ...................................................... 18
5.2 Examples ...................................................... 18 5.1.1 Average Rate Meter .......................................... 18
5.2.1 Average Rate Meter .......................................... 18 5.1.2 Exponential Weighted Moving Average (EWMA) Meter ............ 19
5.2.2 Exponential Weighted Moving Average (EWMA) Meter ............ 19 5.1.3 Two-Parameter Token Bucket Meter ............................ 19
5.2.3 Two-Parameter Token Bucket Meter ............................ 20 5.1.4 Multi-Stage Token Bucket Meter .............................. 20
5.2.4 Multi-Stage Token Bucket Meter .............................. 20 5.1.5 Null Meter .................................................. 21
5.2.5 Null Meter .................................................. 21
6 Action Elements ................................................. 21 6 Action Elements ................................................. 21
6.1 Marker ........................................................ 22 6.1 DSCP Marker ................................................... 22
6.2 Absolute Dropper .............................................. 22 6.2 Absolute Dropper .............................................. 22
6.3 Multiplexor ................................................... 23 6.3 Multiplexor ................................................... 22
6.4 Counter ....................................................... 23 6.4 Counter ....................................................... 23
6.5 Null Action ................................................... 23 6.5 Null Action ................................................... 23
7 Queueing Elements ............................................... 24 7 Queueing Elements ............................................... 23
7.1 Queueing Model ................................................ 24 7.1 Queueing Model ................................................ 24
7.1.1 FIFO Queue .................................................. 25 7.1.1 FIFO Queue .................................................. 24
7.1.2 Scheduler ................................................... 26 7.1.2 Scheduler ................................................... 25
7.1.3 Algorithmic Dropper ......................................... 28 7.1.3 Algorithmic Dropper ......................................... 27
7.1.4 Constructing queueing blocks from the elements .............. 31 7.2 Sharing load among traffic streams using queueing ............. 31
7.2 Sharing load among traffic streams using queueing ............. 32 7.2.1 Load Sharing ................................................ 31
7.2.1 Load Sharing ................................................ 32 7.2.2 Traffic Priority ............................................ 32
7.2.2 Traffic Priority ............................................ 33 8 Traffic Conditioning Blocks (TCBs) .............................. 32
8 Traffic Conditioning Blocks (TCBs) .............................. 34 8.1 TCB ........................................................... 33
8.1 An Example TCB ................................................ 35 8.1.1 Building blocks for Queueing ................................ 34
8.2 An Example TCB to Support Multiple Customers .................. 40 8.2 An Example TCB ................................................ 34
8.3 TCBs Supporting Microflow-based Services ...................... 41 8.3 An Example TCB to Support Multiple Customers .................. 39
8.4 Cascaded TCBs ................................................. 44 8.4 TCBs Supporting Microflow-based Services ...................... 41
9 Security Considerations ......................................... 44 8.5 Cascaded TCBs ................................................. 44
9 Security Considerations ......................................... 45
10 Acknowledgments ................................................ 45 10 Acknowledgments ................................................ 45
11 References ..................................................... 45 11 References ..................................................... 45
12 Appendix A. Simple Token Bucket Discussion and Definition ...... 47 12 Appendix A. Discussion of Token Buckets and Leaky Buckets ...... 47
13 Authors' Addresses ............................................. 50 13 Authors' Addresses ............................................. 52
14. Full Copyright 14. Full Copyright
Copyright (C) The Internet Society (2000). All Rights Reserved. Copyright (C) The Internet Society (2000). All Rights Reserved.
This document and translations of it may be copied and furnished to This document and translations of it may be copied and furnished to
others, and derivative works that comment on or otherwise explain it others, and derivative works that comment on or otherwise explain it
or assist in its implmentation may be prepared, copied, published and or assist in its implmentation may be prepared, copied, published and
distributed, in whole or in part, without restriction of any kind, distributed, in whole or in part, without restriction of any kind,
provided that the above copyright notice and this paragraph are provided that the above copyright notice and this paragraph are
 End of changes. 165 change blocks. 
635 lines changed or deleted 699 lines changed or added

This html diff was produced by rfcdiff 1.34. The latest version is available from http://tools.ietf.org/tools/rfcdiff/