One document matched: draft-ietf-tsvwg-rtcweb-qos-09.xml
<?xml version="1.0" encoding="US-ASCII"?>
<!DOCTYPE rfc SYSTEM "rfc2629.dtd">
<?rfc toc="yes"?>
<?rfc tocompact="yes"?>
<?rfc tocdepth="3"?>
<?rfc tocindent="yes"?>
<?rfc symrefs="yes"?>
<?rfc sortrefs="yes"?>
<?rfc comments="yes"?>
<?rfc inline="yes"?>
<?rfc compact="yes"?>
<?rfc subcompact="yes"?>
<rfc category="std" docName="draft-ietf-tsvwg-rtcweb-qos-09"
ipr="trust200902">
<front>
<title abbrev="WebRTC QoS">DSCP and other packet markings for WebRTC
QoS</title>
<author fullname="Subha Dhesikan" initials="S." surname="Dhesikan">
<organization>Cisco Systems</organization>
<address>
<email>sdhesika@cisco.com</email>
</address>
</author>
<author fullname="Cullen Jennings" initials="C."
surname="Jennings">
<organization>Cisco Systems</organization>
<address>
<email>fluffy@cisco.com</email>
</address>
</author>
<author fullname="Dan Druta" initials="D." role="editor" surname="Druta">
<organization>AT&T</organization>
<address>
<email>dd5826@att.com</email>
</address>
</author>
<author fullname="Paul E. Jones" initials="P." surname="Jones">
<organization>Cisco Systems</organization>
<address>
<email>paulej@packetizer.com</email>
</address>
</author>
<date/>
<abstract>
<t>
Many networks, such as service provider and enterprise networks,
can provide treatment for individual packets based on
Differentiated Services Code Point (DSCP) values on a per-hop
basis. This document provides the recommended DSCP values for
web browsers to use for various classes of WebRTC traffic.
</t>
</abstract>
</front>
<middle>
<section title="Introduction">
<t>
Differentiated Services Code Points (DSCP) <xref target="RFC2474"/>
packet marking can help provide QoS in some environments.
This specification proposes how WebRTC applications can mark
packets, but does not contradict or redefine any
advice from previous IETF RFCs. Rather, it merely provides a
simple set of recommendations for implementers based on the
previous RFCs.
</t>
<t>
There are many use cases where such marking does not help, but it
seldom makes things worse if packets are marked appropriately.
As one example of where it does not help, if too many packets,
say all audio or all audio and video, are marked for a given
network condition then it can prevent desirable results. Either
too much other traffic will be starved, or there is not enough
capacity for the preferentially marked packets (i.e., audio
and/or video).
</t>
<t>
There are some environments where DSCP markings frequently help.
These include:
</t>
<t>
1. Private, wide-area networks.
</t>
<t>
2. Residential Networks. If the congested link is the broadband
uplink in a cable or DSL scenario, often residential routers/NAT
support preferential treatment based on DSCP.
</t>
<t>
3. Wireless Networks. If the congested link is a local wireless
network, marking may help.
</t>
<t>
Traditionally DSCP values have been thought of as being site
specific, with each site selecting its own code points for
controlling per-hop-behavior to influence the QoS for transport-layer
flows. However in the WebRTC use cases, the browsers need to set
them to something when there is no site specific information. In
this document, "browsers" is used synonymously with "Interactive User
Agent" as defined in the HTML specification,
<xref target="W3C.REC-html5-20141028"/>. This document
describes a subset of DSCP code point values drawn from existing
RFCs and common usage for use with WebRTC applications. These
code points are solely defaults.
</t>
<t>
This specification defines some inputs that the browser in a
WebRTC application can consider to aid in determining how to set
the various packet markings and defines the mapping from
abstract QoS policies (flow type, priority level) to those
packet markings.
</t>
</section>
<section title="Relation to Other Standards">
<t>
This document exists as a complement to
<xref target="RFC7657"/>, which describes the interaction between
DSCP and real-time communications. It covers the implications of
using various DSCP values, particularly focusing on Real-time
Transport Protocol (RTP) <xref target="RFC3550"/> streams that
are multiplexed onto a single transport-layer flow.
</t>
<t>
There are a number of guidelines specified in
<xref target="RFC7657"/> that should be followed when marking
traffic sent by WebRTC applications, as it is common for
multiple RTP streams to be multiplexed on the same
transport-layer flow. Generally, the RTP streams would be
marked with a value as appropriate from
<xref target="table-dscp"/>. A WebRTC application might also
multiplex data channel
<xref target="I-D.ietf-rtcweb-data-channel"/> traffic over the
same 5-tuple as RTP streams, which would also be marked as per
that table. The guidance in <xref target="RFC7657"/> says that
all data channel traffic would be marked with a single value
that is typically different than the value(s) used for RTP
streams multiplexed with the data channel traffic over the same
5-tuple, assuming RTP streams are marked with a value other than
default forwarding (DF). This is expanded upon further in the
next section.
</t>
<t>
This specification does not change or override the advice in any
other standards about setting packet markings. It simply selects
a subset of DSCP values that is relevant in the
WebRTC context. This document also specifies the inputs that
are needed by the browser to provide to the media engine.
</t>
<t>
The DSCP value set by the endpoint is not always trusted by
the network. Therefore, the DSCP value may be remarked at any
place in the network for a variety of reasons to any other DSCP
value, including default forwarding (DF) value to provide basic
best effort service. The mitigation for such action is through
an authorization mechanism. Such authorization mechanism is
outside the scope of this document. There is benefit in marking
traffic even if it only benefits the first few hops.
</t>
</section>
<section title="Terminology">
<t>
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL
NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and
"OPTIONAL" in this document are to be interpreted as described
in <xref target="RFC2119"/>.
</t>
</section>
<section title="Inputs">
<t>
WebRTC entities transmit and receive two types of media of
significance to this document:
<list style="symbols">
<t>
media flows which are RTP streams
<xref target="I-D.ietf-rtcweb-rtp-usage"/>
</t>
<t>
data flows which are data channels
<xref target="I-D.ietf-rtcweb-data-channel"/>
</t>
</list>
</t>
<t>
Each of the RTP streams and distinct data channels consists of
all of the packets associated with an independent media entity
and are not always equivalent to a transport-layer flow defined
by a 5-tuple (source address, destination address, source port,
destination port, and protocol). There may be multiple RTP
streams and data channels multiplexed over the same 5-tuple,
with each having a different level of importance to the
application and, therefore, potentially marked using different
DSCP values than another RTP stream or data channel within the
same transport-layer flow. (Note that there are restrictions
with respect to marking different data channels carried within
the same SCTP association as outlined in
<xref target="dscp-mappings"/>.)
</t>
<t>
The following are the inputs that the browser provides to the
media engine:
<list style="symbols">
<t>
Flow Type: The browser provides this input as it knows if the
flow is audio, interactive video with or without audio,
non-interactive video with or without audio, or data.
</t>
<t>
Application Priority: Another input is the relative
importance of an RTP stream or data channel. Many
applications have multiple flows of the same Flow Type and
often some flows are more important than others. For
example, in a video conference where there are usually audio
and video flows, the audio flow may be more important than
the video flow. JavaScript applications can tell the
browser whether a particular flow is high, medium, low or
very low importance to the application.
</t>
</list>
</t>
<t>
<xref target="I-D.ietf-rtcweb-transports"/> defines in more
detail what an individual flow is within the WebRTC
context.
</t>
</section>
<section anchor="dscp-mappings" title="DSCP Mappings">
<t>
The DSCP markings for each flow type of interest to WebRTC given
the application priority is shown in the following
table. The DSCP values for each flow type listed are a
reasonable subset of code point values taken from
<xref target="RFC4594"/>. A web browser SHOULD use these values
to mark the appropriate media packets. More information on EF
can be found in <xref target="RFC3246"/>. More information on
AF can be found in <xref target="RFC2597"/>. DF is default
forwarding which provides the basic best effort service.
</t>
<texttable anchor="table-dscp"
title="Recommended DSCP Values for WebRTC Applications">
<ttcol align="center">Flow Type</ttcol>
<ttcol align="center">Very Low</ttcol>
<ttcol align="center">Low</ttcol>
<ttcol align="center">Medium</ttcol>
<ttcol align="center">High</ttcol>
<c>Audio</c>
<c>CS1 (8)</c>
<c>DF (0)</c>
<c>EF (46)</c>
<c>EF (46)</c>
<c> </c>
<c> </c>
<c> </c>
<c> </c>
<c> </c>
<c>Interactive Video with or without audio</c>
<c>CS1 (8)</c>
<c>DF (0)</c>
<c>AF42, AF43 (36, 38)</c>
<c>AF41, AF42 (34, 36)</c>
<c> </c>
<c> </c>
<c> </c>
<c> </c>
<c> </c>
<c>Non-Interactive Video with or without audio</c>
<c>CS1 (8)</c>
<c>DF (0)</c>
<c>AF32, AF33 (28, 30)</c>
<c>AF31, AF32 (26, 28)</c>
<c> </c>
<c> </c>
<c> </c>
<c> </c>
<c> </c>
<c>Data</c>
<c>CS1 (8)</c>
<c>DF (0)</c>
<c>AF11</c>
<c>AF21</c>
</texttable>
<t>
The application priority, indicated by the columns "very low",
"low", "Medium", and "high", signifies the relative importance
of the flow within the application. It is an input that the
browser receives to assist it in selecting the DSCP value.
Application priority does not refer to priority in the network
transport.
</t>
<t>
The above table assumes that packets marked with CS1 are treated
as "less than best effort". However, the treatment of CS1 is
implementation dependent. If an implementation treats CS1 as
other than "less than best effort", then the actual priority
(or, more precisely, the per-hop-behavior) of the packets may be
changed from what is intended. It is common for CS1 to be
treated the same as DF, so anyone using CS1 cannot assume that
CS1 will be treated differently than DF. Implementers should
also note that excess EF traffic is dropped. This could
mean that a packet marked as EF may not get through as opposed
to a packet marked with a different DSCP value.
</t>
<t>
The browser SHOULD first select the flow type of the flow.
Within the flow type, the relative importance of the flow
SHOULD be used to select the appropriate DSCP value.
</t>
<t>
The combination of flow type and application priority provides
specificity and helps in selecting the right DSCP value for the
flow. All packets within a flow SHOULD have the same application
priority. In some cases, the selected application priority cell
may have multiple DSCP values, such as AF41 and AF42. These offer
different drop precedences. The different drop precedence
values provides additional granularity in classifying packets
within a flow. For example, in a video conference, the video
flow may have medium application priority. If so, either AF42
or AF43 may be selected. If the I-frames in the stream are more
important than the P-frames, then the I-frames can be marked
with AF42 and the P-frames marked with AF43.
</t>
<t>
For reasons discussed in Section 6 of <xref
target="RFC7657"/>, if multiple flows are
multiplexed using a reliable transport (e.g., TCP) then all of
the packets for all flows multiplexed over that
transport-layer flow MUST be marked using the same DSCP value.
Likewise, all WebRTC data channel packets transmitted over an
SCTP association MUST be marked using the same DSCP value,
regardless of how many data channels (streams) exist or what
kind of traffic is carried over the various SCTP streams. In the
event that the browser wishes to change the DSCP value in use
for an SCTP association, it MUST reset the SCTP congestion
controller after changing values. Frequent changes in the DSCP
value used for an SCTP association are discouraged, though, as
this would defeat any attempts at effectively managing
congestion. It should also be noted that any change in DSCP
value that results in a reset of the congestion controller puts
the SCTP association back into slow start, which may have
undesirable effects on application performance.
</t>
<t>
For the data channel traffic multiplexed over an SCTP
association, it is RECOMMENDED that the DSCP value selected be
the one associated with the highest priority requested for all
data channels multiplexed over the SCTP association. Likewise,
when multiplexing multiple flows over a TCP connection,
the DCSP value selected should be the one associated with the
highest priority requested for all multiplexed flows.
</t>
<t>
If a packet enters a QoS domain that has no support for the
above defined flow types/application priority (service class),
then the network node at the edge will remark the DSCP value
based on policies. This could result in the flow not
getting the network treatment it expects based on the original
DSCP value in the packet. Subsequently, if the packet enters a
QoS domain that supports a larger number of service classes,
there may not be sufficient information in the packet to restore
the original markings. Mechanisms for restoring such original
DSCP is outside the scope of this document.
</t>
<t>
In summary, there are no guarantees or promised level of service
with the use of DSCP. The service provided to a packet is
dependent upon the network design along the path, as well as the
congestion levels at every hop.
</t>
</section>
<section title="Security Considerations">
<t>
This specification does not add any additional security implication
other than the normal application use of DSCP. For security
implications on use of DSCP, please refer to Section 6 of <xref
target="RFC4594"/>. Please also see <xref
target="I-D.ietf-rtcweb-security"/> as an additional reference.
</t>
</section>
<section title="IANA Considerations">
<t>
This specification does not require any actions from IANA.
</t>
</section>
<section title="Downward References">
<t>
This specification contains a downwards reference to <xref
target="RFC4594"/>. However, the parts of that RFC used by this
specification are sufficiently stable for this downward
reference.
</t>
</section>
<section title="Acknowledgements">
<t>
Thanks To David Black, Magnus Westerland, Paolo Severini, Jim
Hasselbrook, Joe Marcus, Erik Nordmark, and Michael Tuexen for
their invaluable input.
</t>
</section>
<section title="Dedication">
<t>
This document is dedicated to the memory of James Polk, a
long-time friend and colleague. James made important
contributions to this specification, including being one of its
primary authors. The IETF global community mourns his loss and
he will be missed dearly.
</t>
</section>
<section title="Document History">
<t>
Note to RFC Editor: Please remove this section.
</t>
<t>
This document was originally an individual submission in RTCWeb WG.
The RTCWeb working group selected it to be become a WG document.
Later the transport ADs requested that this be moved to the TSVWG WG
as that seemed to be a better match.
</t>
</section>
</middle>
<back>
<references title="Normative References">
<?rfc include='reference.RFC.4594'?>
<?rfc include='reference.RFC.2119'?>
<?rfc include='reference.RFC.7657'?>
<?rfc include='reference.I-D.ietf-rtcweb-security'?>
<?rfc include='reference.I-D.ietf-rtcweb-transports'?>
<?rfc include='reference.I-D.ietf-rtcweb-data-channel'?>
<?rfc include='reference.I-D.ietf-rtcweb-rtp-usage'?>
</references>
<references title="Informative References">
<?rfc include='reference.RFC.3246'?>
<?rfc include='reference.RFC.2474'?>
<?rfc include='reference.RFC.2597'?>
<?rfc include='reference.RFC.3550'?>
<?rfc include='reference.W3C.REC-html5-20141028.xml'?>
</references>
</back>
</rfc>
| PAFTECH AB 2003-2026 | 2026-04-23 14:47:11 |