Rfc | 3450 |
Title | Asynchronous Layered Coding (ALC) Protocol Instantiation |
Author | M. Luby,
J. Gemmell, L. Vicisano, L. Rizzo, J. Crowcroft |
Date | December 2002 |
Format: | TXT, HTML |
Obsoleted by | RFC5775 |
Status: | EXPERIMENTAL |
|
Network Working Group M. Luby
Request for Comments: 3450 Digital Fountain
Category: Experimental J. Gemmell
Microsoft
L. Vicisano
Cisco
L. Rizzo
Univ. Pisa
J. Crowcroft
Cambridge Univ.
December 2002
Asynchronous Layered Coding (ALC) Protocol Instantiation
Status of this Memo
This memo defines an Experimental Protocol for the Internet
community. It does not specify an Internet standard of any kind.
Discussion and suggestions for improvement are requested.
Distribution of this memo is unlimited.
Copyright Notice
Copyright (C) The Internet Society (2002). All Rights Reserved.
Abstract
This document describes the Asynchronous Layered Coding (ALC)
protocol, a massively scalable reliable content delivery protocol.
Asynchronous Layered Coding combines the Layered Coding Transport
(LCT) building block, a multiple rate congestion control building
block and the Forward Error Correction (FEC) building block to
provide congestion controlled reliable asynchronous delivery of
content to an unlimited number of concurrent receivers from a single
sender.
Table of Contents
1. Introduction.................................................2
1.1 Delivery service models...................................3
1.2 Scalability...............................................5
1.3 Environmental Requirements and Considerations.............6
2. Architecture Definition......................................8
2.1 LCT building block........................................9
2.2 Multiple rate congestion control building block..........10
2.3 FEC building block.......................................11
2.4 Session Description......................................13
2.5 Packet authentication building block.....................14
3. Conformance Statement.......................................14
4. Functionality Definition....................................14
4.1 Packet format used by ALC................................15
4.2 Detailed Example of Packet format used by ALC............16
4.3 Header-Extension Fields..................................23
4.4 Sender Operation.........................................26
4.5 Receiver Operation.......................................27
5. Security Considerations.....................................29
6. IANA Considerations.........................................31
7. Intellectual Property Issues................................31
8. Acknowledgments.............................................31
9. References..................................................31
Authors' Addresses.............................................33
Full Copyright Statement.......................................34
1. Introduction
This document describes a massively scalable reliable content
delivery protocol, Asynchronous Layered Coding (ALC), for multiple
rate congestion controlled reliable content delivery. The protocol
is specifically designed to provide massive scalability using IP
multicast as the underlying network service. Massive scalability in
this context means the number of concurrent receivers for an object
is potentially in the millions, the aggregate size of objects to be
delivered in a session ranges from hundreds of kilobytes to hundreds
of gigabytes, each receiver can initiate reception of an object
asynchronously, the reception rate of each receiver in the session is
the maximum fair bandwidth available between that receiver and the
sender, and all of this can be supported using a single sender.
Because ALC is focused on reliable content delivery, the goal is to
deliver objects as quickly as possible to each receiver while at the
same time remaining network friendly to competing traffic. Thus, the
congestion control used in conjunction with ALC should strive to
maximize use of available bandwidth between receivers and the sender
while at the same time backing off aggressively in the face of
competing traffic.
The sender side of ALC consists of generating packets based on
objects to be delivered within the session and sending the
appropriately formatted packets at the appropriate rates to the
channels associated with the session. The receiver side of ALC
consists of joining appropriate channels associated with the session,
performing congestion control by adjusting the set of joined channels
associated with the session in response to detected congestion, and
using the packets to reliably reconstruct objects. All information
flow in an ALC session is in the form of data packets sent by a
single sender to channels that receivers join to receive data.
ALC does specify the Session Description needed by receivers before
they join a session, but the mechanisms by which receivers obtain
this required information is outside the scope of ALC. An
application that uses ALC may require that receivers report
statistics on their reception experience back to the sender, but the
mechanisms by which receivers report back statistics is outside the
scope of ALC. In general, ALC is designed to be a minimal protocol
instantiation that provides reliable content delivery without
unnecessary limitations to the scalability of the basic protocol.
This document is a product of the IETF RMT WG and follows the general
guidelines provided in RFC 3269 [8].
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 BCP 14, RFC 2119 [2].
Statement of Intent
This memo contains part of the definitions necessary to fully
specify a Reliable Multicast Transport protocol in accordance with
RFC2357. As per RFC2357, the use of any reliable multicast
protocol in the Internet requires an adequate congestion control
scheme.
While waiting for such a scheme to be available, or for an
existing scheme to be proven adequate, the Reliable Multicast
Transport working group (RMT) publishes this Request for Comments
in the "Experimental" category.
It is the intent of RMT to re-submit this specification as an IETF
Proposed Standard as soon as the above condition is met.
1.1 Delivery service models
ALC can support several different reliable content delivery service
models. Some examples are briefly described here.
Push service model.
A push model is a sender initiated concurrent delivery of objects to
a selected set of receivers. A push service model can be used for
example for reliable delivery of a large object such as a 100 GB
file. The sender could send a Session Description announcement to a
control channel and receivers could monitor this channel and join a
session whenever a Session Description of interest arrives. Upon
receipt of the Session Description, each receiver could join the
session to receive packets until enough packets have arrived to
reconstruct the object, at which point the receiver could report back
to the sender that its reception was completed successfully. The
sender could decide to continue sending packets for the object to the
session until all receivers have reported successful reconstruction
or until some other condition has been satisfied. In this example,
the sender uses ALC to generate packets based on the object and send
packets to channels associated with the session, and the receivers
use ALC to receive packets from the session and reconstruct the
object.
There are several features ALC provides to support the push model.
For example, the sender can optionally include an Expected Residual
Time (ERT) in the packet header that indicates the expected remaining
time of packet transmission for either the single object carried in
the session or for the object identified by the Transmission Object
Identifier (TOI) if there are multiple objects carried in the
session. This can be used by receivers to determine if there is
enough time remaining in the session to successfully receive enough
additional packets to recover the object. If for example there is
not enough time, then the push application may have receivers report
back to the sender to extend the transmission of packets for the
object for enough time to allow the receivers to obtain enough
packets to reconstruct the object. The sender could then include an
ERT based on the extended object transmission time in each subsequent
packet header for the object. As other examples, the LCT header
optionally can contain a Close Session flag that indicates when the
sender is about to end sending packet to the session and a Close
Object flag that indicates when the sender is about to end sending
packets to the session for the object identified by the Transmission
Object ID. However, these flags are not a completely reliable
mechanism and thus the Close Session flag should only be used as a
hint of when the session is about to close and the Close Object flag
should only be used as a hint of when transmission of packets for the
object is about to end.
The push model is particularly attractive in satellite networks and
wireless networks. In these environments a session may include one
channel and a sender may send packets at a fixed rate to this
channel, but sending at a fixed rate without congestion control is
outside the scope of this document.
On-demand content delivery model.
For an on-demand content delivery service model, senders typically
transmit for some given time period selected to be long enough to
allow all the intended receivers to join the session and recover a
single object. For example a popular software update might be
transmitted using ALC for several days, even though a receiver may be
able to complete the download in one hour total of connection time,
perhaps spread over several intervals of time. In this case the
receivers join the session at any point in time when it is active.
Receivers leave the session when they have received enough packets to
recover the object. The receivers, for example, obtain a Session
Description by contacting a web server.
Other service models.
There may be other reliable content delivery service models that can
be supported by ALC. The description of the potential applications,
the appropriate delivery service model, and the additional mechanisms
to support such functionalities when combined with ALC is beyond the
scope of this document.
1.2 Scalability
Massive scalability is a primary design goal for ALC. IP multicast
is inherently massively scalable, but the best effort service that it
provides does not provide session management functionality,
congestion control or reliability. ALC provides all of this on top
of IP multicast without sacrificing any of the inherent scalability
of IP multicast. ALC has the following properties:
o To each receiver, it appears as if though there is a dedicated
session from the sender to the receiver, where the reception rate
adjusts to congestion along the path from sender to receiver.
o To the sender, there is no difference in load or outgoing rate if
one receiver is joined to the session or a million (or any number
of) receivers are joined to the session, independent of when the
receivers join and leave.
o No feedback packets are required from receivers to the sender.
o Almost all packets in the session that pass through a bottleneck
link are utilized by downstream receivers, and the session shares
the link with competing flows fairly in proportion to their
utility.
Thus, ALC provides a massively scalable content delivery transport
that is network friendly.
ALC intentionally omits any application specific features that could
potentially limit its scalability. By doing so, ALC provides a
minimal protocol that is massively scalable. Applications may be
built on top of ALC to provide additional features that may limit the
scalability of the application. Such applications are outside the
scope of this document.
1.3 Environmental Requirements and Considerations
All of the environmental requirements and considerations that apply
to the LCT building block [11], the FEC building block [10], the
multiple rate congestion control building block and to any additional
building blocks that ALC uses also apply to ALC.
ALC requires connectivity between a sender and receivers, but does
not require connectivity from receivers to a sender. ALC inherently
works with all types of networks, including LANs, WANs, Intranets,
the Internet, asymmetric networks, wireless networks, and satellite
networks. Thus, the inherent raw scalability of ALC is unlimited.
However, ALC requires receivers to obtain the Session Description
out-of-band before joining a session and some implementations of this
may limit scalability.
If a receiver is joined to multiple ALC sessions then the receiver
MUST be able to uniquely identify and demultiplex packets to the
correct session. The Transmission Session Identifier (TSI) that MUST
appear in each packet header is used for this purpose. The TSI is
scoped by the IP address of the sender, and the IP address of the
sender together with the TSI uniquely identify the session. Thus,
the demultiplexing MUST be done on the basis of the IP address of the
sender and the TSI of the session from that sender.
ALC is presumed to be used with an underlying IP multicast network or
transport service that is a "best effort" service that does not
guarantee packet reception, packet reception order, and which does
not have any support for flow or congestion control. There are
currently two models of multicast delivery, the Any-Source Multicast
(ASM) model as defined in RFC 1112 [3] and the Source-Specific
Multicast (SSM) model as defined in [7]. ALC works with both
multicast models, but in a slightly different way with somewhat
different environmental concerns. When using ASM, a sender S sends
packets to a multicast group G, and an ALC channel address consists
of the pair (S,G), where S is the IP address of the sender and G is a
multicast group address. When using SSM, a sender S sends packets to
an SSM channel (S,G), and an ALC channel address coincides with the
SSM channel address.
A sender can locally allocate unique SSM channel addresses, and this
makes allocation of ALC channel addresses easy with SSM. To allocate
ALC channel addresses using ASM, the sender must uniquely choose the
ASM multicast group address across the scope of the group, and this
makes allocation of ALC channel addresses more difficult with ASM.
ALC channels and SSM channels coincide, and thus the receiver will
only receive packets sent to the requested ALC channel. With ASM,
the receiver joins an ALC channel by joining a multicast group G, and
all packets sent to G, regardless of the sender, may be received by
the receiver. Thus, SSM has compelling security advantages over ASM
for prevention of denial of service attacks. In either case,
receivers SHOULD use mechanisms to filter out packets from unwanted
sources.
Other issues specific to ALC with respect to ASM is the way the
multiple rate congestion control building block interacts with ASM.
The congestion control building block may use the measured difference
in time between when a join to a channel is sent and when the first
packet from the channel arrives in determining the receiver reception
rate. The congestion control building block may also uses packet
sequence numbers per channel to measure losses, and this is also used
to determine the receiver reception rate. These features raise two
concerns with respect to ASM: The time difference between when the
join to a channel is sent and when the first packet arrives can be
significant due to the use of Rendezvous Points (RPs) and the MSDP
protocol, and packets can be lost in the switch over from the (*,G)
join to the RP and the (S,G) join directly to the sender. Both of
these issues could potentially substantially degrade the reception
rate of receivers. To ameliorate these concerns, it is RECOMMENDED
that the RP be as close to the sender as possible. SSM does not
share these same concerns. For a fuller consideration of these
issues, consult the multiple rate congestion control building block.
Some networks are not amenable to some congestion control protocols
that could be used with ALC. In particular, for a satellite or
wireless network, there may be no mechanism for receivers to
effectively reduce their reception rate since there may be a fixed
transmission rate allocated to the session.
ALC is compatible with either IPv4 or IPv6 as no part of the packet
is IP version specific.
2. Architecture Definition
ALC uses the LCT building block [11] to provide in-band session
management functionality. ALC uses a multiple rate congestion
control building block that is compliant with RFC 2357 [12] to
provide congestion control that is feedback free. Receivers adjust
their reception rates individually by joining and leaving channels
associated with the session. ALC uses the FEC building block [10] to
provide reliability. The sender generates encoding symbols based on
the object to be delivered using FEC codes and sends them in packets
to channels associated with the session. Receivers simply wait for
enough packets to arrive in order to reliably reconstruct the object.
Thus, there is no request for retransmission of individual packets
from receivers that miss packets in order to assure reliable
reception of an object, and the packets and their rate of
transmission out of the sender can be independent of the number and
the individual reception experiences of the receivers.
The definition of a session for ALC is the same as it is for LCT. An
ALC session comprises multiple channels originating at a single
sender that are used for some period of time to carry packets
pertaining to the transmission of one or more objects that can be of
interest to receivers. Congestion control is performed over the
aggregate of packets sent to channels belonging to a session. The
fact that an ALC session is restricted to a single sender does not
preclude the possibility of receiving packets for the same objects
from multiple senders. However, each sender would be sending packets
to a a different session to which congestion control is individually
applied. Although receiving concurrently from multiple sessions is
allowed, how this is done at the application level is outside the
scope of this document.
ALC is a protocol instantiation as defined in RFC 3048 [16]. This
document describes version 1 of ALC which MUST use version 1 of LCT
described in [11]. Like LCT, ALC is designed to be used with the IP
multicast network service. This specification defines ALC as payload
of the UDP transport protocol [15] that supports IP multicast
delivery of packets. Future versions of this specification, or
companion documents may extend ALC to use the IP network layer
service directly. ALC could be used as the basis for designing a
protocol that uses a different underlying network service such as
unicast UDP, but the design of such a protocol is outside the scope
of this document.
An ALC packet header immediately follows the UDP header and consists
of the default LCT header that is described in [11] followed by the
FEC Payload ID that is described in [10]. The Congestion Control
Information field within the LCT header carries the required
Congestion Control Information that is described in the multiple rate
congestion control building block specified that is compliant with
RFC 2357 [12]. The packet payload that follows the ALC packet header
consists of encoding symbols that are identified by the FEC Payload
ID as described in [10].
Each receiver is required to obtain a Session Description before
joining an ALC session. As described later, the Session Description
includes out-of-band information required for the LCT, FEC and the
multiple rate congestion control building blocks. The FEC Object
Transmission Information specified in the FEC building block [10]
required for each object to be received by a receiver can be
communicated to a receiver either out-of-band or in-band using a
Header Extension. The means for communicating the Session
Description and the FEC Object Transmission Information to a receiver
is outside the scope of this document.
2.1 LCT building block
LCT requires receivers to be able to uniquely identify and
demultiplex packets associated with an LCT session, and ALC inherits
and strengthens this requirement. A Transport Session Identifier
(TSI) MUST be associated with each session and MUST be carried in the
LCT header of each ALC packet. The TSI is scoped by the sender IP
address, and the (sender IP address, TSI) pair MUST uniquely identify
the session.
The LCT header contains a Congestion Control Information (CCI) field
that MUST be used to carry the Congestion Control Information from
the specified multiple rate congestion control protocol. There is a
field in the LCT header that specifies the length of the CCI field,
and the multiple rate congestion control building block MUST uniquely
identify a format of the CCI field that corresponds to this length.
The LCT header contains a Codepoint field that MAY be used to
communicate to a receiver the settings for information that may vary
during a session. If used, the mapping between settings and
Codepoint values is to be communicated in the Session Description,
and this mapping is outside the scope of this document. For example,
the FEC Encoding ID that is part of the FEC Object Transmission
Information as specified in the FEC building block [10] could vary
for each object carried in the session, and the Codepoint value could
be used to communicate the FEC Encoding ID to be used for each
object. The mapping between FEC Encoding IDs and Codepoints could
be, for example, the identity mapping.
If more than one object is to be carried within a session then the
Transmission Object Identifier (TOI) MUST be used in the LCT header
to identify which packets are to be associated with which objects.
In this case the receiver MUST use the TOI to associate received
packets with objects. The TOI is scoped by the IP address of the
sender and the TSI, i.e., the TOI is scoped by the session. The TOI
for each object is REQUIRED to be unique within a session, but MAY
NOT be unique across sessions. Furthermore, the same object MAY have
a different TOI in different sessions. The mapping between TOIs and
objects carried in a session is outside the scope of this document.
If only one object is carried within a session then the TOI MAY be
omitted from the LCT header.
The default LCT header from version 1 of the LCT building block [11]
MUST be used.
2.2 Multiple rate congestion control building block
Implementors of ALC MUST implement a multiple rate feedback-free
congestion control building block that is in accordance to RFC 2357
[12]. Congestion control MUST be applied to all packets within a
session independently of which information about which object is
carried in each packet. Multiple rate congestion control is
specified because of its suitability to scale massively and because
of its suitability for reliable content delivery. The multiple rate
congestion control building block MUST specify in-band Congestion
Control Information (CCI) that MUST be carried in the CCI field of
the LCT header. The multiple rate congestion control building block
MAY specify more than one format, but it MUST specify at most one
format for each of the possible lengths 32, 64, 96 or 128 bits. The
value of C in the LCT header that determines the length of the CCI
field MUST correspond to one of the lengths for the CCI defined in
the multiple rate congestion control building block, this length MUST
be the same for all packets sent to a session, and the CCI format
that corresponds to the length as specified in the multiple rate
congestion control building block MUST be the format used for the CCI
field in the LCT header.
When using a multiple rate congestion control building block a sender
sends packets in the session to several channels at potentially
different rates. Then, individual receivers adjust their reception
rate within a session by adjusting which set of channels they are
joined to at each point in time depending on the available bandwidth
between the receiver and the sender, but independent of other
receivers.
2.3 FEC building block
The FEC building block [10] provides reliable object delivery within
an ALC session. Each object sent in the session is independently
encoded using FEC codes as described in [9], which provide a more
in-depth description of the use of FEC codes in reliable content
delivery protocols. All packets in an ALC session MUST contain an
FEC Payload ID in a format that is compliant with the FEC building
block [10]. The FEC Payload ID uniquely identifies the encoding
symbols that constitute the payload of each packet, and the receiver
MUST use the FEC Payload ID to determine how the encoding symbols
carried in the payload of the packet were generated from the object
as described in the FEC building block.
As described in [10], a receiver is REQUIRED to obtain the FEC Object
Transmission Information for each object for which data packets are
received from the session. The FEC Object Transmission Information
includes:
o The FEC Encoding ID.
o If an Under-Specified FEC Encoding ID is used then the FEC
Instance ID associated with the FEC Encoding ID.
o For each object in the session, the length of the object in
bytes.
o The additional required FEC Object Transmission Information for
the FEC Encoding ID as prescribed in the FEC building block [10].
For example, when the FEC Encoding ID is 128, the required FEC
Object Transmission Information is the number of source blocks
that the object is partitioned into and the length of each source
block in bytes.
Some of the FEC Object Transmission Information MAY be implicit based
on the implementation. As an example, source block lengths may be
derived by a fixed algorithm from the object length. As another
example, it may be that all source blocks are the same length and
this is what is passed out-of-band to the receiver. As another
example, it could be that the full sized source block length is
provided and this is the length used for all but the last source
block, which is calculated based on the full source block length and
the object length. As another example, it could be that the same FEC
Encoding ID and FEC Instance ID are always used for a particular
application and thus the FEC Encoding ID and FEC Instance ID are
implicitly defined.
Sometimes the objects that will be sent in a session are completely
known before the receiver joins the session, in which case the FEC
Object Transmission Information for all objects in the session can be
communicated to receivers before they join the session. At other
times the objects may not know when the session begins, or receivers
may join a session in progress and may not be interested in some
objects for which transmission has finished, or receivers may leave a
session before some objects are even available within the session.
In these cases, the FEC Object Transmission Information for each
object may be dynamically communicated to receivers at or before the
time packets for the object are received from the session. This may
be accomplished using either an out-of-band mechanism, in-band using
the Codepoint field or a Header Extension, or any combination of
these methods. How the FEC Object Transmission Information is
communicated to receivers is outside the scope of this document.
If packets for more than one object are transmitted within a session
then a Transmission Object Identifier (TOI) that uniquely identifies
objects within a session MUST appear in each packet header. Portions
of the FEC Object Transmission Information could be the same for all
objects in the session, in which case these portions can be
communicated to the receiver with an indication that this applies to
all objects in the session. These portions may be implicitly
determined based on the application, e.g., an application may use the
same FEC Encoding ID for all objects in all sessions. If there is a
portion of the FEC Object Transmission Information that may vary from
object to object and if this FEC Object Transmission Information is
communicated to a receiver out-of-band then the TOI for the object
MUST also be communicated to the receiver together with the
corresponding FEC Object Transmission Information, and the receiver
MUST use the corresponding FEC Object Transmission Information for
all packets received with that TOI. How the TOI and corresponding
FEC Object Transmission Information is communicated out-of-band to
receivers is outside the scope of this document.
It is also possible that there is a portion of the FEC Object
Transmission Information that may vary from object to object that is
carried in-band, for example in the CodePoint field or in Header
Extensions. How this is done is outside the scope of this document.
In this case the FEC Object Transmission Information is associated
with the object identified by the TOI carried in the packet.
2.4 Session Description
The Session Description that a receiver is REQUIRED to obtain before
joining an ALC session MUST contain the following information:
o The multiple rate congestion control building block to be used
for the session;
o The sender IP address;
o The number of channels in the session;
o The address and port number used for each channel in the session;
o The Transport Session ID (TSI) to be used for the session;
o An indication of whether or not the session carries packets for
more than one object;
o If Header Extensions are to be used, the format of these Header
Extensions.
o Enough information to determine the packet authentication scheme
being used, if it is being used.
How the Session Description is communicated to receivers is outside
the scope of this document.
The Codepoint field within the LCT portion of the header CAN be used
to communicate in-band some of the dynamically changing information
within a session. To do this, a mapping between Codepoint values and
the different dynamic settings MUST be included within the Session
Description, and then settings to be used are communicated via the
Codepoint value placed into each packet. For example, it is possible
that multiple objects are delivered within the same session and that
a different FEC encoding algorithm is used for different types of
objects. Then the Session Description could contain the mapping
between Codepoint values and FEC Encoding IDs. As another example,
it is possible that a different packet authentication scheme is used
for different packets sent to the session. In this case, the mapping
between the packet authentication scheme and Codepoint values could
be provided in the Session Description. Combinations of settings can
be mapped to Codepoint values as well. For example, a particular
combination of a FEC Encoding ID and a packet authentication scheme
could be associated with a Codepoint value.
The Session Description could also include, but is not limited to:
o The mappings between combinations of settings and Codepoint
values;
o The data rates used for each channel;
o The length of the packet payload;
o Any information that is relevant to each object being
transported, such as the Object Transmission Information for each
object, when the object will be available within the session and
for how long.
The Session Description could be in a form such as SDP as defined in
RFC 2327 [5], or XML metadata as defined in RFC 3023 [13], or
HTTP/Mime headers as defined in RFC 2068 [4], etc. It might be
carried in a session announcement protocol such as SAP as defined in
RFC 2974 [6], obtained using a proprietary session control protocol,
located on a web page with scheduling information, or conveyed via
E-mail or other out-of-band methods. Discussion of Session
Description formats and methods for communication of Session
Descriptions to receivers is beyond the scope of this document.
2.5 Packet authentication building block
It is RECOMMENDED that implementors of ALC use some packet
authentication scheme to protect the protocol from attacks. An
example of a possibly suitable scheme is described in [14]. Packet
authentication in ALC, if used, is to be integrated through the
Header Extension support for packet authentication provided in the
LCT building block.
3. Conformance Statement
This Protocol Instantiation document, in conjunction with the LCT
building block [11], the FEC building block [10] and with a multiple
rate congestion control building block completely specifies a working
reliable multicast transport protocol that conforms to the
requirements described in RFC 2357 [12].
4. Functionality Definition
This section describes the format and functionality of the data
packets carried in an ALC session as well as the sender and receiver
operations for a session.
4.1 Packet format used by ALC
The packet format used by ALC is the UDP header followed by the
default LCT header followed by the FEC Payload ID followed by the
packet payload. The default LCT header is described in the LCT
building block [11] and the FEC Payload ID is described in the FEC
building block [10]. The Congestion Control Information field in the
LCT header contains the REQUIRED Congestion Control Information that
is described in the multiple rate congestion control building block
used. The packet payload contains encoding symbols generated from an
object. If more than one object is carried in the session then the
Transmission Object ID (TOI) within the LCT header MUST be used to
identify which object the encoding symbols are generated from.
Within the scope of an object, encoding symbols carried in the
payload of the packet are identified by the FEC Payload ID as
described in the FEC building block.
The version number of ALC specified in this document is 1. This
coincides with version 1 of the LCT building block [11] used in this
specification. The LCT version number field should be interpreted as
the ALC version number field.
The overall ALC packet format is depicted in Figure 1. The packet is
an IP packet, either IPv4 or IPv6, and the IP header precedes the UDP
header. The ALC packet format has no dependencies on the IP version
number. The default LCT header MUST be used by ALC and this default
is described in detail in the LCT building block [11].
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| UDP header |
| |
+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+
| Default LCT header |
| |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| FEC Payload ID |
| |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Encoding Symbol(s) |
| ... |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Figure 1 - Overall ALC packet format
In some special cases an ALC sender may need to produce ALC packets
that do not contain any payload. This may be required, for example,
to signal the end of a session or to convey congestion control
information. These data-less packets do not contain the FEC Payload
ID either, but only the LCT header fields. The total datagram
length, conveyed by outer protocol headers (e.g., the IP or UDP
header), enables receivers to detect the absence of the ALC payload
and FEC Payload ID.
4.2 Detailed Example of Packet format used by ALC
A detailed example of an ALC packet starting with the LCT header is
shown in Figure 2. In the example, the LCT header is the first 5
32-bit words, the FEC Payload ID is the next 2 32-bit words, and the
remainder of the packet is the payload.
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| 1 | 0 | 0 |1| 1 |0|1|0|0|0| 5 | 128 |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Congestion Control Information (CCI, length = 32 bits) |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Transport Session Identifier (TSI, length = 32 bits) |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Transport Object Identifier (TOI, length = 32 bits) |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Sender Current Time |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Source Block Number |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Encoding Symbol ID |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Encoding Symbol(s) |
| ... |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Figure 2 - A detailed example of the ALC packet format
The LCT portion of the overall ALC packet header is of variable size,
which is specified by a length field in the third byte of the header.
All integer fields are carried in "big-endian" or "network order"
format, that is, most significant byte (octet) first. Bits
designated as "padding" or "reserved" (r) MUST by set to 0 by senders
and ignored by receivers. Unless otherwise noted, numeric constants
in this specification are in decimal (base 10).
The function and length and particular setting of the value for each
field in this detailed example of the header is the following,
described in the order of their appearance in the header.
ALC version number (V): 4 bits
Indicates the ALC version number.
The ALC version number for this specification is 1 as shown.
This is also the LCT version number.
Congestion control flag (C): 2 bits
The Congestion Control Information (CCI) field specified by the
multiple rate congestion control building block is a multiple
of 32-bits in length. The multiple rate congestion control
building block MUST specify a format for the CCI. The
congestion control building block MAY specify formats for
different CCI lengths, where the set of possible lengths is 32,
64, 96 or 128 bits. The value of C MUST match the length of
exactly one of the possible formats for the congestion control
building block, and this format MUST be used for the CCI field.
The value of C MUST be the same for all packets sent to a
session.
C=0 indicates the 32-bit CCI field format is to be used.
C=1 indicates the 64-bit CCI field format is to be used.
C=2 indicates the 96-bit CCI field format is to be used.
C=3 indicates the 128-bit CCI field format is to be used.
In the example C=0 indicates that a 32-bit format is to be
used.
Reserved (r): 2 bits
Reserved for future use. A sender MUST set these bits to zero
and a receiver MUST ignore these bits.
As required, these bits are set to 0 in the example.
Transport Session Identifier flag (S): 1 bit
This is the number of full 32-bit words in the TSI field. The
TSI field is 32*S + 16*H bits in length. For ALC the length of
the TSI field is REQUIRED to be non-zero. This implies that
the setting S=0 and H=0 MUST NOT be used.
In the example S=1 and H=0, and thus the TSI is 32-bits in
length.
Transport Object Identifier flag (O): 2 bits
This is the number of full 32-bit words in the TOI field. The
TOI field is 32*O + 16*H bits in length. If more than one
object is to be delivered in the session then the TOI MUST be
used, in which case the setting O=0 and H=0 MUST NOT be used.
In the example O=1 and H=0, and thus the TOI is 32-bits in
length.
Half-word flag (H): 1 bit
The TSI and the TOI fields are both multiples of 32-bits plus
16*H bits in length. This allows the TSI and TOI field lengths
to be multiples of a half-word (16 bits), while ensuring that
the aggregate length of the TSI and TOI fields is a multiple of
32-bits.
In the example H=0 which indicates that both TSI and TOI are
both multiples of 32-bits in length.
Sender Current Time present flag (T): 1 bit
T = 0 indicates that the Sender Current Time (SCT) field is not
present.
T = 1 indicates that the SCT field is present. The SCT is
inserted by senders to indicate to receivers how long the
session has been in progress.
In the example T=1, which indicates that the SCT is carried in
this packet.
Expected Residual Time present flag (R): 1 bit
R = 0 indicates that the Expected Residual Time (ERT) field is
not present.
R = 1 indicates that the ERT field is present.
The ERT is inserted by senders to indicate to receivers how
much longer packets will be sent to the session for either the
single object carried in the session or for the object
identified by the TOI if there are multiple objects carried in
the session. Senders MUST NOT set R = 1 when the ERT for the
object is more than 2^32-1 time units (approximately 49 days),
where time is measured in units of milliseconds.
In the example R=0, which indicates that the ERT is not carried
in this packet.
Close Session flag (A): 1 bit
Normally, A is set to 0. The sender MAY set A to 1 when
termination of transmission of packets for the session is
imminent. A MAY be set to 1 in just the last packet
transmitted for the session, or A MAY be set to 1 in the last
few seconds of packets transmitted for the session. Once the
sender sets A to 1 in one packet, the sender SHOULD set A to 1
in all subsequent packets until termination of transmission of
packets for the session. A received packet with A set to 1
indicates to a receiver that the sender will immediately stop
sending packets for the session. When a receiver receives a
packet with A set to 1 the receiver SHOULD assume that no more
packets will be sent to the session.
In the example A=0, and thus this packet does not indicate the
close of the session.
Close Object flag (B): 1 bit
Normally, B is set to 0. The sender MAY set B to 1 when
termination of transmission of packets for an object is
imminent. If the TOI field is in use and B is set to 1 then
termination of transmission for the object identified by the
TOI field is imminent. If the TOI field is not in use and B is
set to 1 then termination of transmission for the one object in
the session identified by out-of-band information is imminent.
B MAY be set to 1 in just the last packet transmitted for the
object, or B MAY be set to 1 in the last few seconds packets
transmitted for the object. Once the sender sets B to 1 in one
packet for a particular object, the sender SHOULD set B to 1 in
all subsequent packets for the object until termination of
transmission of packets for the object. A received packet with
B set to 1 indicates to a receiver that the sender will
immediately stop sending packets for the object. When a
receiver receives a packet with B set to 1 then it SHOULD
assume that no more packets will be sent for the object to the
session.
In the example B=0, and thus this packet does not indicate the
end of sending data packets for the object.
LCT header length (HDR_LEN): 8 bits
Total length of the LCT header in units of 32-bit words. The
length of the LCT header MUST be a multiple of 32-bits. This
field can be used to directly access the portion of the packet
beyond the LCT header, i.e., the FEC Payload ID if the packet
contains a payload, or the end of the packet if the packet
contains no payload.
In the example HDR_LEN=5 to indicate that the length of the LCT
header portion of the overall ALC is 5 32-bit words.
Codepoint (CP): 8 bits
This field is used by ALC to carry the mapping that identifies
settings for portions of the Session Description that can
change within the session. The mapping between Codepoint
values and the settings for portions of the Session Description
is to be communicated out-of-band.
In the example the portion of the Session Description that can
change within the session is the FEC Encoding ID, and the
identity mapping is used between Codepoint values and FEC
Encoding IDs. Thus, CP=128 identifies FEC Encoding ID 128, the
"Small Block, Large Block and Expandable FEC Codes" as
described in the FEC building block [10]. The FEC Payload ID
associated with FEC Encoding ID 128 is 64-bits in length.
Congestion Control Information (CCI): 32, 64, 96 or 128 bits
This is field contains the Congestion Control Information as
defined by the specified multiple rate congestion control
building block. The format of this field is determined by the
multiple rate congestion control building block.
This field MUST be 32 bits if C=0.
This field MUST be 64 bits if C=1.
This field MUST be 96 bits if C=2.
This field MUST be 128 bits if C=3.
In the example, the CCI is 32-bits in length. The format of
the CCI field for the example MUST correspond to the format for
the 32-bit version of the CCI specified in the multiple rate
congestion control building block.
Transport Session Identifier (TSI): 16, 32 or 48 bits
The TSI uniquely identifies a session among all sessions from a
particular sender. The TSI is scoped by the sender IP address,
and thus the (sender IP address, TSI) pair uniquely identify
the session. For ALC, the TSI MUST be included in the LCT
header.
The TSI MUST be unique among all sessions served by the sender
during the period when the session is active, and for a large
period of time preceding and following when the session is
active. A primary purpose of the TSI is to prevent receivers
from inadvertently accepting packets from a sender that belong
to sessions other than sessions receivers are subscribed to.
For example, suppose a session is deactivated and then another
session is activated by a sender and the two sessions use an
overlapping set of channels. A receiver that connects and
remains connected to the first session during this sender
activity could possibly accept packets from the second session
as belonging to the first session if the TSI for the two
sessions were identical. The mapping of TSI field values to
sessions is outside the scope of this document and is to be
done out-of-band.
The length of the TSI field is 32*S + 16*H bits. Note that the
aggregate lengths of the TSI field plus the TOI field is a
multiple of 32 bits.
In the example the TSI is 32 bits in length.
Transport Object Identifier (TOI): 0, 16, 32, 48, 64, 80, 96 or 112
bits.
This field indicates which object within the session this
packet pertains to. For example, a sender might send a number
of files in the same session, using TOI=0 for the first file,
TOI=1 for the second one, etc. As another example, the TOI may
be a unique global identifier of the object that is being
transmitted from several senders concurrently, and the TOI
value may be the output of a hash function applied to the
object. The mapping of TOI field values to objects is outside
the scope of this document and is to be done out-of-band. The
TOI field MUST be used in all packets if more than one object
is to be transmitted in a session, i.e., the TOI field is
either present in all the packets of a session or is never
present.
The length of the TOI field is 32*O + 16*H bits. Note that the
aggregate lengths of the TSI field plus the TOI field is a
multiple of 32 bits.
In the example the TOI is 32 bits in length.
Sender Current Time (SCT): 0 or 32 bits
This field represents the current clock of the sender at the
time this packet was transmitted, measured in units of 1ms and
computed modulo 2^32 units from the start of the session.
This field MUST NOT be present if T=0 and MUST be present if
T=1.
In this example the SCT is present.
Expected Residual Time (ERT): 0 or 32 bits
This field represents the sender expected residual transmission
time of packets for either the single object carried in the
session or for the object identified by the TOI if there are
multiple objects carried in the session.
This field MUST NOT be present if R=0 and MUST be present if
R=1.
In this example the ERT is not present.
FEC Payload ID: X bits
The length and format of the FEC Payload ID depends on the FEC
Encoding ID as described in the FEC building block [10]. The
FEC Payload ID format is determined by the FEC Encoding ID that
MUST be communicated in the Session Description. The Session
Description MAY specify that more than one FEC Encoding ID is
used in the session, in which case the Session Description MUST
contain a mapping that identifies which Codepoint values
correspond to which FEC Encoding IDs. This mapping, if used,
is outside the scope of this document.
The example packet format corresponds to the format for "Small
Block, Large Block and Expandable FEC Codes" as described in
the FEC building block, for which the associated FEC Encoding
ID 128. For FEC Encoding ID 128, the FEC Payload ID consists
of the following two fields that in total are X = 64 bits in
length:
Source Block Number (SBN): 32 bits
The Source Block Number identifies from which source block
of the object the encoding symbol(s) in the payload are
generated. These blocks are numbered consecutively from
0 to N-1, where N is the number of source blocks in the
object.
Encoding Symbol ID (ESI): 32 bits
The Encoding Symbol ID identifies which specific encoding
symbol(s) generated from the source block are carried in the
packet payload. The exact details of the correspondence
between Encoding Symbol IDs and the encoding symbol(s) in
the packet payload are dependent on the particular encoding
algorithm used as identified by the FEC Encoding ID and by
the FEC Instance ID.
Encoding Symbol(s): Y bits
The encoding symbols are what the receiver uses to reconstruct
an object. The total length Y of the encoding symbol(s) in the
packet can be determined by the receiver of the packet by
computing the total length of the received packet and
subtracting off the length of the headers.
4.3 Header-Extension Fields
Header Extensions can be used to extend the LCT header portion of the
ALC header to accommodate optional header fields that are not always
used or have variable size. Header Extensions are not used in the
example ALC packet format shown in the previous subsection. Examples
of the use of Header Extensions include:
o Extended-size versions of already existing header fields.
o Sender and Receiver authentication information.
The presence of Header Extensions can be inferred by the LCT header
length (HDR_LEN): if HDR_LEN is larger than the length of the
standard header then the remaining header space is taken by Header
Extension fields.
If present, Header Extensions MUST be processed to ensure that they
are recognized before performing any congestion control procedure or
otherwise accepting a packet. The default action for unrecognized
Header Extensions is to ignore them. This allows the future
introduction of backward-compatible enhancements to ALC without
changing the ALC version number. Non backward-compatible Header
Extensions CANNOT be introduced without changing the ALC version
number.
There are two formats for Header Extension fields, as depicted below.
The first format is used for variable-length extensions, with Header
Extension Type (HET) values between 0 and 127. The second format is
used for fixed length (one 32-bit word) extensions, using HET values
from 127 to 255.
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| HET (<=127) | HEL | |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +
. .
. Header Extension Content (HEC) .
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| HET (>=128) | Header Extension Content (HEC) |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Figure 3 - Format of additional headers
The explanation of each sub-field is the following.
Header Extension Type (HET): 8 bits
The type of the Header Extension. This document defines a
number of possible types. Additional types may be defined in
future versions of this specification. HET values from 0 to
127 are used for variable-length Header Extensions. HET values
from 128 to 255 are used for fixed-length 32-bit Header
Extensions.
Header Extension Length (HEL): 8 bits
The length of the whole Header Extension field, expressed in
multiples of 32-bit words. This field MUST be present for
variable-length extensions (HET between 0 and 127) and MUST NOT
be present for fixed-length extensions (HET between 128 and
255).
Header Extension Content (HEC): variable length
The content of the Header Extension. The format of this sub-
field depends on the Header Extension type. For fixed-length
Header Extensions, the HEC is 24 bits. For variable-length
Header Extensions, the HEC field has variable size, as
specified by the HEL field. Note that the length of each
Header Extension field MUST be a multiple of 32 bits. Also
note that the total size of the LCT header, including all
Header Extensions and all optional header fields, cannot exceed
255 32-bit words.
Header Extensions are further divided between general LCT extensions
and Protocol Instantiation specific extensions (PI-specific).
General LCT extensions have HET in the ranges 0:63 and 128:191
inclusive. PI-specific extensions have HET in the ranges 64:127 and
192:255 inclusive.
General LCT extensions are intended to allow the introduction of
backward-compatible enhancements to LCT without changing the LCT
version number. Non backward-compatible Header Extensions CANNOT be
introduced without changing the LCT version number.
PI-specific extensions are reserved for PI-specific use with semantic
and default parsing actions defined by the PI.
The following general LCT Header Extension types are defined:
EXT_NOP=0 No-Operation extension.
The information present in this extension field MUST be
ignored by receivers.
EXT_AUTH=1 Packet authentication extension
Information used to authenticate the sender of the
packet. The format of this Header Extension and its
processing is outside the scope of this document and is
to be communicated out-of-band as part of the Session
Description.
It is RECOMMENDED that senders provide some form of
packet authentication. If EXT_AUTH is present,
whatever packet authentication checks that can be
performed immediately upon reception of the packet
SHOULD be performed before accepting the packet and
performing any congestion control-related action on it.
Some packet authentication schemes impose a delay of
several seconds between when a packet is received and
when the packet is fully authenticated. Any congestion
control related action that is appropriate MUST NOT be
postponed by any such full packet authentication.
All senders and receivers implementing ALC MUST support the EXT_NOP
Header Extension and MUST recognize EXT_AUTH, but MAY NOT be able to
parse its content.
For this version of ALC, the following PI-specific extension is
defined:
EXT_FTI=64 FEC Object Transmission Information extension
The purpose of this extension is to carry in-band the
FEC Object Transmission Information for an object. The
format of this Header Extension and its processing is
outside the scope of this document and is to be
communicated out-of-band as part of the Session
Description.
4.4 Sender Operation
The sender operation when using ALC includes all the points made
about the sender operation when using the LCT building block [11],
the FEC building block [10] and the multiple rate congestion control
building block.
A sender using ALC MUST make available the required Session
Description as described in Section 2.4. A sender also MUST make
available the required FEC Object Transmission Information as
described in Section 2.3.
Within a session a sender transmits a sequence of packets to the
channels associated with the session. The ALC sender MUST obey the
rules for filling in the CCI field in the packet headers and MUST
send packets at the appropriate rates to the channels associated with
the session as dictated by the multiple rate congestion control
building block.
The ALC sender MUST use the same TSI for all packets in the session.
Several objects MAY be delivered within the same ALC session. If
more than one object is to be delivered within a session then the
sender MUST use the TOI field and each object MUST be identified by a
unique TOI within the session, and the sender MUST use corresponding
TOI for all packets pertaining to the same object. The FEC Payload
ID MUST correspond to the encoding symbol(s) for the object carried
in the payload of the packet.
Objects MAY be transmitted sequentially within a session, and they
MAY be transmitted concurrently. However, it is good practice to
only send objects concurrently in the same session if the receivers
that participate in that portion of the session have interest in
receiving all the objects. The reason for this is that it wastes
bandwidth and networking resources to have receivers receive data for
objects that they have no interest in. However, there are no rules
with respect to mixing packets for different objects carried within
the session. Although this issue affects the efficiency of the
protocol, it does not affect the correctness nor the inter-
operability of ALC between senders and receivers.
Typically, the sender(s) continues to send packets in a session until
the transmission is considered complete. The transmission may be
considered complete when some time has expired, a certain number of
packets have been sent, or some out-of-band signal (possibly from a
higher level protocol) has indicated completion by a sufficient
number of receivers.
It is RECOMMENDED that packet authentication be used. If packet
authentication is used then the Header Extensions described in
Section 4.3 MUST be used to carry the authentication.
This document does not pose any restriction on packet sizes.
However, network efficiency considerations recommend that the sender
uses as large as possible packet payload size, but in such a way that
packets do not exceed the network's maximum transmission unit size
(MTU), or fragmentation coupled with packet loss might introduce
severe inefficiency in the transmission. It is RECOMMENDED that all
packets have the same or very similar sizes, as this can have a
severe impact on the effectiveness of the multiple rate congestion
control building block.
4.5 Receiver Operation
The receiver operation when using ALC includes all the points made
about the receiver operation when using the LCT building block [11],
the FEC building block [10] and the multiple rate congestion control
building block.
To be able to participate in a session, a receiver MUST obtain the
REQUIRED Session Description as listed in Section 2.4. How receivers
obtain a Session Description is outside the scope of this document.
To be able to be a receiver in a session, the receiver MUST be able
to process the ALC header. The receiver MUST be able to discard,
forward, store or process the other headers and the packet payload.
If a receiver is not able to process the ALC header, it MUST drop
from the session.
To be able to participate in a session, a receiver MUST implement the
multiple rate congestion control building block using the Congestion
Control Information field provided in the LCT header. If a receiver
is not able to implement the multiple rate congestion control
building block it MUST NOT join the session.
Several objects can be carried either sequentially or concurrently
within the same session. In this case, each object is identified by
a unique TOI. Note that even if a sender stops sending packets for
an old object before starting to transmit packets for a new object,
both the network and the underlying protocol layers can cause some
reordering of packets, especially when sent over different channels,
and thus receivers SHOULD NOT assume that the reception of a packet
for a new object means that there are no more packets in transit for
the previous one, at least for some amount of time.
As described in Section 2.3, a receiver MUST obtain the required FEC
Object Transmission Information for each object for which the
receiver receives and processes packets.
A receiver MAY concurrently join multiple ALC sessions from one or
more senders. The receiver MUST perform congestion control on each
such session. The receiver MAY make choices to optimize the packet
flow performance across multiple sessions, as long as the receiver
still adheres to the multiple rate congestion control building block
for each session individually.
Upon receipt of each packet the receiver proceeds with the following
steps in the order listed.
(1) The receiver MUST parse the packet header and verify that it is a
valid header. If it is not valid then the packet MUST be
discarded without further processing. If multiple packets are
received that cannot be parsed then the receiver SHOULD leave the
session.
(2) The receiver MUST verify that the sender IP address together with
the TSI carried in the header matches one of the (sender IP
address, TSI) pairs that was received in a Session Description
and that the receiver is currently joined to. If there is not a
match then the packet MUST be discarded without further
processing. If multiple packets are received with non-matching
(sender IP address, TSI) values then the receiver SHOULD leave
the session. If the receiver is joined to multiple ALC sessions
then the remainder of the steps are performed within the scope of
the (sender IP address, TSI) session of the received packet.
(3) The receiver MUST process and act on the CCI field in accordance
with the multiple rate congestion control building block.
(4) If more than one object is carried in the session, the receiver
MUST verify that the TOI carried in the LCT header is valid. If
the TOI is not valid, the packet MUST be discarded without
further processing.
(5) The receiver SHOULD process the remainder of the packet,
including interpreting the other header fields appropriately, and
using the FEC Payload ID and the encoding symbol(s) in the
payload to reconstruct the corresponding object.
It is RECOMMENDED that packet authentication be used. If packet
authentication is used then it is RECOMMENDED that the receiver
immediately check the authenticity of a packet before proceeding with
step (3) above. If immediate checking is possible and if the packet
fails the check then the receiver MUST discard the packet and reduce
its reception rate to a minimum before continuing to regulate its
reception rate using the multiple rate congestion control.
Some packet authentication schemes such as TESLA [14] do not allow an
immediate authenticity check. In this case the receiver SHOULD check
the authenticity of a packet as soon as possible, and if the packet
fails the check then it MUST be discarded before step (5) above and
reduce its reception rate to a minimum before continuing to regulate
its reception rate using the multiple rate congestion control.
5. Security Considerations
The same security consideration that apply to the LCT, FEC and the
multiple rate congestion control building blocks also apply to ALC.
Because of the use of FEC, ALC is especially vulnerable to denial-
of-service attacks by attackers that try to send forged packets to
the session which would prevent successful reconstruction or cause
inaccurate reconstruction of large portions of the object by
receivers. ALC is also particularly affected by such an attack
because many receivers may receive the same forged packet. There are
two ways to protect against such attacks, one at the application
level and one at the packet level. It is RECOMMENDED that prevention
be provided at both levels.
At the application level, it is RECOMMENDED that an integrity check
on the entire received object be done once the object is
reconstructed to ensure it is the same as the sent object. Moreover,
in order to obtain strong cryptographic integrity protection a
digital signature verifiable by the receiver SHOULD be used to
provide this application level integrity check. However, if even one
corrupted or forged packet is used to reconstruct the object, it is
likely that the received object will be reconstructed incorrectly.
This will appropriately cause the integrity check to fail and in this
case the inaccurately reconstructed object SHOULD be discarded.
Thus, the acceptance of a single forged packet can be an effective
denial of service attack for distributing objects, but an object
integrity check at least prevents inadvertent use of inaccurately
reconstructed objects. The specification of an application level
integrity check of the received object is outside the scope of this
document.
At the packet level, it is RECOMMENDED that a packet level
authentication be used to ensure that each received packet is an
authentic and uncorrupted packet containing FEC data for the object
arriving from the specified sender. Packet level authentication has
the advantage that corrupt or forged packets can be discarded
individually and the received authenticated packets can be used to
accurately reconstruct the object. Thus, the effect of a denial of
service attack that injects forged packets is proportional only to
the number of forged packets, and not to the object size. Although
there is currently no IETF standard that specifies how to do
multicast packet level authentication, TESLA [14] is a known
multicast packet authentication scheme that would work.
In addition to providing protection against reconstruction of
inaccurate objects, packet level authentication can also provide some
protection against denial of service attacks on the multiple rate
congestion control. Attackers can try to inject forged packets with
incorrect congestion control information into the multicast stream,
thereby potentially adversely affecting network elements and
receivers downstream of the attack, and much less significantly the
rest of the network and other receivers. Thus, it is also
RECOMMENDED that packet level authentication be used to protect
against such attacks. TESLA [14] can also be used to some extent to
limit the damage caused by such attacks. However, with TESLA a
receiver can only determine if a packet is authentic several seconds
after it is received, and thus an attack against the congestion
control protocol can be effective for several seconds before the
receiver can react to slow down the session reception rate.
Reverse Path Forwarding checks SHOULD be enabled in all network
routers and switches along the path from the sender to receivers to
limit the possibility of a bad agent injecting forged packets into
the multicast tree data path.
A receiver with an incorrect or corrupted implementation of the
multiple rate congestion control building block may affect health of
the network in the path between the sender and the receiver, and may
also affect the reception rates of other receivers joined to the
session. It is therefore RECOMMENDED that receivers be required to
identify themselves as legitimate before they receive the Session
Description needed to join the session. How receivers identify
themselves as legitimate is outside the scope of this document.
Another vulnerability of ALC is the potential of receivers obtaining
an incorrect Session Description for the session. The consequences
of this could be that legitimate receivers with the wrong Session
Description are unable to correctly receive the session content, or
that receivers inadvertently try to receive at a much higher rate
than they are capable of, thereby disrupting traffic in portions of
the network. To avoid these problems, it is RECOMMENDED that
measures be taken to prevent receivers from accepting incorrect
Session Descriptions, e.g., by using source authentication to ensure
that receivers only accept legitimate Session Descriptions from
authorized senders. How this is done is outside the scope of this
document.
6. IANA Considerations
No information in this specification is directly subject to IANA
registration. However, building blocks components used by ALC may
introduce additional IANA considerations. In particular, the FEC
building block used by ALC does require IANA registration of the FEC
codecs used.
7. Intellectual Property Issues
The IETF has been notified of intellectual property rights claimed in
regard to some or all of the specification contained in this
document. For more information consult the online list of claimed
rights.
8. Acknowledgments
Thanks to Vincent Roca, Justin Chapweske and Roger Kermode for their
detailed comments on this document.
9. References
[1] Bradner, S., "The Internet Standards Process -- Revision 3", BCP
9, RFC 2026, October 1996.
[2] Bradner, S., "Key words for use in RFCs to Indicate Requirement
Levels", BCP 14, RFC 2119, March 1997.
[3] Deering, S., "Host Extensions for IP Multicasting", STD 5, RFC
1112, August 1989.
[4] Fielding, R., Gettys, J., Mogul, J., Frystyk, H. and T.
Berners-Lee, "Hypertext Transfer Protocol -- HTTP/1.1", RFC
2616, January 1997.
[5] Handley, M. and V. Jacobson, "SDP: Session Description
Protocol", RFC 2327, April 1998.
[6] Handley, M., Perkins, C. and E. Whelan, "Session Announcement
Protocol", RFC 2974, October 2000.
[7] Holbrook, H. W., "A Channel Model for Multicast", Ph.D.
Dissertation, Stanford University, Department of Computer
Science, Stanford, California, August 2001.
[8] Kermode, R., Vicisano, L., "Author Guidelines for Reliable
Multicast Transport (RMT) Building Blocks and Protocol
Instantiation documents", RFC 3269, April 2002.
[9] Luby, M., Vicisano, L., Gemmell, J., Rizzo, L., Handley, M. and
J. Crowcroft, "The Use of Forward Error Correction (FEC) in
Reliable Multicast", RFC 3453, December 2002.
[10] Luby, M., Vicisano, L., Gemmell, J., Rizzo, L., Handley, M., and
J. Crowcroft, "Forward Error Correction (FEC) Building Block",
RFC 3452, December 2002.
[11] Luby, M., Gemmell, J., Vicisano, L., Rizzo, L., Handley, M. and
J. Crowcroft, "Layered Coding Transport (LCT) Building Block",
RFC 3451 December 2002.
[12] Mankin, A., Romanow, A., Bradner, S. and V. Paxson, "IETF
Criteria for Evaluating Reliable Multicast Transport and
Application Protocols", RFC 2357, June 1998.
[13] Murata, M., St.Laurent, S. and D. Kohn, "XML Media Types", RFC
3023, January 2001.
[14] Perrig, A., Canetti, R., Song, D. and J.D. Tygar, "Efficient and
Secure Source Authentication for Multicast", Network and
Distributed System Security Symposium, NDSS 2001, pp. 35-46,
February 2001.
[15] Postel, J., "User Datagram Protocol", STD 6, RFC 768, August
1980.
[16] Whetten, B., Vicisano, L., Kermode, R., Handley, M., Floyd, S.
and M. Luby, "Reliable Multicast Transport Building Blocks for
One-to-Many Bulk-Data Transfer", RFC 3048, January 2001.
Authors' Addresses
Michael Luby
Digital Fountain
39141 Civic Center Dr.
Suite 300
Fremont, CA, USA, 94538
EMail: luby@digitalfountain.com
Jim Gemmell
Microsoft Research
455 Market St. #1690
San Francisco, CA, 94105
EMail: jgemmell@microsoft.com
Lorenzo Vicisano
cisco Systems, Inc.
170 West Tasman Dr.
San Jose, CA, USA, 95134
EMail: lorenzo@cisco.com
Luigi Rizzo
Dip. Ing. dell'Informazione,
Univ. di Pisa
via Diotisalvi 2, 56126 Pisa, Italy
EMail: luigi@iet.unipi.it
Jon Crowcroft
Marconi Professor of Communications Systems
University of Cambridge
Computer Laboratory
William Gates Building
J J Thomson Avenue
Cambridge CB3 0FD, UK
EMail: Jon.Crowcroft@cl.cam.ac.uk
Full Copyright Statement
Copyright (C) The Internet Society (2002). All Rights Reserved.
This document and translations of it may be copied and furnished to
others, and derivative works that comment on or otherwise explain it
or assist in its implementation may be prepared, copied, published
and distributed, in whole or in part, without restriction of any
kind, provided that the above copyright notice and this paragraph are
included on all such copies and derivative works. However, this
document itself may not be modified in any way, such as by removing
the copyright notice or references to the Internet Society or other
Internet organizations, except as needed for the purpose of
developing Internet standards in which case the procedures for
copyrights defined in the Internet Standards process must be
followed, or as required to translate it into languages other than
English.
The limited permissions granted above are perpetual and will not be
revoked by the Internet Society or its successors or assigns.
This document and the information contained herein is provided on an
"AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING
TASK FORCE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
Acknowledgement
Funding for the RFC Editor function is currently provided by the
Internet Society.