SAM J. Buford, Panasonic
Internet Draft S. Kadadi, Motorola
Expires: June 30, 2007 December 31, 2006
SAM Problem Statement
draft-irtf-sam-problem-statement-01.txt
Status of this Memo
By submitting this Internet-Draft, each author represents that
any applicable patent or other IPR claims of which he or she is
aware have been or will be disclosed, and any of which he or she
becomes aware will be disclosed, in accordance with Section 6 of
BCP 79.
Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF), its areas, and its working groups. Note that
other groups may also distribute working documents as Internet-
Drafts.
Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as "work in progress."
The list of current Internet-Drafts can be accessed at
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
This Internet-Draft will expire on June 30, 2007.
Copyright Notice
Copyright (C) The Internet Society (2006). All Rights Reserved.
Abstract
We describe the generally expected behavior of a scalable and
adaptive multicast architecture, leaving further details to separate
documents on requirements and the SAM design space. This document is
Buford Expires June 30, 2007 [Page 1]
Internet-Draft SAM Problem Statement December 30, 2006
a starting point for discussions of feasibility, priority, and
deployability.
Conventions used in this document
In examples, "C:" and "S:" indicate lines sent by the client and
server respectively.
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 RFC-2119 [1].
Table of Contents
1. Introduction...................................................2
2. Heterogeneous Multicast Infrastructure.........................3
2.1. Varying Infrastructure by Network Region..................3
2.2. Regional Transitions......................................4
3. Quality of Service.............................................4
3.1. Native QOS, No Native Multicast...........................4
3.2. Other Combinations........................................5
4. Mobility.......................................................5
4.1. Multicast Service Selection...............................6
4.2. Transitions between ALM and Native Multicast..............7
4.3. Other Considerations......................................7
5. Security Considerations........................................7
6. Conclusions....................................................8
7. References.....................................................8
7.1. Normative References......................................8
7.2. Informative References....................................8
Author's Addresses................................................8
Intellectual Property Statement...................................9
Disclaimer of Validity............................................9
Copyright Statement...............................................9
Acknowledgment....................................................9
1. Introduction
The concept of scalable adaptive multicast includes both scaling
properties and adaptability properties. Scalability is intended to
cover:
o large group size
o large numbers of small groups
Buford Expires June 30, 2007 [Page 2]
Internet-Draft SAM Problem Statement December 30, 2006
o rate of group membership change
o admission control for QoS
o use with network layer QoS mechanisms
o varying degrees of reliability
o trees connect nodes over global internet
Adaptability includes
o use of different control mechanisms for different multicast trees
depending on initial application parameters or application class
o changing multicast tree structure depending on changes in
application requirements, network conditions, and membership
o use of different control mechanisms and tree structure in
different regions of network depending on native multicast
support, network characteristics, and node behavior
The following sections describe some adaptation scenarios. After the
base scenarios are elaborated, then scenarios for scalability and
dynamic adaptation should be added.
2. Heterogeneous Multicast Infrastructure
2.1. Varying Infrastructure by Network Region
Regions A, B, C are disjoint areas of the network with some type of
native multicast support. Region Z is all other areas of the network
with no native multicast support. Region Z may be partitioned by A,
B, and/or C.
A multicast connection between nodes in A, B, C, and Z is needed. In
each region A, B, C, the respective native multicast mechanism is
used.
Multicast topology choices include:
o Multicast applications see an end-to-end multicast application
layer which is mapped to a native layer transparently in the
regions that it is available. The overlay’s group management
mechansisms hold for all nodes, and are mapped transparently to
the native layer mechanisms in the appropriate regions. All nodes
have addresses in the overlay.
Buford Expires June 30, 2007 [Page 3]
Internet-Draft SAM Problem Statement December 30, 2006
o Multicast applications see an end-to-end native multicast, where
nodes in region Z connect to native regions using tunnels. The
native group management mechanisms hold for all nodes.
Homogeneous sub-case: regions A, B, C may use the same native
multicast protocol.
2.2. Regional Transitions
A node in a new region D joins the multicast tree. Region D has
native support.
What is the minimum number of nodes in a region needed for native
support to be used in that part of the tree?
3. Quality of Service
3.1. Native QOS, No Native Multicast
Each endpoint in the multicast tree specifies QOS constraints such as
bandwidth, delay, and jitter for a given source. Multicast join
includes admission control step for the selected QOS mechanism. This
means that the join decision combines both multicast tree
considerations (eg., best metrics) and an admission control decision.
Paths to different endpoints from a given source might have different
QOS constraints. A given multicast tree may mix QOS delivery and
best effort delivery to different receivers.
Available IP QOS mechanisms include Intserv, Diffserv, and MPLS.
Assume all regions of network have interoperable native QOS
mechanism. Assume all receivers have homogenous capabilities.
The topology of the overlay is not assumed to be isomorphic to
available QOS paths. The overlay must be sophisticated enough to
determine what paths are available and arrange its tree construction
and routing behaviour accordingly.
In order to enforce QOS, a measurement mechanism is needed. The
scalability of the measurement, feedback and policing mechanism is an
important issue. RTP is such a measurement and feedback protocol for
UDP.
A source might adapt its bit rate and quality depending on feedback
from receivers. There might be graceful degradation mechanisms such
as multi-description coding over different multicast paths. This
behavior is application dependent.
Buford Expires June 30, 2007 [Page 4]
Internet-Draft SAM Problem Statement December 30, 2006
3.2. Other Combinations
Heterogeneous QOS refers to either 1) portions of the network where
no QOS mechanism exists at native level, or 2) receivers which have
heterogeneous capabilities.
These combinations need further elaboration.
o Native QOS with Regional Native Multicast
o Heterogeneous QOS, No Native Multicast
o Heterogeneous QOS, Regional Native Multicast
4. Mobility
We assume mobile nodes use Mobile IP (MIP), and that regions of the
network that mobile nodes operate in are MIP enabled. Any node in the
multicast tree may be mobile, and we consider source node mobility as
a special case.
A mobile node’s home address (HoA) is associated with its overlay
address (if this is an overlay) or group multicast address. As the
node moves to another network, multicast messages are routed to it
via the home agent (HA). In addition to increased latency, node
mobility can impact robustness of multicast delivery due to loss of
connectivity during mobility transitions. Some link layer solutions
may mitigate or eliminate connectivity loss, but may require sending
packets to both old and new care-of addresses during the transition.
If the node uses its care-of address (CoA) in the overlay or
multicast tree, then any mobility transition will be disruptive,
causing a leave-join sequence.
Forwarding of packets can be through the home agent. If the source
address is the care-of address, these might be rejected by nodes
expecting packets only from overlay-registered addresses.
In general, mobile node transitions to another network lead to lost
packets during the transition, and downstream nodes in the tree will
also be disconnected. Possibile solutions are bi-casting the packets
to both old and new CoA, or buffering packets at the HA or old or new
anchored points.
If the overlay is aware that the node is mobile, then it could
construct a mesh rather than tree to connect to. The mesh might
provide redundant paths to the mobile node’s children in the tree.
Buford Expires June 30, 2007 [Page 5]
Internet-Draft SAM Problem Statement December 30, 2006
The overlay might use knowledge about node movement to make a set of
target anchor points prejoin the multicast group/service so that the
handoff delay can be reduced. Solutions similar to low latency
handoffs/Fast MIPv6 (prereg) could be used.
There can be different scenarios depending on whether all nodes in
multicast tree are mobile or a subset of nodes.
4.1. Multicast Service Selection
As the node moves from one network to another network, it can get
multicast service in the new network in three ways: 1) from the new
foreign network, 2) from the home network via a tunnel, 3) from the
old foreign network via a tunnel, if the transition was from one
foreign network to another. The selection may depend on what is
available in the new foreign network and which of the three
mechanisms was used in the previous foreign network.
1. Multicast service to/from the new foreign network
o In the case of native multicast, this means that the new foreign
network has a multicast router which the mobile node uses. The
mobile node can use the new IP address (obtained in the new
network). This causes a leave-join sequence. If the router in the
new network is not already a part the multicast tree, there will
be additional delay to join the multicast tree in the foreign
network. The mobile node uses either the CoA as the source address
of control messages or its HoA.
2. Multicast service via home network
o Multicast packets are tunneled to/from the mobile node by the HA.
The mobile node uses HoA for multicast control messages. No need
to join/leave the multicast group during handoff. The transmission
path is not optimal.
o If the tunnel end-point is not a mobile node, this may result in
duplicate packets. Consider the case where packets of the same
multicast group are tunneled to the new network. This means two
HAs are tunneling packets for the same multicast group to the same
foreign network. Possible solutions to this are: 1) applications
in mobile node takes care of duplicate packets, 2) multicast
packets are sent to mobile node as unicast packets (e.g., Mobile
IPv4 uses this solution).
3. Multicast service from old foreign network
Buford Expires June 30, 2007 [Page 6]
Internet-Draft SAM Problem Statement December 30, 2006
o The mobile node gets the packets from its old multicast service
anchor point until it registers/joins new multicast service anchor
point in new network. Once it starts getting packets from new
anchor point, it leaves its old anchor point. This means that
packets are tunneled from the old foreign network until the mobile
node gets multicast service from the new foreign network.
o There is minimal packet loss. There may be duplicate packets
during the transition. A leave and join sequence results.
4.2. Transitions between ALM and Native Multicast
Transitions of mobile nodes between heterogeneous multicast networks
(say, from a native multicast region to OL/ALM multicast region or
vice versa) need to be considered.
4.3. Other Considerations
o Multicast source mobility: Mobile source nodes may have more
impact than other cases, and overlay tree/mesh may be reorganized
when the multicast source moves to a new network. There may be
solutions specific to source node mobility that may not scale to
mobile nodes in general.
o Scalability of advertisement mechanisms. Multicast advertisements
are also multicast packets with well known multicast group address
and port number. If mobile nodes which are in the foreign network
want to know about multicast services in home network, these
advertisement packets should be sent to foreign network. The home
agent can tunnel packets to the foreign network, but it can
increase the load on the HA.
o Network topology supported by the access network.
5. Security Considerations
[RESC2006] surveys the security issues specific to overlay networks
which include:
o Correctness of routing due to malicious nodes acting individually
or collectively
o Node impersonation due to lack of secure routing and identity
o Fairness enforcement since each node acts autonomously, it can
chose to limit its resource contribution to the operation of the
overlay
Buford Expires June 30, 2007 [Page 7]
Internet-Draft SAM Problem Statement December 30, 2006
o Denial of service (DOS)
o Using overlays for launching DDoS attacks [ROSS2006]
SAM will not solve the overlay security problems, but should work
with overlays that provide security mechanisms.
6. Conclusions
Using this discussion with the separately developed SAM Design Space,
we will be able to enumerate those ares of the problem space for
which solutions exist and those which are open problems. This will
suggest the steps by which the SAM Framework is designed.
7. References
7.1. Normative References
[1] Bradner, S., "Key words for use in RFCs to Indicate Requirement
Levels", BCP 14, RFC 2119, March 1997.
7.2. Informative References
[MUR2006] E. Muramoto, Y. Imai, N. Kawaguchi. Requirements for
Scalable Adaptive Multicast Framework in Non-GIG Networks.
November 2006. Internet Draft draft-muramoto-irtf-sam-
generic-require-01.txt, work in progress.
[RESC2006] E. Rescorla. Introduction to Distributed Hash Tables.
IETF-65 Technical Plenary, March 2006.
www3.ietf.org/proceedings/06mar/slides/plenaryt-2.pdf
[ROSS2006] K. Ross. Exploiting P2P Systems for DDOS Attacks. IETF
65 P2PRG CORE Subgroup. www.cs.uml.edu/~buford/irtf-
p2prg/ietf65/ietf65-irtf-p2prg-core-ddos.pdf
Author's Addresses
John Buford
Panasonic Princeton Laboratory
rd
2 Research Way, 3 Floor
Princeton, NJ 08540, USA
Email: buford@research.panasonic.com
Shivanand Kadadi
Motorola Bangalore India
Email: a22063@motorola.com
Buford Expires June 30, 2007 [Page 8]
Internet-Draft SAM Problem Statement December 30, 2006
Intellectual Property Statement
The IETF takes no position regarding the validity or scope of any
Intellectual Property Rights or other rights that might be claimed to
pertain to the implementation or use of the technology described in
this document or the extent to which any license under such rights
might or might not be available; nor does it represent that it has
made any independent effort to identify any such rights. Information
on the procedures with respect to rights in RFC documents can be
found in BCP 78 and BCP 79.
Copies of IPR disclosures made to the IETF Secretariat and any
assurances of licenses to be made available, or the result of an
attempt made to obtain a general license or permission for the use of
such proprietary rights by implementers or users of this
specification can be obtained from the IETF on-line IPR repository at
http://www.ietf.org/ipr.
The IETF invites any interested party to bring to its attention any
copyrights, patents or patent applications, or other proprietary
rights that may cover technology that may be required to implement
this standard. Please address the information to the IETF at
ietf-ipr@ietf.org
Disclaimer of Validity
This document and the information contained herein are provided on an
"AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET
ENGINEERING TASK FORCE DISCLAIM 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.
Copyright Statement
Copyright (C) The Internet Society (2006).
This document is subject to the rights, licenses and restrictions
contained in BCP 78, and except as set forth therein, the authors
retain all their rights.
Acknowledgment
Funding for the RFC Editor function is currently provided by the
Internet Society.
Buford Expires June 30, 2007 [Page 9]
Internet-Draft SAM Problem Statement December 30, 2006
Buford Expires June 30, 2007 [Page 10]