[e2e] Measurements of P2P traffic

[Ayyasamy, Senthilkumar (UMKC-Student)]

> I would appreciate if someone can point me any references or papers on
> measurements of current Peer-to-Peer traffic patterns in the 
> Internet. For
> example, the percentage and behaviours of P2P routing, lookup/query
> traffic in the Internet. 

1. A excellent measurement study of the two peer-to-peer file systems 
napster and gnutella.
*A Measurement Study of Peer-to-Peer File Sharing Systems*
by Stefan Saroiu, P. Krishna Gummadi, Steven D. Gribble
(http://www.cs.washington.edu/homes/gribble/papers/mmcn.pdf)
They even characterize the end host number in addition to traditional
traffic measures.
 A bottleneck BW measurement tool for nash type of network from 
 the same group (http://sprobe.cs.washington.edu/)

2. A flow-level analysis of napster traffic across a campus network.
*An Analysis of Napster and Other IP Flow Sizes* by dave plonka
http://net.doit.wisc.edu/~plonka/NATimes/
http://net.doit.wisc.edu/data/Napster/

and tool employed by them to get the flow level details can be found at:
http://net.doit.wisc.edu/~plonka/FlowScan/

3. A measurement study on non-co operative distributed arch for p2p 
systems.
*Early Measurements of a Cluster-based Architecture for P2P Systems*
by krishamurthy et all.
(http://www.icir.org/vern/imw-2001/imw2001-papers/29.pdf)

4. IMW 2002 should also have some works on peer to peer analysis. 
keep looking at their site for proceedings 
(http://www.icir.org/vern/imw-2002/papers.html)
I have found this title :
*Analyzing Peer-to-Peer Traffic Across Large Networks*
   - Subhabrata Sen and Jia Wang; AT&T Labs -- Research 


I think these are the only papers studying at measurement level. But also
it depends on what you want to do with the data: If you want to infer
topology, you can indirectly find helpful information in lot of AS topology 
inference studies. It also differs whether you want to do at application
level or network level measurement study.

But i dont think caida or any other traffic archive repositories 
have peer to peer traffic stats.

>I am interested in knowing how the current and
>proposed End-to-End TCP/IP implementation/design will impact P2P network
>and the Internet.

Actually the question should be reverse: *what impacts will p2p network
have on end to end design principle of internet*

Refer *Rethinking the design of the Internet: The end to end arguments 
vs. the brave new world* by clark et all.
(http://ana-www.lcs.mit.edu/anaweb/PDF/Rethinking_2001.pdf)

and also a SIGCOMM'02 position paper from clark et all:
*Tussle in Cyberspace: Defining Tomorrow's Internet* by clark et all
(http://www.acm.org/sigcomm/sigcomm2002/papers/tussle.html)

You can get some useful pointers from some of the papers in IPTPS'02 
proceedings. But this again depends on what you expect? overlay routing
or workload analysis etc...
http://www.cs.rice.edu/Conferences/IPTPS02/

>  I can't find  any P2P mailing list, is there any?

There is a peer to peer mailing list from  p2p group of internet2.
Its home http://www.internet2.edu/p2p/ and its archives can 
be found at  (http://listserv.utk.edu/archives/p2p.html)

- senthil