<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=iso-8859-1">
<META content="MSHTML 6.00.2900.2180" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=#ffffff>
<DIV>Alexander Löser wrote:<BR>> Hey all,<BR>> structured overlay
networks based on DHT's, such as Pastry and Chord<BR>> among
others, have been investigated in the past to construct scalable<BR>> and
performance orientated peer-to-peer networks. However,
unstructured<BR>> networks, such as Gnutella or Kazaa, are still widely
used among the<BR>> file sharing community. Recently researchers proposed
extensions to<BR>> unstructured networks networks based on the small
world idea: peers<BR>> dynamically create shortcuts to other peers based on
their interests.<BR>> Over a while peers with the same interests became
direct neighbors<BR>> through its shortcuts and build interest based
clusters. Hence peers<BR>> no longer flood messages but partly
route it's queries via a interested<BR>> based/semantic overlay.
Examples are described in [1] [2] among<BR>> others.<BR>> <BR>>
Comparing small world and DHT approaches is a difficult task,
since<BR>> simulations usually differ in scenarios, data sets or
simulation<BR>> methodology. I'm interested in scenarios
and arguments PRO small<BR>> world overlays for unstructured networks. Does
anybody now actual<BR>> theoretic or practical work that compares both
approaches in different<BR>> scenarios (high churn, no super peers, key word
based search, meta data<BR>> based search)? Which scenarios or
arguments support small world<BR>> approaches for unstructured
networks?<BR>> <BR>> Alex<BR>> <BR>> <BR>> <BR>> <BR>> [1]
Gia - Making Gnutella like P2P Systems Scalable<BR>> <A
href="http://berkeley.intel-research.net/sylvia/1103-chawathe.pdf">http://berkeley.intel-research.net/sylvia/1103-chawathe.pdf</A><BR>>
<A
href="http://seattle.intel-research.net/people/yatin/publications/talks/sigcomm2003-gia.ppt">http://seattle.intel-research.net/people/yatin/publications/talks/sigcomm2003-gia.ppt</A><BR>>
<BR>> [2] Efficient Content Location Using Interest Based Locality
in<BR>> Peer-to-Peer Systems<BR>> <A
href="http://www.ieee-infocom.org/2003/papers/53_01.PDF">http://www.ieee-infocom.org/2003/papers/53_01.PDF</A><BR>>
--<BR>> ___________________________________________________________<BR>>
<BR>> Alexander Löser<BR>> Technische Universitaet
Berlin<BR>> hp: <A
href="http://cis.cs.tu-berlin.de/~aloeser/">http://cis.cs.tu-berlin.de/~aloeser/</A><BR>>
office: +49- 30-314-25551<BR>> fax : +49-
30-314-21601<BR>>
___________________________________________________________<BR>> <BR>>
<BR>> _______________________________________________<BR>> p2p-hackers
mailing list<BR>> <A
href="mailto:p2p-hackers@zgp.org">p2p-hackers@zgp.org</A><BR>> <A
href="http://zgp.org/mailman/listinfo/p2p-hackers">http://zgp.org/mailman/listinfo/p2p-hackers</A><BR></DIV>
<DIV><FONT face=Arial>Interesting discussion Alex.</FONT></DIV>
<DIV> </DIV>
<DIV>From the practical and system challenges that faced researchers working on
DHTs (long time for the network to become stable, updates and maintenance for
nodes join and leave, high cost of messaging when adding an object to the
network, ..), it has become the norm to think about the application when trying
to decide to use structured (DHTs) or unstructured (gnutella-like) p2p
topologies. That is probably one of the reasons why people did not compare both
structures in an analysis similar to what you are asking for. Thus, small world
and power-law have emerged to bridge the gap between a total random network and
a "rigid" DHT. Note that super-peers in Kazaa and Gnutella do actually help the
network become more like a small-world. We also have worked in this area and
created a power-law distribution P2P network that might interest you: </DIV>
<DIV>- Rita H. Wouhaybi, and Andrew T. Campbell, <SPAN
style="FONT-SIZE: 12pt; FONT-FAMILY: 'Times New Roman'; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA">“<A
href="http://www.comet.columbia.edu/~rita/pdfs/phenix-infocom04.pdf">Phenix:
Supporting Resilient Low-Diameter Peer-to-Peer Topologies</A>”, IEEE
INFOCOM'2004, Hong Kong, China, March 7-11, 2004.</SPAN> </DIV>
<DIV> </DIV>
<DIV><FONT face=Arial></FONT> </DIV>
<DIV><FONT face=Arial>Rita H. Wouhaybi<BR><A
href="mailto:rita@comet.columbia.edu">rita@comet.columbia.edu</A><BR><A
href="http://comet.columbia.edu/~rita/">http://comet.columbia.edu/~rita/</A></FONT></DIV></BODY></HTML>