[Nrg-l] FW: TOC seminar 5/7: Lenore Cowen

Mark Crovella crovella at cs.bu.edu
Thu May 3 16:00:37 EDT 2007


Looks like an interesting seminar!
 
Mark

________________________________

From: leonid.reyzin at gmail.com [mailto:leonid.reyzin at gmail.com] On Behalf
Of Leonid Reyzin
Sent: Wednesday, May 02, 2007 9:54 AM
To: Azer Bestavros; Abraham Matta; John Byers; Mark Crovella
Subject: Fwd: TOC seminar 5/7: Lenore Cowen



Below is a seminar announcement that may be of interest to the
networking community:


---------- Forwarded message ----------
From: Salil Vadhan <salil at eecs.harvard.edu>
Date: May 2, 2007 12:39 AM 
Subject: TOC seminar 5/7: Lenore Cowen
To: theory-seminars at eecs.harvard.edu


Theory of Computation Seminar
-----------------------------
Speaker: Lenore Cowen, Tufts University
Place and Time: Monday, 5/7, Maxwell Dworkin 221
Refreshments at 2:30, Talk at 2:45

Title: Compact Routing from Theory to Practice

Abstract:
The compact routing problem is a classical problem in the theory of 
distributed algorithms. Compact routing achieves routing tables whose
size
grows sublinearly with the number of network nodes in exchange for near
optimal, rather than optimal shortest path routes.In addition, some
recent 
schemes achieve a "name independence" property where node are
arbitrarily
assigned persistent identifiers in a way that is decoupled from the
underlying network topology. `Name independence''
(also called ``flat addressing'' or a ``flat address space'' in the
Networking community) has recently gained traction as one approach to
clean
slate Internet redesign that may be required to handle increasing number
of 
mobile nodes (i.e. laptops; wireless devices) and dynamic changes in
network
connectivity.  In fact, several recent proposals for a complete redesign
of
the Internet architecture have called for separating the naming layer
from 
the packet forwarding layer.

We show that compact routing schemes perform even better on power-law
networks that may more realistically model the Internet's inter-AS
graph,
and give both theoretical and experimental results. We also discuss 
remaining questions about dynamic networks, congestion control, traffic
engineering, and policy that must also be addressed for any compact
routing
scheme to be extended into a realistic candidate for a next-generation 
routing protocol.


Upcoming TOC seminars and related talks:
----------------------------------------
5/7, 12:30pm: Shien Jin Ong Ph.D. defense
5/7: Lenore Cowen, Tufts
5/14: Allan Borodin, Toronto 


_______________________________________________
Theory-seminars mailing list
Theory-seminars at eecs.harvard.edu
Subscribe/unsubscribe/customize at
https://www.eecs.harvard.edu/mailman/listinfo/theory-seminars

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://cs-mailman.bu.edu/pipermail/nrg-l/attachments/20070503/5d6fb08d/attachment.html


More information about the Nrg-l mailing list