<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=us-ascii">
<META content="MSHTML 6.00.6000.16735" name=GENERATOR></HEAD>
<BODY>
<DIV dir=ltr align=left><FONT face=Arial color=#0000ff size=2><SPAN
class=989373014-05112008>Chang Kim will present his preFPO on Monday November 10
at 9AM in Room 402.</SPAN></FONT></DIV>
<DIV dir=ltr align=left><FONT face=Arial color=#0000ff size=2><SPAN
class=989373014-05112008>The members of his committee are: Jennifer
Rexford, advisor; Michael Freedman </SPAN></FONT></DIV>
<DIV dir=ltr align=left><FONT face=Arial color=#0000ff size=2><SPAN
class=989373014-05112008>and Albert Greenberg (MSR), readers; Vivek Pai and
Margaret Martonosi, nonreaders. </SPAN></FONT></DIV>
<DIV dir=ltr align=left><FONT face=Arial color=#0000ff size=2><SPAN
class=989373014-05112008>Everyone is invited to attend his talk. The
abstract follows below.</SPAN></FONT></DIV>
<DIV dir=ltr align=left><FONT face=Arial color=#0000ff size=2><SPAN
class=989373014-05112008>---------------------------------</SPAN></FONT></DIV>
<DIV><FONT size=4><BR><STRONG>Scalable and Efficient Self-Configuring
Networks</STRONG></FONT></DIV>
<DIV><FONT size=4></FONT> </DIV>
<DIV><FONT size=4>Abstract:</FONT></DIV>
<DIV><FONT size=4></FONT> </DIV>
<DIV><FONT size=4>Maintaining scalability and efficiency in conventional
networks often requires<BR>huge amount of administrative effort to configure
network devices <BR>and protocols. Large configuration overhead increases
operational expenditures </FONT></DIV>
<DIV><FONT size=4>and leads to frequent failures due to configuration mistakes.
<BR>On the other hand, self-configuring networks, such as Ethernet, work only
<BR>on a small scale, because increasing the size of the network rapidly
degrades <BR>its performance (e.g., skews link utilization, stretches
traffic-forwarding paths, <BR>lowers availability, delays failover, etc.).
</FONT></DIV>
<DIV><FONT size=4></FONT> </DIV>
<DIV><FONT size=4>This thesis proposes a simple and generic set of architectural
primitives <BR>that enable self-configuring in a large network without
sacrificing performance. <BR>Those primitives are flat addressing, traffic
indirection, and caching of routing state. <BR>By building upon these
primitives, the thesis proposes and evaluates three specific <BR>architectural
solutions designed respectively for enterprises, data centers, and
providers: <BR>i) a scalable Ethernet architecture for large enterprises
(SEATTLE),<BR>ii) a next-generation data-center network architecture (Monsoon),
and <BR>iii) a scalable VPN routing architecture for large tier-1 providers
(Relaying). <BR>The resulting architectures collectively demonstrate the
efficacy of the <BR>primitives in resolving the tussle among scalability,
efficiency, and lower <BR>management complexity.</FONT></DIV></BODY></HTML>