Dear Colleagues,
We are writing to announce the Cowles Foundation Conference on Optimization,
which will take place on March 20th and 21st, 2006.
The list of speakers will include:
Farid Alizadeh (Rutgers)
Sanjeev Arora (Princeton)
Dmitri Bertsimas (MIT)
Lisa Fleischer (IBM Watson)
Rob Freund (MIT)
Baruch Schieber (IBM Watson)
Cliff Stein (Columbia)
Maxim Sviridenko (IBM Watson)
Michael Overton (NYU)
Michael Todd (Cornell)
Santosh Vempala (MIT)
David Williamson (Cornell)
Mihalis Yannakakis (Columbia)
The talks on Monday the 20th will take place in LOM 201,
and the talks on Tuesday the 21st will take place in AKW 200.
We will announce a precise schedule of talks in the next few days.
There will be a dinner for the conference attendees on March 20th.
Although seating is limited, please let us know if you are interested
in attending.
Sincerely yours,
Herb Scarf and Dan Spielman
Yale's Applied Math Program has pioneered a novel clustered computer architecture that balances the needs for CPU intensive ('compute bound') and data I/O intensive ('I/O bound') computing. A detailed asymptotic mathematical analysis of the new system shows that it avoids the common pitfall of parallel, cluster computing, often called 'Amdahl's law', in which bottlenecks appear and cluster performance saturates quickly to a fixed level, even as the number of compute nodes. This new cluster is the basis for the design of a computational environment for massive dataset analysis, a so-called 'petabyte-scale storage facility', which will enable path-breaking data-intensive analyses in fields as diverse as biomedical systems, financial analysis, image processing, next-generation search engines, etc.