Fat servers are the dual of thin clients. Many forces are pushing us to build huge compute and storage servers. The hardware to build such servers is from commodity components is available today. Even more extraordinary hardware is expected soon. The required software has been slow to arrive. There are a few success stories: transaction processing systems, parallel database systems for datamining, and more recently web servers. This talk explains the key properties that enabled these successes: a simple programming model and parallelism that comes from many small requests or from massive dataflows.
About the speaker (updated):
On January 28, 2007, Jim Gray went sailing on San Francisco Bay and disappeared. A massive search mounted by Coast Guard and the Bay Area technical community found no traces of him or his boat, the Tenacious. See Tenacious Search for additional information. Information Week published an article on March 31, 2007, which provides some additional information and the latest news.
Jim Gray was a Senior Researcher at Microsoft Research, http://research.microsoft.com/barc/gray .
Dr. Gray was a specialist in database and transaction processing computer systems. At Microsoft his research focused on scaleable computing: building super-servers and workgroup systems from commodity software and hardware. Prior to joining Microsoft, he worked at Digital, Tandem, IBM and AT&T on database and transaction processing systems. He is editor of the Performance Handbook for Database and Transaction Processing Systems, and coauthor of Transaction Processing Concepts and Techniques. He holds doctorates from Berkeley and Stuttgart, is a Member of the National Academy of Engineering, Fellow of the ACM, a member of the National Research Council's Computer Science and Telecommunications Board, Trustee of the VLDB Foundation, and Editor of the Morgan Kaufmann series on Data Management and an active editorial advisor for ACM's Queue.