Building Petabyte Databases -- Commodity SuperComputing and SuperStorage

Jim Gray
Microsoft Research


I have been building inexpensive mega-servers from commodity components. The most successful of these have been the TerraServer, the PetaBumps demo, and the 5,000$ Terabyte (see I have also been working with the Sloan Digital Sky Survey and more generally with the National Virtual Observatory to build a distributed system to store and analyze historical and current astronomy datasets. This talk will survey each of these efforts and discuss some of our successes and failures.


Jim Gray is part of Microsoft's research group. His work focuses on databases and transaction processing. Jim is active in the research community, is an ACM and NAE Fellow and received the Turing Award for his work on transaction processing. He is also a member of the PITAC, and an editor of a series of books on data management.