Manish Gupte wrote: > > Hi ! > > we are planning to set up a linux cluster. > We intend to store a large amount of data on it. we > would need multiple passes of the database for the > algorithm that would run on the cluster. >
If you're going to make some passes on a database, you might want something more than a distributed file system. Have a look at the parallel I/O lib in LAM distribution, it has a debian package. > Can anyone suggest which file system we could use? > how is the file system plugged into linux? You could use GFS, PVFS and other older stuff, but those would not address the parallel application you suggest. Thanks, -- Eray (exa) Ozkural Comp. Sci. Dept., Bilkent University, Ankara e-mail: [EMAIL PROTECTED] www: http://www.cs.bilkent.edu.tr/~erayo
