On Fri, 7 Nov 2003 Michael.James@csiro.au wrote: > We have a problem with 66 nodes becoming NFS bound > when blasting many (>10,000) sequences > against the same database set. > > One approach (which we are trying) is to cache database files locally, > so nodes can re-read their files without bottlenecking on NFS. > > A totally different approach, with even better performance potential, > would be if a blast process could start up, load its database(s) > and process multiple queries until told to exit. Solaris has a CacheFS filesystem used to cache NFS mostly read filesystems on the local disk for slow connections including PPP. This would be ideal for your situation, however I don't know if Linux has anything similar. http://docs.sun.com/db/doc/806-4073/6jd67r9jd?a=view Of course this requires a local disk. If you're running diskless, it's going to be a bit more difficult.