Thanks for your interest in FastBit. There are a few different possibilities of parallelizing FastBit. Unfortunately, we don't have any of them in place yet. A trivial approach that might be useful for some tasks is to split your input data into multiple data partitions (i.e., splitting them into multiple directories).
John On 6/10/2010 9:57 AM, Matthieu Dorier wrote: > Hello, > > I am a new user of FastBit and I would like to know if it is possible to > split an index into small fixe-sized files. > The idea is that I have very large datasets and indexing them will > generate also very large index files. I have a storage layer that allows > me to store fixe-sized pages among a number of nodes on a supercomputer, > and I would like to be able to perform my queries in parallel on a > distributed index. Is it possible? If not, do you think it can be > implemented, and what files should I take a look at? > Thank you > > Best regards > > Matthieu Dorier > > -- > Matthieu Dorier > ENS Cachan, antenne de Bretagne > Département informatique et télécommunication > http://perso.eleves.bretagne.ens-cachan.fr/~mdori307/ > > > > _______________________________________________ > FastBit-users mailing list > [email protected] > https://hpcrdm.lbl.gov/cgi-bin/mailman/listinfo/fastbit-users _______________________________________________ FastBit-users mailing list [email protected] https://hpcrdm.lbl.gov/cgi-bin/mailman/listinfo/fastbit-users
