rishi pathak wrote: > I came across this www.gluster.org <http://www.gluster.org> > Has any one tried it . > Is it a true parallel file system allowing concurrent read and write to > a file by many processes. > Will it be suitable for HPC applications. > >
I wouldn't call GlusterFS a parallel filesystem in the same way I would refer to Lustre or PVFS. GlusterFS is a distributed filesystem, where complete files are contained on one of multiple servers. Although it supports striping, even they say striping for their implementation is bad (http://www.gluster.org/docs/index.php/GlusterFS_FAQ#Why_is_striping_bad.3F). Because of GlusterFS's modular architecture it was easy for them to implement. They do have MPI-IO support on their roadmap, so maybe they are planning to work around the issues described in the link above in user space. GlusterFS is much more like Ibrix or Netapp/GX than Lustre. It seems best as a distributed NFS replacement. In my minimal testing, performance scales linearly as you add data servers. Metadata performance is reasonable (by feel, not by actual measurements). Some of the more interesting features that GlusterFS supports include automatic file replication (AFR), layered performance translators for both client and server side, ability to support heterogeneous storage servers, and is really easy to setup and maintain. I don't have a Lustre setup ready to make any apples to apples comparisons though. However, I believe that the two products fit two different needs. Also, file systems are hard and take a long time to stabilize. Lustre has put in its time, and we are now seeing the benefits. GlusterFS is less mature. Note, comments above are from some basic testing that I have done. I am not a GlusterFS developer. Craig > -- > Regards-- > Rishi Pathak > > > ------------------------------------------------------------------------ > > _______________________________________________ > Lustre-discuss mailing list > [email protected] > http://lists.lustre.org/mailman/listinfo/lustre-discuss -- Craig Tierney ([EMAIL PROTECTED]) _______________________________________________ Lustre-discuss mailing list [email protected] http://lists.lustre.org/mailman/listinfo/lustre-discuss
