On Monday 24 January 2011 22:58 Jeff Sturm wrote > > -----Original Message----- > > From: linux-cluster-bounces@xxxxxxxxxx > > [mailto:linux-cluster-bounces@xxxxxxxxxx] > > > On Behalf Of Wendy Cheng > > Subject: Re: gfs2 v. zfs? > > > > I would love to get an education here. From usage model point of view, > > what is the > > difference between a "parallel file system" and a "cluster file > > system" ? i.e., when to > > use a parallel file system and when to use a cluster file system ? > > Getting off-topic but I'd also like to hear who uses a parallel > distributed FS, and what problem space they work well in. HPC where you need very high bandwidth/throughput to disk (usually scratch filesystem). > GFS works fine, but we have trouble scaling it up due solely to the > limits of central storage. I'm sure there are esoteric storage designs > to get around this, but at some point I also suspect there's an easier > way... >From a HW point of view, DDN has very good storage arrays ... but you pay for it ;) They promise 6 GB/s and you get those 6 GB/s. You can also use more storage arrays and use some volume manager to group them together. This is a bit tricky because it won't guarantee a load balance on the different storage arrays. HTH Rafa -- "We cannot treat computers as Humans. Computers need love." Happily using KDE 4.5.4 :) -- Linux-cluster mailing list Linux-cluster@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/linux-cluster