Hi Arnold, you are right that comparing gluster against a local filesystem is quite unfair. We just used the iozone results to get a number to compare against. What concerns us most is the FINODELK value in gluster profile listing, which has a great latency. What does FINODELK mean and is it something we should be concerned about? So what would be the best use case for gluster in a system with small/medium and often accessed files? Thank you very much Jens Am 09.05.2012 um 11:35 schrieb Arnold Krille: > On Wednesday 09 May 2012 11:01:00 Jens Nauber | tyclipso. net wrote: >> we have set up a replicated gluster volume on 3 glusterservers, 2 of them >> act as well as gluster clients using NFS-Mode. >> >> It seems we have a performance problem especially in read access, compared >> by Iozone-Test for GlusterFS and local filesystem (testfiles attached) > > Comparing a network-filesystem to a local filesystem is a bit unfair, isn't it? > > You "think" reading happens via local disk, but actually reading happens > across the network when you use nfs-mount. > Even when you use the fuse-mount locally, you still have a userspace-component > (which has only a very little influence). And while reading of big contents all > happens locally and is thus fast, accessing files (small or big, read or write) > has an overhead as the attributes of the remote bricks are checked too to see > if the volume is consistent and trigger a self-heal if needed. So while > reading big files from local fuse-mount is fast, reading many small files still > has the network-penalty of any distributed and/or replicated system. > > A better comparison is kernel-nfs against gluster with its userspace-nfs. > And maybe also compare gluster to moosefs or ceph or lustre or any fs above > drbd... > > Have fun, > > Arnold_______________________________________________ > Gluster-users mailing list > Gluster-users at gluster.org > http://gluster.org/cgi-bin/mailman/listinfo/gluster-users -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 4897 bytes Desc: not available URL: <http://gluster.org/pipermail/gluster-users/attachments/20120509/d800c285/attachment.bin>
