Re: [Gluster-users] GlusterFS Preformance
On Thu, 9 Jul 2009 09:33:59 +0100 Hiren Joshi j...@moonfruit.com wrote: -Original Message- From: Stephan von Krawczynski [mailto:sk...@ithnet.com] Sent: 09 July 2009 09:08 To: Liam Slusser Cc: Hiren Joshi; gluster-users@gluster.org Subject: Re: [Gluster-users] GlusterFS Preformance On Wed, 8 Jul 2009 10:05:58 -0700 Liam Slusser lslus...@gmail.com wrote: You have to remember that when you are writing with NFS you're writing to one node, where as your gluster setup below is copying the same data to two nodes; so you're doubling the bandwidth. Dont expect nfs like performance on writing with multiple storage bricks. However read performance should be quite good. liam Do you think this problem can be solved by using 2 storage bricks on two different network cards on the client? I'd be surprised if the bottleneck here was the network. I'm testing on a xen network but I've only been given one eth per slice. Do you mean your clients and servers are virtual XEN installations (on the same physical box) ? Regards, Stephan ___ Gluster-users mailing list Gluster-users@gluster.org http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users
Re: [Gluster-users] GlusterFS Preformance
-Original Message- From: Stephan von Krawczynski [mailto:sk...@ithnet.com] Sent: 09 July 2009 13:50 To: Hiren Joshi Cc: Liam Slusser; gluster-users@gluster.org Subject: Re: [Gluster-users] GlusterFS Preformance On Thu, 9 Jul 2009 09:33:59 +0100 Hiren Joshi j...@moonfruit.com wrote: -Original Message- From: Stephan von Krawczynski [mailto:sk...@ithnet.com] Sent: 09 July 2009 09:08 To: Liam Slusser Cc: Hiren Joshi; gluster-users@gluster.org Subject: Re: [Gluster-users] GlusterFS Preformance On Wed, 8 Jul 2009 10:05:58 -0700 Liam Slusser lslus...@gmail.com wrote: You have to remember that when you are writing with NFS you're writing to one node, where as your gluster setup below is copying the same data to two nodes; so you're doubling the bandwidth. Dont expect nfs like performance on writing with multiple storage bricks. However read performance should be quite good. liam Do you think this problem can be solved by using 2 storage bricks on two different network cards on the client? I'd be surprised if the bottleneck here was the network. I'm testing on a xen network but I've only been given one eth per slice. Do you mean your clients and servers are virtual XEN installations (on the same physical box) ? They are on different boxes and using different disks (don't ask), this seemed like a good way to evaluate as I setup an NFS server using the same equipment to get relative timings. The plan is to roll it out onto new physical boxes in a month or 2 Regards, Stephan ___ Gluster-users mailing list Gluster-users@gluster.org http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users
Re: [Gluster-users] GlusterFS Preformance
Just a not, we have seen a pretty significant increase in speed from this latest 2.03 release. Doing a test read over afr we are seeing speeds between 200-320 mB a second. (over infiniband, ib-verbs) This is with direct IO disabled too. Oddly putting performance translators on the clients made no difference. We cued up 10 servers to read a single file simultaneously and got about ~30-50 mB a sec from each client, totaling up to ~400 mB a sec on a single file (read from 2 servers, by 10 servers). We are really happy with these numbers for a virtual cluster so, most single drives won't read at those speeds. Thanks! -Mic Hiren Joshi wrote: -Original Message- From: Stephan von Krawczynski [mailto:sk...@ithnet.com] Sent: 09 July 2009 13:50 To: Hiren Joshi Cc: Liam Slusser; gluster-users@gluster.org Subject: Re: [Gluster-users] GlusterFS Preformance On Thu, 9 Jul 2009 09:33:59 +0100 Hiren Joshi j...@moonfruit.com wrote: -Original Message- From: Stephan von Krawczynski [mailto:sk...@ithnet.com] Sent: 09 July 2009 09:08 To: Liam Slusser Cc: Hiren Joshi; gluster-users@gluster.org Subject: Re: [Gluster-users] GlusterFS Preformance On Wed, 8 Jul 2009 10:05:58 -0700 Liam Slusser lslus...@gmail.com wrote: You have to remember that when you are writing with NFS you're writing to one node, where as your gluster setup below is copying the same data to two nodes; so you're doubling the bandwidth. Dont expect nfs like performance on writing with multiple storage bricks. However read performance should be quite good. liam Do you think this problem can be solved by using 2 storage bricks on two different network cards on the client? I'd be surprised if the bottleneck here was the network. I'm testing on a xen network but I've only been given one eth per slice. Do you mean your clients and servers are virtual XEN installations (on the same physical box) ? They are on different boxes and using different disks (don't ask), this seemed like a good way to evaluate as I setup an NFS server using the same equipment to get relative timings. The plan is to roll it out onto new physical boxes in a month or 2 Regards, Stephan ___ Gluster-users mailing list Gluster-users@gluster.org http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users -- ___ Gluster-users mailing list Gluster-users@gluster.org http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users
Re: [Gluster-users] GlusterFS Preformance
You have to remember that when you are writing with NFS you're writing to one node, where as your gluster setup below is copying the same data to two nodes; so you're doubling the bandwidth. Dont expect nfs like performance on writing with multiple storage bricks. However read performance should be quite good. liam On Wed, Jul 8, 2009 at 5:22 AM, Hiren Joshi j...@moonfruit.com wrote: Hi, I'm currently evaluating gluster with the intention of replacing our current setup and have a few questions: At the moment, we have a large SAN which is split into 10 partitions and served out via NFS. For gluster, I was thinking 12 nodes to make up about 6TB (mirrored so that's 1TB per node) and served out using gluster. What sort of filesystem should I be using for the nodes (currently on ext3) to give me the best performance and recoverability? Also, I setup a test with a simple mirrored pair with a client that looks like: volume glust3 type protocol/client option transport-type tcp/client option remote-host glust3 option remote-port 6996 option remote-subvolume brick end-volume volume glust4 type protocol/client option transport-type tcp/client option remote-host glust4 option remote-port 6996 option remote-subvolume brick end-volume volume mirror1 type cluster/replicate subvolumes glust3 glust4 end-volume volume writebehind type performance/write-behind option window-size 1MB subvolumes mirror1 end-volume volume cache type performance/io-cache option cache-size 512MB subvolumes writebehind end-volume I ran a basic test by writing 1G to an NFS server and this gluster pair: [r...@glust1 ~]# time dd if=/dev/zero of=/mnt/glust2_nfs/nfs_test bs=65536 count=15625 15625+0 records in 15625+0 records out 102400 bytes (1.0 GB) copied, 1718.16 seconds, 596 kB/s real28m38.278s user0m0.010s sys 0m0.650s [r...@glust1 ~]# time dd if=/dev/zero of=/mnt/glust/glust_test bs=65536 count=15625 15625+0 records in 15625+0 records out 102400 bytes (1.0 GB) copied, 3572.31 seconds, 287 kB/s real59m32.745s user0m0.010s sys 0m0.010s With it taking almost twice as long, can I expect this sort of performance degradation on 'real' servers? Also, what sort of setup would you recommend for us? Can anyone help? Thanks, Josh. ___ Gluster-users mailing list Gluster-users@gluster.org http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users ___ Gluster-users mailing list Gluster-users@gluster.org http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users