On Thu, 2005-11-10 at 08:37 -0800, David Kirchner wrote: > On 11/10/05, Ron Johnson <[EMAIL PROTECTED]> wrote: > > The issue was performance, though. Are you getting good speed? > > I must have missed that email then, sorry about that. We use RAID5 on I don know if it was my message, but I am having major performance problems in PostgreSQL doing long *updates* (NOT inserts) in largish tables (1.100.0000 records). In general I see only ~100/200 writes/second, totaling ~2000/3000 KILOBYTE/second.
Reading and (batch) inserts perform reasonable well. This all on a Debian Sarge Testing 2.6.12-1-amd64-k8-smp, 2 Opteron 248, 4GB, 5 disk RAID5 array with a 3Ware 9500S-8 > $ dd if=/dev/zero of=/data2-2/bigfile bs=131072 count=100000 > 100000+0 records in > 100000+0 records out > 13107200000 bytes transferred in 119.269480 seconds (109895675 bytes/sec) [EMAIL PROTECTED]:~/tmp$ dd if=/dev/zero of=/home/jkr/tmp/bigfile bs=131072 count=10000 10000+0 records in 10000+0 records out 1310720000 bytes transferred in 14.468310 seconds (90592473 bytes/sec) I did notice that changing the I/O scheduler's nr_request from the default 128 to 1024 or even 4096 made a remarkable performance improvement. I suspect that experimenting with other I/O schedululers could improve performance. More more info see also: http://www.issociate.de/board/goto/894541/3ware_+_RAID5_ +_xfs_performance.html#msg_894541 http://lkml.org/lkml/2005/4/20/110 http://seclists.org/lists/linux-kernel/2005/Oct/1171.html -- Groeten, Joost Kraaijeveld Askesis B.V. Molukkenstraat 14 6524NB Nijmegen tel: 024-3888063 / 06-51855277 fax: 024-3608416 e-mail: [EMAIL PROTECTED] web: www.askesis.nl -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
