* [linux-lvm] lvm2 performance data with linux-2.6 @ 2004-03-08 19:16 ` markw 0 siblings, 0 replies; 14+ messages in thread From: markw @ 2004-03-08 14:15 UTC (permalink / raw) To: linux-lvm; +Cc: linux-kernel I've started collecting various data (including oprofile) using our DBT-2 (OLTP) workload with lvm2 on linux 2.6.2 and 2.6.3 on ia32 and ia64 platforms: http://developer.osdl.org/markw/lvm2/ So far I've only varied the stripe width with lvm, from 8 KB to 512 KB, for PostgreSQL that is using 8 KB sized blocks with ext2. It appears that a stripe width of 16 KB through 128KB on the ia64 system gives the best throughput for the DBT-2 workload on a volume that should be doing mostly sequential writes. I'm going to run through more tests varying the block size that PostgreSQL uses, but I wanted to share what I had so far in case there were other suggestions or recommendations. -- Mark Wong - - markw@osdl.org Open Source Development Lab Inc - A non-profit corporation 12725 SW Millikan Way - Suite 400 - Beaverton, OR 97005 (503) 626-2455 x 32 (office) (503) 626-2436 (fax) http://developer.osdl.org/markw/ ^ permalink raw reply [flat|nested] 14+ messages in thread
* lvm2 performance data with linux-2.6 @ 2004-03-08 19:16 ` markw 0 siblings, 0 replies; 14+ messages in thread From: markw @ 2004-03-08 19:16 UTC (permalink / raw) To: linux-lvm; +Cc: linux-kernel I've started collecting various data (including oprofile) using our DBT-2 (OLTP) workload with lvm2 on linux 2.6.2 and 2.6.3 on ia32 and ia64 platforms: http://developer.osdl.org/markw/lvm2/ So far I've only varied the stripe width with lvm, from 8 KB to 512 KB, for PostgreSQL that is using 8 KB sized blocks with ext2. It appears that a stripe width of 16 KB through 128KB on the ia64 system gives the best throughput for the DBT-2 workload on a volume that should be doing mostly sequential writes. I'm going to run through more tests varying the block size that PostgreSQL uses, but I wanted to share what I had so far in case there were other suggestions or recommendations. -- Mark Wong - - markw@osdl.org Open Source Development Lab Inc - A non-profit corporation 12725 SW Millikan Way - Suite 400 - Beaverton, OR 97005 (503) 626-2455 x 32 (office) (503) 626-2436 (fax) http://developer.osdl.org/markw/ ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: lvm2 performance data with linux-2.6 2004-03-08 19:16 ` markw (?) @ 2004-03-08 19:46 ` Miquel van Smoorenburg -1 siblings, 0 replies; 14+ messages in thread From: Miquel van Smoorenburg @ 2004-03-08 19:46 UTC (permalink / raw) To: linux-kernel In article <200403081916.i28JGgE25794@mail.osdl.org>, <markw@osdl.org> wrote: >I've started collecting various data (including oprofile) using our >DBT-2 (OLTP) workload with lvm2 on linux 2.6.2 and 2.6.3 on ia32 and >ia64 platforms: > http://developer.osdl.org/markw/lvm2/ For write-heavy loads, you might want to see if 2.6.4-rc2-mm1 makes any difference. It fixes a case where both pdflush and a writing process itself would queue writes to an LVM device, resulting in less than optimal I/O ordering for some cases. The fix will probably be in 2.6.4 proper as well. Mike. ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: lvm2 performance data with linux-2.6 2004-03-08 19:16 ` markw (?) (?) @ 2004-03-08 22:01 ` Matthias Andree -1 siblings, 0 replies; 14+ messages in thread From: Matthias Andree @ 2004-03-08 22:01 UTC (permalink / raw) To: linux-kernel On Mon, 08 Mar 2004, markw@osdl.org wrote: > I've started collecting various data (including oprofile) using our > DBT-2 (OLTP) workload with lvm2 on linux 2.6.2 and 2.6.3 on ia32 and > ia64 platforms: > http://developer.osdl.org/markw/lvm2/ > > So far I've only varied the stripe width with lvm, from 8 KB to 512 KB, > for PostgreSQL that is using 8 KB sized blocks with ext2. It appears Does ext2 write 8 KByte sized blocks atomically on IA32? Or is this no requirement for PostgreSQL consistency? -- Matthias Andree Encrypt your mail: my GnuPG key ID is 0x052E7D95 ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: lvm2 performance data with linux-2.6 2004-03-08 19:16 ` markw @ 2004-03-12 10:38 ` Bill Davidsen -1 siblings, 0 replies; 14+ messages in thread From: Bill Davidsen @ 2004-03-11 22:12 UTC (permalink / raw) To: markw; +Cc: linux-lvm, linux-kernel markw@osdl.org wrote: > I've started collecting various data (including oprofile) using our > DBT-2 (OLTP) workload with lvm2 on linux 2.6.2 and 2.6.3 on ia32 and > ia64 platforms: > http://developer.osdl.org/markw/lvm2/ > > So far I've only varied the stripe width with lvm, from 8 KB to 512 KB, > for PostgreSQL that is using 8 KB sized blocks with ext2. It appears > that a stripe width of 16 KB through 128KB on the ia64 system gives the > best throughput for the DBT-2 workload on a volume that should be doing > mostly sequential writes. > > I'm going to run through more tests varying the block size that > PostgreSQL uses, but I wanted to share what I had so far in case there > were other suggestions or recommendations. > Here's one thought: look at the i/o rates on individual drives using each stripe size. You *might* see that one size does far fewer seeks than others, which is a secondary thing to optimize after throughput IMHO. If you don't have a tool for this I can send you the latest diorate which does stuff like this, io rate perdrive or per partition, something I occasionally find revealing. -bill ^ permalink raw reply [flat|nested] 14+ messages in thread
* [linux-lvm] Re: lvm2 performance data with linux-2.6 @ 2004-03-12 10:38 ` Bill Davidsen 0 siblings, 0 replies; 14+ messages in thread From: Bill Davidsen @ 2004-03-12 10:38 UTC (permalink / raw) To: markw; +Cc: linux-lvm, linux-kernel markw@osdl.org wrote: > I've started collecting various data (including oprofile) using our > DBT-2 (OLTP) workload with lvm2 on linux 2.6.2 and 2.6.3 on ia32 and > ia64 platforms: > http://developer.osdl.org/markw/lvm2/ > > So far I've only varied the stripe width with lvm, from 8 KB to 512 KB, > for PostgreSQL that is using 8 KB sized blocks with ext2. It appears > that a stripe width of 16 KB through 128KB on the ia64 system gives the > best throughput for the DBT-2 workload on a volume that should be doing > mostly sequential writes. > > I'm going to run through more tests varying the block size that > PostgreSQL uses, but I wanted to share what I had so far in case there > were other suggestions or recommendations. > Here's one thought: look at the i/o rates on individual drives using each stripe size. You *might* see that one size does far fewer seeks than others, which is a secondary thing to optimize after throughput IMHO. If you don't have a tool for this I can send you the latest diorate which does stuff like this, io rate perdrive or per partition, something I occasionally find revealing. -bill ^ permalink raw reply [flat|nested] 14+ messages in thread
* [linux-lvm] Re: lvm2 performance data with linux-2.6 2004-03-12 10:38 ` [linux-lvm] " Bill Davidsen @ 2004-03-11 22:25 ` Mark Wong -1 siblings, 0 replies; 14+ messages in thread From: Mark Wong @ 2004-03-11 17:23 UTC (permalink / raw) To: Bill Davidsen; +Cc: linux-lvm, linux-kernel On Thu, Mar 11, 2004 at 05:12:35PM -0500, Bill Davidsen wrote: > markw@osdl.org wrote: > > I've started collecting various data (including oprofile) using our > > DBT-2 (OLTP) workload with lvm2 on linux 2.6.2 and 2.6.3 on ia32 and > > ia64 platforms: > > http://developer.osdl.org/markw/lvm2/ > > > > So far I've only varied the stripe width with lvm, from 8 KB to 512 KB, > > for PostgreSQL that is using 8 KB sized blocks with ext2. It appears > > that a stripe width of 16 KB through 128KB on the ia64 system gives the > > best throughput for the DBT-2 workload on a volume that should be doing > > mostly sequential writes. > > > > I'm going to run through more tests varying the block size that > > PostgreSQL uses, but I wanted to share what I had so far in case there > > were other suggestions or recommendations. > > > Here's one thought: look at the i/o rates on individual drives using > each stripe size. You *might* see that one size does far fewer seeks > than others, which is a secondary thing to optimize after throughput IMHO. > > If you don't have a tool for this I can send you the latest diorate > which does stuff like this, io rate perdrive or per partition, something > I occasionally find revealing. Yeah, please do send me a copy. I'd be interested to see what that might turn up. I've just been using iostat -x so far. Thanks, Mark ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: lvm2 performance data with linux-2.6 @ 2004-03-11 22:25 ` Mark Wong 0 siblings, 0 replies; 14+ messages in thread From: Mark Wong @ 2004-03-11 22:25 UTC (permalink / raw) To: Bill Davidsen; +Cc: linux-lvm, linux-kernel On Thu, Mar 11, 2004 at 05:12:35PM -0500, Bill Davidsen wrote: > markw@osdl.org wrote: > > I've started collecting various data (including oprofile) using our > > DBT-2 (OLTP) workload with lvm2 on linux 2.6.2 and 2.6.3 on ia32 and > > ia64 platforms: > > http://developer.osdl.org/markw/lvm2/ > > > > So far I've only varied the stripe width with lvm, from 8 KB to 512 KB, > > for PostgreSQL that is using 8 KB sized blocks with ext2. It appears > > that a stripe width of 16 KB through 128KB on the ia64 system gives the > > best throughput for the DBT-2 workload on a volume that should be doing > > mostly sequential writes. > > > > I'm going to run through more tests varying the block size that > > PostgreSQL uses, but I wanted to share what I had so far in case there > > were other suggestions or recommendations. > > > Here's one thought: look at the i/o rates on individual drives using > each stripe size. You *might* see that one size does far fewer seeks > than others, which is a secondary thing to optimize after throughput IMHO. > > If you don't have a tool for this I can send you the latest diorate > which does stuff like this, io rate perdrive or per partition, something > I occasionally find revealing. Yeah, please do send me a copy. I'd be interested to see what that might turn up. I've just been using iostat -x so far. Thanks, Mark ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [linux-lvm] Re: lvm2 performance data with linux-2.6 2004-03-11 22:25 ` Mark Wong @ 2004-03-11 23:01 ` Chris Croswhite -1 siblings, 0 replies; 14+ messages in thread From: Chris Croswhite @ 2004-03-11 18:01 UTC (permalink / raw) To: linux-lvm; +Cc: Bill Davidsen, linux-kernel Is this generally available to all? Where can it be had? TIA On Thu, 2004-03-11 at 14:25, Mark Wong wrote: > On Thu, Mar 11, 2004 at 05:12:35PM -0500, Bill Davidsen wrote: > > markw@osdl.org wrote: > > > I've started collecting various data (including oprofile) using our > > > DBT-2 (OLTP) workload with lvm2 on linux 2.6.2 and 2.6.3 on ia32 and > > > ia64 platforms: > > > http://developer.osdl.org/markw/lvm2/ > > > > > > So far I've only varied the stripe width with lvm, from 8 KB to 512 KB, > > > for PostgreSQL that is using 8 KB sized blocks with ext2. It appears > > > that a stripe width of 16 KB through 128KB on the ia64 system gives the > > > best throughput for the DBT-2 workload on a volume that should be doing > > > mostly sequential writes. > > > > > > I'm going to run through more tests varying the block size that > > > PostgreSQL uses, but I wanted to share what I had so far in case there > > > were other suggestions or recommendations. > > > > > Here's one thought: look at the i/o rates on individual drives using > > each stripe size. You *might* see that one size does far fewer seeks > > than others, which is a secondary thing to optimize after throughput IMHO. > > > > If you don't have a tool for this I can send you the latest diorate > > which does stuff like this, io rate perdrive or per partition, something > > I occasionally find revealing. > > Yeah, please do send me a copy. I'd be interested to see what that might > turn up. I've just been using iostat -x so far. > > Thanks, > Mark > > _______________________________________________ > linux-lvm mailing list > linux-lvm@redhat.com > https://www.redhat.com/mailman/listinfo/linux-lvm > read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/ ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [linux-lvm] Re: lvm2 performance data with linux-2.6 @ 2004-03-11 23:01 ` Chris Croswhite 0 siblings, 0 replies; 14+ messages in thread From: Chris Croswhite @ 2004-03-11 23:01 UTC (permalink / raw) To: linux-lvm; +Cc: Bill Davidsen, linux-kernel Is this generally available to all? Where can it be had? TIA On Thu, 2004-03-11 at 14:25, Mark Wong wrote: > On Thu, Mar 11, 2004 at 05:12:35PM -0500, Bill Davidsen wrote: > > markw@osdl.org wrote: > > > I've started collecting various data (including oprofile) using our > > > DBT-2 (OLTP) workload with lvm2 on linux 2.6.2 and 2.6.3 on ia32 and > > > ia64 platforms: > > > http://developer.osdl.org/markw/lvm2/ > > > > > > So far I've only varied the stripe width with lvm, from 8 KB to 512 KB, > > > for PostgreSQL that is using 8 KB sized blocks with ext2. It appears > > > that a stripe width of 16 KB through 128KB on the ia64 system gives the > > > best throughput for the DBT-2 workload on a volume that should be doing > > > mostly sequential writes. > > > > > > I'm going to run through more tests varying the block size that > > > PostgreSQL uses, but I wanted to share what I had so far in case there > > > were other suggestions or recommendations. > > > > > Here's one thought: look at the i/o rates on individual drives using > > each stripe size. You *might* see that one size does far fewer seeks > > than others, which is a secondary thing to optimize after throughput IMHO. > > > > If you don't have a tool for this I can send you the latest diorate > > which does stuff like this, io rate perdrive or per partition, something > > I occasionally find revealing. > > Yeah, please do send me a copy. I'd be interested to see what that might > turn up. I've just been using iostat -x so far. > > Thanks, > Mark > > _______________________________________________ > linux-lvm mailing list > linux-lvm@redhat.com > https://www.redhat.com/mailman/listinfo/linux-lvm > read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/ ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [linux-lvm] Re: lvm2 performance data with linux-2.6 2004-03-11 23:01 ` Chris Croswhite @ 2004-03-14 14:02 ` Bill Davidsen -1 siblings, 0 replies; 14+ messages in thread From: Bill Davidsen @ 2004-03-14 14:02 UTC (permalink / raw) To: csc; +Cc: linux-kernel, linux-lvm Chris Croswhite wrote: > Is this generally available to all? Where can it be had? I stuck it on the page I use to avoid beating my fractional T1 to death. http://pages.prodigy.net/davidsen/diorate.pl Enjoy, any misfeatures please feed back to me. -- bill davidsen <davidsen@tmr.com> CTO TMR Associates, Inc Doing interesting things with small computers since 1979 ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [linux-lvm] Re: lvm2 performance data with linux-2.6 @ 2004-03-14 14:02 ` Bill Davidsen 0 siblings, 0 replies; 14+ messages in thread From: Bill Davidsen @ 2004-03-14 14:02 UTC (permalink / raw) To: csc; +Cc: linux-lvm, linux-kernel Chris Croswhite wrote: > Is this generally available to all? Where can it be had? I stuck it on the page I use to avoid beating my fractional T1 to death. http://pages.prodigy.net/davidsen/diorate.pl Enjoy, any misfeatures please feed back to me. -- bill davidsen <davidsen@tmr.com> CTO TMR Associates, Inc Doing interesting things with small computers since 1979 ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: lvm2 performance data with linux-2.6 2004-03-11 22:25 ` Mark Wong (?) (?) @ 2004-03-16 21:42 ` bill davidsen 2004-03-16 23:17 ` markw -1 siblings, 1 reply; 14+ messages in thread From: bill davidsen @ 2004-03-16 21:42 UTC (permalink / raw) To: linux-kernel In article <20040311142515.A27177@osdlab.pdx.osdl.net>, Mark Wong <markw@osdl.org> wrote: | On Thu, Mar 11, 2004 at 05:12:35PM -0500, Bill Davidsen wrote: | > Here's one thought: look at the i/o rates on individual drives using | > each stripe size. You *might* see that one size does far fewer seeks | > than others, which is a secondary thing to optimize after throughput IMHO. | > | > If you don't have a tool for this I can send you the latest diorate | > which does stuff like this, io rate perdrive or per partition, something | > I occasionally find revealing. | | Yeah, please do send me a copy. I'd be interested to see what that might | turn up. I've just been using iostat -x so far. Okay, I posted the pointer a few days ago to LKML, did you get a chance to try it? And if so, did it tell you anything? -- bill davidsen <davidsen@tmr.com> CTO, TMR Associates, Inc Doing interesting things with little computers since 1979. ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: lvm2 performance data with linux-2.6 2004-03-16 21:42 ` bill davidsen @ 2004-03-16 23:17 ` markw 0 siblings, 0 replies; 14+ messages in thread From: markw @ 2004-03-16 23:17 UTC (permalink / raw) To: davidsen; +Cc: linux-kernel On 16 Mar, bill davidsen wrote: > In article <20040311142515.A27177@osdlab.pdx.osdl.net>, > Mark Wong <markw@osdl.org> wrote: > | On Thu, Mar 11, 2004 at 05:12:35PM -0500, Bill Davidsen wrote: > > | > Here's one thought: look at the i/o rates on individual drives using > | > each stripe size. You *might* see that one size does far fewer seeks > | > than others, which is a secondary thing to optimize after throughput IMHO. > | > > | > If you don't have a tool for this I can send you the latest diorate > | > which does stuff like this, io rate perdrive or per partition, something > | > I occasionally find revealing. > | > | Yeah, please do send me a copy. I'd be interested to see what that might > | turn up. I've just been using iostat -x so far. > > Okay, I posted the pointer a few days ago to LKML, did you get a chance > to try it? And if so, did it tell you anything? I've grabbed it but I haven't had a chance to try it yet. Hopefully later today or tomorrow. Mark ^ permalink raw reply [flat|nested] 14+ messages in thread
end of thread, other threads:[~2004-03-16 23:17 UTC | newest] Thread overview: 14+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2004-03-08 14:15 [linux-lvm] lvm2 performance data with linux-2.6 markw 2004-03-08 19:16 ` markw 2004-03-08 19:46 ` Miquel van Smoorenburg 2004-03-08 22:01 ` Matthias Andree 2004-03-11 22:12 ` Bill Davidsen 2004-03-12 10:38 ` [linux-lvm] " Bill Davidsen 2004-03-11 17:23 ` Mark Wong 2004-03-11 22:25 ` Mark Wong 2004-03-11 18:01 ` [linux-lvm] " Chris Croswhite 2004-03-11 23:01 ` Chris Croswhite 2004-03-14 14:02 ` Bill Davidsen 2004-03-14 14:02 ` Bill Davidsen 2004-03-16 21:42 ` bill davidsen 2004-03-16 23:17 ` markw
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.