Re: [HACKERS] pg_test_fsync performance

2012-02-15 Thread Magnus Hagander
On Wed, Feb 15, 2012 at 02:23, Bruce Momjian br...@momjian.us wrote: On Wed, Feb 15, 2012 at 01:35:05AM +0200, Marko Kreen wrote: On Tue, Feb 14, 2012 at 05:59:06PM -0500, Tom Lane wrote: Bruce Momjian br...@momjian.us writes: On Mon, Feb 13, 2012 at 08:28:03PM -0500, Tom Lane wrote: +1,

Re: [HACKERS] pg_test_fsync performance

2012-02-15 Thread Bruce Momjian
On Wed, Feb 15, 2012 at 09:54:04AM +0100, Magnus Hagander wrote: On Wed, Feb 15, 2012 at 02:23, Bruce Momjian br...@momjian.us wrote: On Wed, Feb 15, 2012 at 01:35:05AM +0200, Marko Kreen wrote: On Tue, Feb 14, 2012 at 05:59:06PM -0500, Tom Lane wrote: Bruce Momjian br...@momjian.us

Re: [HACKERS] pg_test_fsync performance

2012-02-15 Thread Bruce Momjian
On Tue, Feb 14, 2012 at 08:23:10PM -0500, Bruce Momjian wrote: On Wed, Feb 15, 2012 at 01:35:05AM +0200, Marko Kreen wrote: On Tue, Feb 14, 2012 at 05:59:06PM -0500, Tom Lane wrote: Bruce Momjian br...@momjian.us writes: On Mon, Feb 13, 2012 at 08:28:03PM -0500, Tom Lane wrote: +1, I

Re: [HACKERS] pg_test_fsync performance

2012-02-15 Thread Magnus Hagander
On Wed, Feb 15, 2012 at 16:14, Bruce Momjian br...@momjian.us wrote: On Wed, Feb 15, 2012 at 09:54:04AM +0100, Magnus Hagander wrote: On Wed, Feb 15, 2012 at 02:23, Bruce Momjian br...@momjian.us wrote: On Wed, Feb 15, 2012 at 01:35:05AM +0200, Marko Kreen wrote: On Tue, Feb 14, 2012 at

Re: [HACKERS] pg_test_fsync performance

2012-02-14 Thread Bruce Momjian
On Mon, Feb 13, 2012 at 09:54:06PM -0500, Bruce Momjian wrote: On Mon, Feb 13, 2012 at 08:28:03PM -0500, Tom Lane wrote: Robert Haas robertmh...@gmail.com writes: Instead of or in addition to a fixed number operations per test, maybe we should cut off each test after a certain amount of

Re: [HACKERS] pg_test_fsync performance

2012-02-14 Thread Tom Lane
Bruce Momjian br...@momjian.us writes: On Mon, Feb 13, 2012 at 08:28:03PM -0500, Tom Lane wrote: +1, I was about to suggest the same thing. Running any of these tests for a fixed number of iterations will result in drastic degradation of accuracy as soon as the machine's behavior changes

Re: [HACKERS] pg_test_fsync performance

2012-02-14 Thread Marko Kreen
On Tue, Feb 14, 2012 at 05:59:06PM -0500, Tom Lane wrote: Bruce Momjian br...@momjian.us writes: On Mon, Feb 13, 2012 at 08:28:03PM -0500, Tom Lane wrote: +1, I was about to suggest the same thing. Running any of these tests for a fixed number of iterations will result in drastic

Re: [HACKERS] pg_test_fsync performance

2012-02-14 Thread Bruce Momjian
On Wed, Feb 15, 2012 at 01:35:05AM +0200, Marko Kreen wrote: On Tue, Feb 14, 2012 at 05:59:06PM -0500, Tom Lane wrote: Bruce Momjian br...@momjian.us writes: On Mon, Feb 13, 2012 at 08:28:03PM -0500, Tom Lane wrote: +1, I was about to suggest the same thing. Running any of these tests

[HACKERS] pg_test_fsync performance

2012-02-13 Thread Bruce Momjian
I have heard complaints that /contrib/pg_test_fsync is too slow. I thought it was impossible to speed up pg_test_fsync without reducing its accuracy. However, now that I some consumer-grade SATA 2 drives, I noticed that the slowness is really in the open_sync test: Compare open_sync

Re: [HACKERS] pg_test_fsync performance

2012-02-13 Thread Robert Haas
On Mon, Feb 13, 2012 at 7:42 PM, Bruce Momjian br...@momjian.us wrote: I have heard complaints that /contrib/pg_test_fsync is too slow.  I thought it was impossible to speed up pg_test_fsync without reducing its accuracy. However, now that I some consumer-grade SATA 2 drives, I noticed that

Re: [HACKERS] pg_test_fsync performance

2012-02-13 Thread Tom Lane
Robert Haas robertmh...@gmail.com writes: Instead of or in addition to a fixed number operations per test, maybe we should cut off each test after a certain amount of wall-clock time, like 15 seconds. +1, I was about to suggest the same thing. Running any of these tests for a fixed number of

Re: [HACKERS] pg_test_fsync performance

2012-02-13 Thread Bruce Momjian
On Mon, Feb 13, 2012 at 08:28:03PM -0500, Tom Lane wrote: Robert Haas robertmh...@gmail.com writes: Instead of or in addition to a fixed number operations per test, maybe we should cut off each test after a certain amount of wall-clock time, like 15 seconds. +1, I was about to suggest