Re: fallocate / posix_fallocate for new WAL file creation (etc...)

On 6/30/13 9:28 PM, Jon Nelson wrote:
> The performance of the latter (new) test sometimes seems to perform
> worse and sometimes seems to perform better (usually worse) than
> either of the other two. In all cases, posix_fallocate performs
> better, but I don't have a sufficiently old kernel to test with.

This updated test program looks reliable now. The numbers are very
tight when I'd expect them to be, and there's nowhere with the huge
differences I saw in the earlier test program.

Here's results from a few sets of popular older platforms:

RHEL5, ext3

method: classic. 10 open/close iterations, 10 rewrite in 22.6949s
method: posix_fallocate. 10 open/close iterations, 10 rewrite in 23.0113s
method: glibc emulation. 10 open/close iterations, 10 rewrite in 22.4921s

method: classic. 10 open/close iterations, 10 rewrite in 23.2808s
method: posix_fallocate. 10 open/close iterations, 10 rewrite in 22.4736s
method: glibc emulation. 10 open/close iterations, 10 rewrite in 23.9871s

method: classic. 10 open/close iterations, 10 rewrite in 22.4812s
method: posix_fallocate. 10 open/close iterations, 10 rewrite in 22.2393s
method: glibc emulation. 10 open/close iterations, 10 rewrite in 23.6940s

RHEL6, ext4

method: classic. 10 open/close iterations, 10 rewrite in 56.0483s
method: posix_fallocate. 10 open/close iterations, 10 rewrite in 61.5092s
method: glibc emulation. 10 open/close iterations, 10 rewrite in 53.8569s

method: classic. 10 open/close iterations, 10 rewrite in 57.0361s
method: posix_fallocate. 10 open/close iterations, 10 rewrite in 55.9840s
method: glibc emulation. 10 open/close iterations, 10 rewrite in 64.9437sb

RHEL6, ext3

method: classic. 10 open/close iterations, 10 rewrite in 14.4080s
method: posix_fallocate. 10 open/close iterations, 10 rewrite in 16.1395s
method: glibc emulation. 10 open/close iterations, 10 rewrite in 16.9657s

method: classic. 10 open/close iterations, 10 rewrite in 15.2825s
method: posix_fallocate. 10 open/close iterations, 10 rewrite in 16.5315s
method: glibc emulation. 10 open/close iterations, 10 rewrite in 14.8115s

The win for posix_fallocate is there in most cases, but it's pretty hard
to see in these older systems. That could be OK. As long as the
difference is no more than noise, and that is the case, this could be
good enough to commit. If there are significantly better results on the
new platforms, the old ones need to just not get worse.

--
Greg Smith 2ndQuadrant US greg(at)2ndQuadrant(dot)com Baltimore, MD
PostgreSQL Training, Services, and 24x7 Support www.2ndQuadrant.com