From: | Jon Nelson <jnelson+pgsql(at)jamponi(dot)net> |
---|---|
To: | Greg Smith <greg(at)2ndquadrant(dot)com> |
Cc: | Robert Haas <robertmhaas(at)gmail(dot)com>, Andres Freund <andres(at)2ndquadrant(dot)com>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: fallocate / posix_fallocate for new WAL file creation (etc...) |
Date: | 2013-07-01 15:24:44 |
Message-ID: | CAKuK5J0sCoLOZ=COdptKQbjO-GTSznB2G5Sc_7-=2bgC1vjSRQ@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Sun, Jun 30, 2013 at 11:52 PM, Greg Smith <greg(at)2ndquadrant(dot)com> wrote:
> 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:
If you found yourself with a spare moment, could you run these again
with the number of open/close cycles set high (say, 100) and the
number of rewrites set to 0 and also to 1? Most of the time spent is
actually spent overwriting the files so by reducing or eliminating
that aspect it might be easier to get a handle on the actual
performance difference.
--
Jon
From | Date | Subject | |
---|---|---|---|
Next Message | Bernd Helmle | 2013-07-01 15:34:24 | Re: Passing fdw_private data from PlanForeignScan to PlanForeignModify |
Previous Message | David Fetter | 2013-07-01 15:15:17 | Re: Documentation/help for materialized and recursive views |