| From: | Jon Nelson <jnelson+pgsql(at)jamponi(dot)net> |
|---|---|
| To: | |
| Cc: | pgsql-performance(at)postgresql(dot)org |
| Subject: | Re: odd postgresql performance (excessive lseek) |
| Date: | 2010-10-19 13:38:30 |
| Message-ID: | AANLkTimJmKx+3PZjO=7Z_d2Ts144RjnVh+RSYwfe9Aw5@mail.gmail.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-performance |
On Tue, Oct 19, 2010 at 8:25 AM, Merlin Moncure <mmoncure(at)gmail(dot)com> wrote:
> On Tue, Oct 19, 2010 at 9:10 AM, Jon Nelson <jnelson+pgsql(at)jamponi(dot)net> wrote:
>> No replies?
>>
>> This is another situation where using pread would have saved a lot of
>> time and sped things up a bit, but failing that, keeping track of the
>> file position ourselves and only lseek'ing when necessary would also
>> help. Postgresql was spending 37% of it's time in redundant lseek!
>
> 37% of cpu time? Is that according to strace -T? how did you measure it?
Per the original post, it (redundant lseek system calls) accounted for
37% of the time spent in the kernel.
strace -f -p <pid> -c
--
Jon
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Greg Spiegelberg | 2010-10-19 14:12:48 | Re: how to get the total number of records in report |
| Previous Message | Merlin Moncure | 2010-10-19 13:25:20 | Re: odd postgresql performance (excessive lseek) |