| From: | Stefan Kaltenbrunner <stefan(at)kaltenbrunner(dot)cc> |
|---|---|
| To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
| Cc: | Andrew Dunstan <andrew(at)dunslane(dot)net>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: lock contention on parallel COPY ? |
| Date: | 2008-09-26 17:06:31 |
| Message-ID: | 48DD1697.4090505@kaltenbrunner.cc |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Tom Lane wrote:
> Stefan Kaltenbrunner <stefan(at)kaltenbrunner(dot)cc> writes:
>> heh no log archiving - I actually said that I'm now playing with
>> --truncate-before-load which seems to cause a noticeable performance (as
>> in IO generated) increase but I still see >130000 context switches/s and
>> a profile that looks like:
>
>> samples % symbol name
>> 55526 16.5614 LWLockAcquire
>> 29721 8.8647 DoCopy
>> 26581 7.9281 CopyReadLine
>> 25105 7.4879 LWLockRelease
>> 15743 4.6956 PinBuffer
>> 14725 4.3919 heap_formtuple
>
> Still a lot of contention for something, then. You might try turning on
> LWLOCK_STATS (this only requires recompiling storage/lmgr/lwlock.c) to
> get some evidence about what.
that one generates a huge amount of logs - output for ~60s into the load
is available here:
http://www.kaltenbrunner.cc/files/lwstats.txt (21MB!)
Stefan
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Brendan Jurd | 2008-09-26 17:25:48 | Meridiem markers (was: [BUGS] Incorrect "invalid AM/PM string" error from to_timestamp) |
| Previous Message | Tom Lane | 2008-09-26 16:38:14 | Re: lock contention on parallel COPY ? |