| From: | Merlin Moncure <mmoncure(at)gmail(dot)com> |
|---|---|
| To: | Kelphet Xiong <kelphet(at)gmail(dot)com> |
| Cc: | Kevin Grittner <kgrittn(at)ymail(dot)com>, ktm(at)rice(dot)edu, ants(dot)aasma(at)eesti(dot)ee, "pgsql-performance(at)postgresql(dot)org" <pgsql-performance(at)postgresql(dot)org> |
| Subject: | Re: Question about postmaster's CPU usage |
| Date: | 2013-04-01 13:42:42 |
| Message-ID: | CAHyXU0ynObZ78d5Wjm+FSbqr7pB4k_eBBXubiSf7fOp2t69uwQ@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-performance |
On Sat, Mar 30, 2013 at 11:00 PM, Kelphet Xiong <kelphet(at)gmail(dot)com> wrote:
> I guess it is because postgres only uses a single thread to read
> the data or “pushing the data around in RAM” according to Kevin’s statement.
> Then my question is actually why postgres can not use the remaining
> 93.4%CPU.
postgres can use an arbitrary amount of threads to read data, but only
one per database connection.
> Btw, I also tried the command suggested by Ants Aasma, but got an error:
>
> explain (analyze on, timing off) select * from inventory;
> ERROR: syntax error at or near "analyze"
>
> LINE 1: explain (analyze on, timing off) select * from inventory;
>
> ^
Ability to manipulate timing was added in 9.2.
merlin
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Armand du Plessis | 2013-04-01 22:35:32 | Problems with pg_locks explosion |
| Previous Message | Tom Lane | 2013-04-01 00:52:59 | Re: query plan estimate |