| From: | Rodrigo Gonzalez <rjgonzale(at)estrads(dot)com(dot)ar> |
|---|---|
| To: | <david(dot)sahagian(at)emc(dot)com> |
| Cc: | <pgsql-general(at)postgresql(dot)org> |
| Subject: | Re: 9.0 EXPLAIN Buffers: written=nnnn |
| Date: | 2012-02-02 17:37:34 |
| Message-ID: | 20120202153734.47a3c21c@estrads.com.ar |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
El Thu, 2 Feb 2012 12:03:20 -0500
<david(dot)sahagian(at)emc(dot)com> escribió:
> Do EXPLAIN ANALYZE:
>
> written=2325 !!!!!
>
> "Total runtime: 375542.347 ms"
>
>
> Then Do EXPLAIN ANALYZE again:
> In general, can you tell me why [written=2325] is displayed by the
> first EXPLAIN, but not the second EXPLAIN ? The query is a SELECT, so
> what is getting "written" ?
>
> Note that both EXPLAINs have the exact same plan. Bottom up is: Seq
> Scan on YYYY, Hash, Seq Scan on XXXX, Hash Join, HashAggregate
>
> I am wondering if it is a clue as to why the "actual time" is so
> different from one run to the next.
>
I can be completely wrong, but maybe it is about hint bits.
Read http://wiki.postgresql.org/wiki/Hint_Bits
Thanks
Rodrigo
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Christophe Pettus | 2012-02-02 18:59:58 | xlog min recovery request ... is past current point ... |
| Previous Message | david.sahagian | 2012-02-02 17:03:20 | 9.0 EXPLAIN Buffers: written=nnnn |