| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Marios Vodas <mvodas(at)gmail(dot)com> |
| Cc: | pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: Temporary tables and in-memory use |
| Date: | 2011-09-29 16:55:29 |
| Message-ID: | 401.1317315329@sss.pgh.pa.us |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Marios Vodas <mvodas(at)gmail(dot)com> writes:
> If I'm not wrong, temporary tables stay in memory if they do not go over
> temp_buffers limit (e.g. if temp_buffers is 2GB and the size of the table is
> 300MB the table will remain in memory).
> What if a column is variable length (e.g. text), how does this column stay
> in-memory since it should be stored in TOAST?
Well, the toast table is also temp, so it'll get cached in temp_buffers
as well, as long as it fits.
> When I build a GiST index on a temporary table does the index stay in memory
> as well?
Same answer.
Keep in mind that temp_buffers is per process, not global. Just as with
work_mem, you need to be careful about setting it sky-high.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Steve Crawford | 2011-09-29 16:55:54 | Re: pg_upgrade - add config directory setting |
| Previous Message | Marios Vodas | 2011-09-29 16:47:53 | Temporary tables and in-memory use |