| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Samuel Gendler <sgendler(at)ideasculptor(dot)com> |
| Cc: | pgsql-performance(at)postgresql(dot)org |
| Subject: | Re: in-memory sorting |
| Date: | 2010-08-19 13:29:39 |
| Message-ID: | 26184.1282224579@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-performance |
Samuel Gendler <sgendler(at)ideasculptor(dot)com> writes:
> Answered my own question. Cranking work_mem up to 350MB revealed that
> the in-memory sort requires more memory than the disk sort.
Yeah. The on-disk representation of sortable data is tighter than the
in-memory representation for various reasons, mostly that we're willing
to work at making it small. Datums aren't necessarily properly aligned
for example, and there's also palloc overhead to consider in-memory.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Kevin Grittner | 2010-08-19 13:41:38 | Re: Fwd: Vacuum Full + Cluster + Vacuum full = non removable dead rows |
| Previous Message | Alexandre de Arruda Paes | 2010-08-19 12:57:12 | Re: Fwd: Vacuum Full + Cluster + Vacuum full = non removable dead rows |