| From: | "Simon Riggs" <simon(at)2ndquadrant(dot)com> |
|---|---|
| To: | "Timothy J(dot) Kordas" <tkordas(at)greenplum(dot)com> |
| Cc: | "Tom Lane" <tgl(at)sss(dot)pgh(dot)pa(dot)us>, <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: hash join hashtable size and work_mem |
| Date: | 2007-03-14 21:41:49 |
| Message-ID: | 1173908510.4160.76.camel@silverbirch.site |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Wed, 2007-03-14 at 10:28 -0700, Timothy J. Kordas wrote:
> I would expect for the same data a hash-join with a work_mem of 256MB
> to run faster than one run with 32MB; even if the inner relation is
> only 30MB.
Certainly not for all data, but for some distrubutions yes, probably.
The easiest thing to do is prove thats true and then work out how to
spot that case ahead of time, or at least find a place where you can
adjust your assumptions cheaply enough to improve things.
--
Simon Riggs
EnterpriseDB http://www.enterprisedb.com
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Alvaro Herrera | 2007-03-14 22:45:55 | how to add seconds to a TimestampTz |
| Previous Message | Eddie Stanley | 2007-03-14 19:24:59 | Re: My honours project - databases using dynamically attached entity-properties |