| From: | Yeb Havinga <yebhavinga(at)gmail(dot)com> |
|---|---|
| To: | Bart Samwel <bart(at)samwel(dot)tk> |
| Cc: | Robert Haas <robertmhaas(at)gmail(dot)com>, Jeroen Vermeulen <jtv(at)xs4all(dot)nl>, pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: Avoiding bad prepared-statement plans. |
| Date: | 2010-02-11 13:41:02 |
| Message-ID: | 4B7408EE.5020905@gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Bart Samwel wrote:
> Perhaps this could be based on a (configurable?) ratio of observed
> planning time and projected execution time. I mean, if planning it the
> first time took 30 ms and projected execution time is 1 ms, then by
> all means NEVER re-plan.
IMHO looking at ms is bad for this 'possible replan' decision. The only
comparable numbers invariant to system load are the planners costs (not
in ms but unitless) and maybe actual number of processed tuples, but
never actual ms.
Regards,
Yeb Havinga
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Dimitri Fontaine | 2010-02-11 13:41:15 | Re: Re: [COMMITTERS] pgsql: Make standby server continuously retry restoring the next WAL |
| Previous Message | Heikki Linnakangas | 2010-02-11 13:28:51 | Re: Re: [COMMITTERS] pgsql: Make standby server continuously retry restoring the next WAL |