| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Paul Lindner <lindner(at)inuus(dot)com> |
| Cc: | pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: MIN() performance regression 8.0 -> 8.1 |
| Date: | 2005-12-05 03:05:35 |
| Message-ID: | 12394.1133751935@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Paul Lindner <lindner(at)inuus(dot)com> writes:
> I believe I've found a performance regression between 8.0 and 8.1 for
> some cases. For some queries it appears that the old MIN() ran much
> faster than the new optimized MIN().
When you are complaining about planner mistakes, showing EXPLAIN rather
than EXPLAIN ANALYZE output is an entirely ineffective way of making
your point.
I suspect the problem is that there is a correlation between the
moo_summary_date and moo_summary_b_id columns, which the planner
will not realize --- is that the case?
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2005-12-05 03:07:38 | Re: Reducing relation locking overhead |
| Previous Message | Tom Lane | 2005-12-05 02:50:15 | port/snprintf.c (was Re: Numeric 508 datatype) |