> The plans are different and I suspect thats where the differences lie.
> For brevity (and readability) I won't copy the plans here but I'll
> provide the links. Search for 'PERF1.POWER.Q2' in the file, it's the
> second query executed and you'll notice the differences under the
Yeah, the difference seems to be an arbitrary choice of which table to join on
first. The error is only +10 ms, but when it's being executed 3000
Will delve into this when I can re-format it, and try to figure out why the
planner is choosing what it is.
Aglio Database Solutions
pgsql-hackers by date
|Next:||From: Thomas F.O'Connell||Date: 2004-10-27 00:09:22|
|Subject: Re: [PATCHES] [HACKERS] ARC Memory Usage analysis|
|Previous:||From: Tom Lane||Date: 2004-10-26 22:44:18|
|Subject: Should bgwriter log checkpoint start/end?|