Re: Query help

From: "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov>
To: "Subbiah Stalin-XCGF84" <SSubbiah(at)Motorola(dot)com>, <pgsql-performance(at)postgresql(dot)org>
Subject: Re: Query help
Date: 2009-08-03 19:48:00
Message-ID: 4A76F8A002000025000292B2@gw.wicourts.gov
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

"Subbiah Stalin-XCGF84" <SSubbiah(at)Motorola(dot)com> wrote:

> Not sure what's wrong in below execution plan but at times the query
> runs for 5 minutes to complete and after a while it runs within a
> second or two.

The plan doesn't look entirely unreasonable for the given query,
although it's hard to be sure of that without seeing the table
definitions. Given the plan, the times look to be about what I'd
expect for uncached and cached timings. (That is, on subsequent runs,
the data is sitting in RAM, so you don't need to access the hard
drives.)

If the initial run time is unacceptable for your environment, and
there's no way to have the cached "primed" when it matters, please
give more details on your table layouts, and perhaps someone can make
a useful suggestion.

> Pg 8.2.7, Sol10.

One quick suggestion -- upgrade your PostgreSQL version if at all
possible. The latest bug-fix version of 8.2 is currently 8.2.13, and
there are significant performance improvements in 8.3 and the newly
released 8.4.

-Kevin

In response to

  • Query help at 2009-08-03 17:09:40 from Subbiah Stalin-XCGF84

Responses

Browse pgsql-performance by date

  From Date Subject
Next Message Subbiah Stalin-XCGF84 2009-08-03 20:12:50 Re: Query help
Previous Message Scott Carey 2009-08-03 19:28:32 Re: PostgreSQL 8.4 performance tuning questions