Re: GiST index performance

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Matthew Wakeling <matthew(at)flymine(dot)org>
Cc: Kevin Grittner <Kevin(dot)Grittner(at)wicourts(dot)gov>, pgsql-performance(at)postgresql(dot)org
Subject: Re: GiST index performance
Date: 2009-04-16 17:59:32
Message-ID: 26222.1239904772@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

Matthew Wakeling <matthew(at)flymine(dot)org> writes:
> On Thu, 16 Apr 2009, Tom Lane wrote:
>> Also, what are the physical sizes of the two indexes?

> location_object_start_gist | 193 MB
> location_object_start | 75 MB

>> I notice that the inner nestloop join gets slower too, when it's not
>> changed at all --- that suggests that the overall I/O load is a lot
>> worse, so maybe the reason the query is falling off a performance cliff
>> is that the GIST index fails to fit in cache.

> Memory in the machine is 16GB.

Hmm, and what is shared_buffers set to? How big are the tables and
other indexes used in the query? We still have to explain why the
inner nestloop got slower, and it's hard to see that unless something
stopped fitting in cache.

regards, tom lane

In response to

Responses

Browse pgsql-performance by date

  From Date Subject
Next Message Matthew Wakeling 2009-04-16 18:05:23 Re: GiST index performance
Previous Message Matthew Wakeling 2009-04-16 17:54:05 Re: GiST index performance