Re: rtree/gist index taking enormous amount of space in 8.2.3

From: "Dolafi, Tom" <dolafit(at)janelia(dot)hhmi(dot)org>
To: "Teodor Sigaev" <teodor(at)sigaev(dot)ru>, "Tom Lane" <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: <pgsql-performance(at)postgresql(dot)org>, "Oleg Bartunov" <oleg(at)sai(dot)msu(dot)su>
Subject: Re: rtree/gist index taking enormous amount of space in 8.2.3
Date: 2007-07-09 21:48:21
Message-ID: AE9860225100F14D87B26D0D4D6766DB4EA627@EXCHANGE03.janelia.priv
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

Thank you for the patch. The index size is back down to 500MB and there
are no performance issues with queries against the table.

-----Original Message-----
From: Teodor Sigaev [mailto:teodor(at)sigaev(dot)ru]
Sent: Friday, July 06, 2007 8:08 AM
To: Tom Lane
Cc: Dolafi, Tom; pgsql-performance(at)postgresql(dot)org; Oleg Bartunov
Subject: Re: [PERFORM] rtree/gist index taking enormous amount of space
in 8.2.3

> Oleg, Teodor, can this be improved?
Attached patch improves creation of index for similar corner cases. And
split
algorithm still demonstrates O(n).

It possible to make fallback to Guttman's split algorithm in corner
cases, but I
don't like this: used linear algorithm is much faster and usually has
better
performance in search.

--
Teodor Sigaev E-mail: teodor(at)sigaev(dot)ru
WWW:
http://www.sigaev.ru/

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Heiko W.Rupp 2007-07-10 08:00:43 Re: PostgreSQL publishes first real benchmark
Previous Message Jim C. Nasby 2007-07-09 20:53:08 Re: PostgreSQL publishes first real benchmark