Re: proposal for smaller indexes on index-ordered tables - Mailing list pgsql-hackers

From Tom Lane
Subject Re: proposal for smaller indexes on index-ordered tables
Date
Msg-id 6669.1214343510@sss.pgh.pa.us
Whole thread Raw
In response to Re: proposal for smaller indexes on index-ordered tables  ("Jeffrey Baker" <jwbaker@gmail.com>)
Responses Re: proposal for smaller indexes on index-ordered tables  ("Jeffrey Baker" <jwbaker@gmail.com>)
List pgsql-hackers
"Jeffrey Baker" <jwbaker@gmail.com> writes:
> I'm quite aware of the problems of maintaining such a table and index, but
> the fact is that data warehouse type tables may never be updated after being
> created.  The particular application I'm struggling with does a SELECT ...
> INTO ... ORDER BY to make an ordered table for querying every night.  The
> problem is it takes longer, much longer, to create the index than to create
> the table, and in the end the index is as big as half the table anyway.

There's something wrong with that: sorting the table rows surely ought
to take longer than sorting the same number of (smaller) index entries.
Have you done any profiling to find out what the problem is?  Perhaps
there's something wrong with the setting of maintenance_work_mem (vs
work_mem).
        regards, tom lane


pgsql-hackers by date:

Previous
From: Zoltan Boszormenyi
Date:
Subject: Re: proposal for smaller indexes on index-ordered tables
Next
From: "Jeffrey Baker"
Date:
Subject: Re: proposal for smaller indexes on index-ordered tables