Re: optimizing postgres - Mailing list pgsql-general

From Tom Lane
Subject Re: optimizing postgres
Date
Msg-id 7825.1184290747@sss.pgh.pa.us
Whole thread Raw
In response to optimizing postgres  (lawpoop@gmail.com)
List pgsql-general
lawpoop@gmail.com writes:
> It turned out he was right for our current set up. When I needed to
> empty the project table to re-parse data, doing a cascading delete
> could take up to 10 minutes!

You mean ON CASCADE DELETE foreign keys?  Usually the reason that's
slow is you forgot to put an index on the referencing column.  PG
doesn't force you to have such an index, but unless the referenced
table is nearly static you'll want one.

I too am fairly suspicious of the N-tables-are-faster-than-another-
key-column mindset, but you'd need to do some actual experimentation
(with correctly optimized table definitions ;-)) to be sure.

            regards, tom lane

pgsql-general by date:

Previous
From: Stephen Frost
Date:
Subject: Re: question on scan of clustered index
Next
From: Tom Lane
Date:
Subject: Re: How to pg_dumpall without root password