pg_dump ANALYZE statements - Mailing list pgsql-hackers

From Simon Riggs
Subject pg_dump ANALYZE statements
Date
Msg-id 1169480417.3776.324.camel@silverbirch.site
Whole thread Raw
Responses Re: pg_dump ANALYZE statements  (Tom Lane <tgl@sss.pgh.pa.us>)
Re: pg_dump ANALYZE statements  (Stefan Kaltenbrunner <stefan@kaltenbrunner.cc>)
Re: pg_dump ANALYZE statements  (Gregory Stark <stark@enterprisedb.com>)
Re: pg_dump ANALYZE statements  ("Jim C. Nasby" <jim@nasby.net>)
List pgsql-hackers
There is currently a performance tip to run ANALYZE after a pg_dump
script has been restored.

ISTM that pg_dump should default to have ANALYZE statements for a table,
when both table definition and data are dumped. If only data (-a) or
only table definition (-s) is dumped the default should be to *not* add
an ANALYZE statement.

There doesn't seem to be any reason to skip the ANALYZE, but I'll
implement it as an option. 
-z on | off 
--analyze=on | off

This would add a table-specific ANALYZE statement following each table's
actions.

I'm not aware of a strong argument against such an option. Performance
surely can't be one because the time saved on the ANALYZE will quickly
bite back on time lost on poorly planned queries.

What does the panel think?

--  Simon Riggs              EnterpriseDB   http://www.enterprisedb.com




pgsql-hackers by date:

Previous
From: "Merlin Moncure"
Date:
Subject: Re: savepoint improvements
Next
From: Tom Lane
Date:
Subject: Re: savepoint improvements