Re: is pg_autovacuum so effective ? - Mailing list pgsql-performance

From Gaetano Mendola
Subject Re: is pg_autovacuum so effective ?
Date
Msg-id 421E556E.6010202@mbigroup.it
Whole thread Raw
In response to Re: is pg_autovacuum so effective ?  (Markus Schaber <schabios@logi-track.com>)
Responses Re: is pg_autovacuum so effective ?
List pgsql-performance
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Markus Schaber wrote:
> Hi, Gaetano,
>
> Gaetano Mendola schrieb:
>
>
>>I have the same requirement too. Actually pg_autovacuum can not be
>>instructed "per table" so some time the global settings are not good
>>enough. I have a table of logs with 6 milions rows ( 3 years logs )
>>I insert on that page ~ 6000 rows for day. I'm running pg_autovacuum
>>with setting to ANALYZE or VACUUM table if the 10% is touched.
>>With this setting pg_autovacuum will analyze that table each 3 months!!!
>
>
> If you have only inserts, and only so few on a large table, you do not
> need to vacuum such often. Not to reclaim space, only to prevent
> transaction ID wraparound (which is ensured by pg_autovacuum).
>
> And if the data distribution does not change, frequently calling ANALYZE
> does not help much, either.

Yes, I'm  aware about it indeed I need the analyze because usualy I do on that
table select regarding last 24 ours so need to analyze it in order to
collect the statistics for this period.
Beside that I tried to partition that table, I used both tecnique on
my knowledge

1) A view with UNION ALL on all tables collecting these logs
2) Using inheritance

and both cases are working in theory but in practice are not ( the index scan
is lost as soon you use this view/table inside others views or joining them)

I heard that next version of pg_autovacuum can be instructed "per table";
is it true ?


Regards
Gaetano Mendola










-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (MingW32)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFCHlVu7UpzwH2SGd4RAqQfAKCatX9qbf5fmTN7RbapWj6BgAcwQgCfRy2R
ApeFl9jezm/4YyVN/4fY3Jg=
=wBIK
-----END PGP SIGNATURE-----


pgsql-performance by date:

Previous
From: Glenn Kusardi
Date:
Subject: PostgreSQL 7.4.3 Performance issues on OpenBSD
Next
From: Tom Arthurs
Date:
Subject: PG block sizes