Re: V4 of PITR performance improvement for 8.4 - Mailing list pgsql-hackers

From Gregory Stark
Subject Re: V4 of PITR performance improvement for 8.4
Date
Msg-id 8763k3lgyy.fsf@oxford.xeocode.com
Whole thread Raw
In response to Re: V4 of PITR performance improvement for 8.4  (Koichi Suzuki <koichi.szk@gmail.com>)
Responses Re: V4 of PITR performance improvement for 8.4  (Koichi Suzuki <koichi.szk@gmail.com>)
Re: V4 of PITR performance improvement for 8.4  (Robert Haas <robertmhaas@gmail.com>)
List pgsql-hackers
Koichi Suzuki <koichi.szk@gmail.com> writes:

> Please find enclosed 2nd patch of pg_readahead which include a patch
> to bufer manager to skip prefetch of pages already in shared buffer.

I'm a bit confused by this comment. PrefetchBuffer already checks if the page
is in shared buffers. 

What is tricky to avoid is prefetching the same page twice -- since the first
prefetch doesn't actually put it in shared buffers there's no way to avoid
prefetching it again unless you keep some kind of hash of recently prefetched
buffers.

For the index scan case I'm debating about whether to add such a cache
directly to PrefetchBuffer -- in which case it would remember if some other
scan prefetched the same buffer -- or to keep it in the index scan code.

--  Gregory Stark EnterpriseDB          http://www.enterprisedb.com Ask me about EnterpriseDB's On-Demand Production
Tuning


pgsql-hackers by date:

Previous
From: Grzegorz Jaskiewicz
Date:
Subject: Re: Hot standby, dropping a tablespace
Next
From: KaiGai Kohei
Date:
Subject: Re: SE-PostgreSQL Updated Revision (r1460)