Re: Cached Query Plans (was: global prepared statements)

From: Martijn van Oosterhout <kleptog(at)svana(dot)org>
To: PFC <lists(at)peufeu(dot)com>
Cc: "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Cached Query Plans (was: global prepared statements)
Date: 2008-04-13 15:09:07
Message-ID: 20080413150907.GB32735@svana.org
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Sun, Apr 13, 2008 at 02:26:04PM +0200, PFC wrote:
> * global plan cache in shared memory, implemented as hashtable, hash key
> being the (search_path, query_string)
> Doubt : Can a plan be stored in shared memory ? Will it have to be copied
> to local memory before being executed ?

Frankly, I think you're better off storing them in a table. Shared
memory is a limited resource and you cannot change how much you've
allocated after the server has started. It does mean you'll have to
serialise/deserialise them, but this will be cheaper than replanning,
right?

I liked your global prepared statements idea much better. Named the
statements is no problem: DB frontends do that for you anyway
sometimes.

Have a nice day,
--
Martijn van Oosterhout <kleptog(at)svana(dot)org> http://svana.org/kleptog/
> Please line up in a tree and maintain the heap invariant while
> boarding. Thank you for flying nlogn airlines.

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Dawid Kuroczko 2008-04-13 15:31:52 Re: Cached Query Plans (was: global prepared statements)
Previous Message Kris Jurka 2008-04-13 14:25:54 Re: [Pljava-dev] stack depth limit exceeded - patch possible?