Re: Prepared Statements

From: Kris Jurka <books(at)ejurka(dot)com>
To: mljv(at)planwerk6(dot)de
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: Prepared Statements
Date: 2008-01-09 20:16:18
Message-ID: Pine.BSO.4.64.0801091507500.27353@leary.csoft.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Wed, 9 Jan 2008, mljv(at)planwerk6(dot)de wrote:

> - I know there is a PREPARE Statement in Postgresql and read the docs.
> - in PostgresqlJDBC i have a prepareThreshold parameter which i left to
> default of 5.
> - in DBCP i have a property "poolPreparedStatements", set to true. Does ist
> just configure prepareThreshold of JDBC or does it maintain a statementPool
> of it's own?

The postgresql JDBC driver does not have a statement pool of its own, so
the two options prepareThreshold and poolPreparedStatements are
complementary. By itself the JDBC driver will switch to a long term
prepared plan once you've used the exact same PreparedStatement object
prepareThreshold number of times. Since it doesn't pool things behind the
scenes, you've got to retain a reference to the same PreparedStatement
object which is difficult or impossible in many applications. This is
where the DBCP statement pool comes in. It proxies PreparedStatement
wrappers to one underlying PG PreparedStatement object so that you can hit
prepareThreshold.

> In my Log files of postgresql each query is called like this:
>
> EXECUTE <unnamed> [PREPARE: select ...]
>
> I have not found anything about preparing "unnamed" statements. What
> does it mean?
>

Unnamed statements are what the driver uses before it hits the
prepareThreshold limit. Once it has determined the statement will be
reused many times it changes to a named statement that has a longer
lifespan.

Kris Jurka

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Nikolay Samokhvalov 2008-01-09 20:47:54 Re: XML path function
Previous Message Ivan Sergio Borgonovo 2008-01-09 20:09:36 Re: count(*) and bad design was: Experiences with extensibility