Re: The performance problem of server side PreparedStatement?
Posted by: Li wy
Date: December 14, 2004 07:24PM

Mark Matthews wrote:
> The JDBC spec doesn't let the driver implicitly
> cache the creation of a prepared statement, as
> most implementations (including ours) have
> prepared statements tied to a particular
> connection, even inside the server. Therefore
> when you close the connection, they go away.

Mark, Thanks for your answers.
In my original text, I mean closing PreparedStatement, not closing connection. I ever looked the source code of Connector/J 3.0.14. Though that version driver doesn't support sever-side PreparedStatement, it will cache PreparedStatement in the driver with the SQL string as key. But now, for the server-side PreparedStatement, it does nothing.
I know the JDBC spec doesn't require that. However, since your document has described related connection properties, such as cachePrepStmts, prepStmtCacheSize, etc. I think you should keep them effective either for the driver-side or the server-side prepared statement implementations.

Regards,
Liwy

Options: ReplyQuote




Sorry, you can't reply to this topic. It has been closed.

Content reproduced on this site is the property of the respective copyright holders. It is not reviewed in advance by Oracle and does not necessarily represent the opinion of Oracle or any other party.