Hi all,
We are using resin 2.1, MSSQL Server and the OPTA 2000 jdbc driver on
our production systems. We use hibernate for our persistence strategy.
The application is mostly read only with very few write operations to
the database.
The DBA's have started complaining that the number of queries being
fired is too high. Every 'real' call to the database is accompanied by a
call to spPrepare, spExecute and spUnprepare. I am guessing this means
that all queries being fired from the application are getting complied
at the sqlserver and then being executed. We need a way to fix this
problem
Our current thought is mainly to bring in prepared statement caching so
that these extra calls to the database (spPrepare and spUnprepare) can
be avoided.
My Questions
1. Is my understanding correct that with the current configuration resin
would return Driver connections and we not get the benefits of statement
caching?
2. Looking into resin 2 documentation I don't see an easy way to
configure prepared statement cache. It seems that prepared statement
caching was introduced in resin 3.0.4. Is there a way to leverage the
same concept on resin 2. The only acceptable value for res-type seems to
be javax.sql.DataSource and javax.sql.XADataSource. We dont have a need
for XADataSource and would like a way to configure a plain simple
ConnectionPoolDataSource. Our resin.conf looks like this
<resource-ref>
<!-- refactor this -->
<res-ref-name>jdbc/DBPOOL</res-ref-name>
<res-type>javax.sql.DataSource</res-type>
<init-param driver-name="com.inet.tds.TdsDriver"/>
<init-param
url="jdbc:inetdae7:10.209.25.128:1433?databaseName=APP_DEV"/>
<init-param user="APP_dev"/>
<init-param password="APP_dev"/>
<init-param max-connections="20"/>
<init-param max-idle-time="30"/>
</resource-ref>
3. Any other recommendation / suggestions ?
Thanks for your time in advance
Anand Raman
_______________________________________________
resin-interest mailing list
resin-interest@caucho.com
http://maillist.caucho.com/mailman/listinfo/resin-interest
