dcsimg
resolving the database access bottlenecks in stateless architectures
0 posts in topic
Flat View  Flat View
TOPIC ACTIONS:
 

Posted By:   Anonymous
Posted On:   Monday, April 2, 2007 05:24 PM

I'm working with an applications that uses stateless session beans (EJB 2.1), no entity beans and component managed database access via JDBC connection pooling. I would like some ideas for minimizing the number of times the app server accesses the database (considering that the technical architect does not want to use features which would tie the architecture to a specific vendors database). So far I've come up with:- 1. Using JDBC connection pooling 2. Tuning the JDBC statement cache 3. Using JDBC statement batching 4. Moving the database as close to the app server as possible 5. Using caching software such as that from Gigaspaces and Tangosol 6. Use of ANSI standard statements such as materialized views to bulk more logic into single stat   More>>

I'm working with an applications that uses stateless session beans (EJB 2.1), no entity beans and component managed database access via JDBC connection pooling. I would like some ideas for minimizing the number of times the app server accesses the database (considering that the technical architect does not want to use features which would tie the architecture to a specific vendors database). So far I've come up with:-

1. Using JDBC connection pooling 2. Tuning the JDBC statement cache 3. Using JDBC statement batching 4. Moving the database as close to the app server as possible 5. Using caching software such as that from Gigaspaces and Tangosol
6. Use of ANSI standard statements such as materialized views to bulk more logic into single statements.

Is there anything else I could consider also would 3rd party caching software such as that listed in my final point help or am I barking up the wrong tree. Note that I don't want to use entity beans.



Please use html tags to format code blocks.

   <<Less
About | Sitemap | Contact