From: Sekhar Vajjhala <Sekhar.Vajjhala_at_Sun.COM>
Date: Mon, 19 Nov 2007 10:50:41 -0500

Moving the discussion of the issue raised on
http://forums.java.net/jive/thread.jspa?messageID=245496
to this alias. Refer to the thread for a description of the issue.

It looks to me that there are two concepts in WebLogic. - MultiPool
and multi datasources.

a. MultiPool has been replaced by multi data sources - para 3,
   http://e-docs.bea.com/wls/docs90/jdbc_admin/jdbc_intro.html#1044158 .

b. Multi data source is an abstraction over dat sources and provides
   load balancing or failover between data sources. See the para
   under the diagram in
   http://e-docs.bea.com/wls/docs90/jdbc_admin/config.html#1064269 .
   When an application looks up multi data source, a data source is
   selected based on the algorithm configured (
   
http://e-docs.bea.com/wls/docs90/jdbc_admin/jdbc_multidatasources.html#1108831
).

GF has one concept - clustered pool, proposed in
http://wiki.glassfish.java.net/attach/OnePagersOrFunctionalSpecs/ConnectionPoolOnePager.html
.
. Is clustered pool similar to multi data source or MulitPool ? It
seems similar to MutliPool to me. Can you clarify ?

Thanks,
Sekhar