I need to integrate WebLogic Application Server 6.1 with MQ Series 5.2. There are 2 ways I can connect to IBM MQ Series:
1. Use IBM java client within WL to do PUT and GET from IBM MQ. This option does not use JMS (either WL or IBM). All it does is to get hold of IBM MQ via their Java library and then does a PUT and GET operations on that queue.
2. Use Weblogic JMS along with MDB. MessageDrivenBeans in turn gets hold of IBM MQ via the IBM java client in the same fashion as option 1. However, the use of MDBs in WL allows me to parallelize the PUT operation on IBM MQ. The first options does that serially since it does not use MDB. Even though the second option allows for concurrency of PUT operation, using JMS and MDB may add lot of overhead.
So, my question is would the use of WL JMS and MDB be faster and more scalable than simply using IBM Java client even though using JMS and MDB adds to the overhead. Which option is more scalable and has faster throughput?
This question is very product-specific and you might have better luck visiting the support newsgroup at news: weblogic.developer.interest.jms
Also, the Weblogic support site provides a white paper and some software related to integrating Weblogic with IBM MQ.
Dig Deeper on Topics Archive
Have a question for an expert?
Please add a title for your question
Get answers from a TechTarget expert on whatever's puzzling you.