Object Pooling

[ashok ks]

What is Object Pooling ? Is it possible thru weblogic 7.0 ? If yes, how?

[Vinod John]

It would be much more useful for you to search for these topics on the net and read the information than posting on a forum.
Here is what I found when I "Googled" on "object pool"
http://www.google.com/search?sourceid=navclient&ie=UTF-8&oe=UTF-8&q=object+pool

[Lasse Koskela]

Object pooling is the act of reusing a limited number of objects to serve a much larger number of clients.
For example, a web container might have 100 threads (threads are also objects for a Java application) at its disposal but it still might have to serve 1000 end-users with their web browsers. The way this is done is that the incoming HTTP requests are put into a queue and they stay there until the container allocates one of the 100 threads for processing that particular HTTP request. Once the thread is done with the request, it returns back to the pool and the container allocates it to the next waiting HTTP request in the queue.
You could take a look at the Jakarta Commons Pool project, which provides an API for implementing object pools into your application.

[Chris Mathews]

For the majority of uses object pooling is a red herring and adds an unnecessary level of complexity to the application. VMs are designed to create objects and they do so very well. Adding a articficial layer of object caching on top of this is not going to help performance much if it helps at all.
The obvious exception to this is things like JDBC Connection Pools, or any situation where the creation of an object is very expensive and yet once it is created it can live on and serve multiple clients for a long period of time.