In general you use JMS for three purposes:
1. The message receiver is not always accessible (due for instance to the
unreliable nature of the internet) but you do want guaranteed delivery.
2. You would like to give clients the option to subscribe at will
(allthough you probably can do this using rcp's as well, but it is very
easy to implement with JMS).
3. It's so easy ...
Using JMS is pretty straightforward. Message driven beans are even more
straightforward to understand.
You probably don't want to use JMS when you want the two object work
together in a synchronous fashion. A calls B, B throws an exception which A
handles. This kind of stuff is more natural using beans / rmi / etc.
Further more, I have no idea about performance penalties using JMS, so
either is would benchmark is or be cautious with high volumes of messages
with little or none processsing.
My 2cts,
FE
On Friday, May 11, 2001 5:09 PM, Eddie [SMTP:[EMAIL PROTECTED]] wrote:
> SV: CLUSTERING PROBLEM FINALLY SOLVED!! <web-module>Hope to get an answer
on this question:
> Can someone please tell me when/why to use JMS ? and what his impact is
on performance to comparision to for example a function-call of another
bean and application ??
>
> I like to use it but the above is still a bit unclear to me.
>
> Thanks,
> Eddie
> << File: ATT00000.html >>
