I apologize if I'm repeating myself, but I received no confirmation from
the list that my last few posts have gone through.
I'm still a little confused on what problems EJB is supposed to solve:
Let me illustrate my confusion by discussing the two application servers I
know something about, NAS/Kiva and WebLogic.
The Netscape Application Server offers me a whole slew of features that
enable me to write highly scalable high performance web applications. To
name a few:
1. High speed distributed state layer, continuously updated on all machines
via "heartbeats"
2. Database result-set caching, minimizing round trips to the DB
3. Highly customizable HTML caching -- when a request comes into the
system, if it has already been
computed it will be transparently redirected to the machine that has it
in cache. Virtually no
processing occurs. Raw data also has a relocatable cache.
Sophisticated cache control API.
4. Highly configurable load balancing algorithms.
Now I'm using WebLogic, whose EJB support, as I understand it, is supposed
to offer me a more open way of accessing these kind of services. But I
don't see where EJB addresses any of these questions.
1. Distributed state layer -- not part of EJB.
2. Result set caching is a property of JDBC driver, not EJB. Of course the
rows themselves can be cached in a session bean, if I write it.
3. Caching -- WebLogic supports single row caching to some degree using
CMP, or I could implement it with BMP. This cache appears not to be
distributed. There is no API for any moderately sophisticated cache controls.
4. Load balancing -- it seems almost meaningless without distributed state.
My question is, what do EJBs actually do (besides distributed transactions)?
-Ben
===========================================================================
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff EJB-INTEREST". For general help, send email to
[EMAIL PROTECTED] and include in the body of the message "help".
