> When making comparisons like this, one has to be specific about what
they
> are comparing.  I think the following are fair statements:
>
> "An object graph comprised of entity beans will ordinarily be slightly
> slower at CRUD operations than the same object graph implemented using
> hand-written and tuned SQL"

Agree, but Java developers do not have to be good SQL programmers able to
tune their SQL statements.

>
> "A remote call to an entity bean will have the same performance overhead
as
> a remote call to another type of server, all other things constant"

Yes and no. Many app servers have the ability to transform these remote
calls to local calls when running in the same JVM.

> "From a performance and development-time point of view, other O/R
frameworks
> would be expected to provide similar performance"
>
> and
>
> "Using entity beans allows the changing of application servers (and
hence
> EJB implementations) without a complete code rewrite"

True, The Server Side has a good example running. They run their site
currently of a cluster of 2 different J2EE 1.3 app servers (WLS & Oracle).
AFIK, they only had to alter the vendor deployment descriptors.

===========================================================================
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".

Reply via email to