> The client side session bean pool would be for client side remote
references
> to session beans--(client side pooling of remote references), not to the
> actual bean class, which is the container's job (server side pooling of
> beans). Creating them ahead of time would be benefical to performance
since
> your incurring your JNDI lookup, etc, ahead of time.
I get it now. So, instead of a on-demand lookup, you are considering an
early lookup.
> I'm not sure what part of the spec your referring to: I checked page 39 of
> the 1.1 spec. Can you point out the paragraph your referring to? A
client
> uses the remote interface. The remote interface points to a
> java.ejb.EJBObject running on the server. The EJBObject is dedicated to
the
> client. The EJBObject wraps the actual bean. In the case of stateless
> beans, that bean is swapped between EJBObjects.
>
> So, in the case of a stateless session bean, it is swapped to server many
> clients, but the wrapper EJBObject is dedicated to the one client.
I interpreted your mail as referring to the client side - not the server
side swapping.
This clarifies.
Cheers.
Subbu
===========================================================================
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".
