>> >I have a question on EJB performance over WAN, in two distinct
scenarios:
>> >
>> >1. Server-to-server, over high capacity WAN links at both ends, and
>> >2. Browser-to-server, over a dial-up or low capacity link at the client
>> end.
>> >
>> >Does anyone have experience regarding performance, reliability, and how
>> >bandwidth efficiency?
>>
>> You'll be a lot happier with the XML-over-HTTP solution
>
>This confuses me. Why would XML-RPC be inherently faster than RMI (over
JRMP),
>RMI-over-IIOP, or CORBA?
There is a misunderstanding here. I do not say XML-RPC is faster
than RMI/JRMP or CORBA/IIOP.
Avi had a concern:
>I am less than pleased with
>these solutions because of the large development overhead they incur.
XML-RPC addresses this concern about **development overhead**
not about **performance**. The idea is that XML-RPC is a lot easier
to develop than XML-and-do-it-yourself method invocation,
marshalling-unmarsahalling
parameters and return values etc.
Actually my full post was:
"You'll be a lot happier with the XML-over-HTTP solution **IF** you check
out
XML-RPC "
So **IF** you must choose the XML-over-HTTP solution for whetever reason, be
it
firewall traversal for instance, **THEN** you'll be happier because of
XML-RPC.
>I can see the benefit of XML-RPC when clients are behind a firewall or
>when your non-Java clients don't want to deal with the complexity of Corba.
But
>other than that, not only would I think that RMI would be faster in these
>situations, it'd be far simpler to write code for.
So we agree in this. XML-RPC is appropiate if you must use protocols other
than JRMP/IIOP/T3 etc Think abour XML-RPC over JMS/MQSeries
Best regards
Javier
===========================================================================
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".
