Hi,
> > > >efficient to serialize in Java. Apparently JRMP is
> quite a bit more
> > > >efficient/fast than IIOP, and some third-party RMI protocols
> > > (WebSphere's,
> > > >Orion's) are potentially even more efficient/fast.
>
> > Have you run any benchmarks to "prove" this!?! I'm writing
> a generic load
> > test suite with plugins for SOAP, CORBA, RMI, EJB, and local method
> > invocations.
>
> Alas, no... just something I saw in a newsgroup or mailing
> list somewhere.
> I will replace that vague answer with your specific benchmarks :-)
Or even better... run some tests yourself ;-)
> In your JRMP test, did the client load the classes remotely,
> or were they in
> the client's classpath? And does this matter?
I generated the client stubs using rmic and then the JVM loaded them from
the the classpath. How the stubs are obtained should (AFAIK) not matter at
all.
I just re-ran the test, this time using "true" remote invocations over a 10
MB/s LAN (the previous test was between two JVM:s on one host). The servers
ran on RedHat Linux 7.0 using JDK1.3.0 and the clients ran on WinNT 4 sp 5
using JDK1.3.0.
It still seems like RMI/JRMP and CORBA/IIOP are equal WRT performance (and
yes I still believe that, in general, other factors and features such as
interoperability, scalability, being able to dynamically download stubs and
so on... are more important than raw performance...).
Regards,
Per
> Kyle Cordes
> www.kylecordes.com
--
--------------------------------------------------------------
To subscribe: [EMAIL PROTECTED]
To unsubscribe: [EMAIL PROTECTED]
Problems?: [EMAIL PROTECTED]
