Hi,
1) My opinion is that you couldn't compare SOAP
and RMI because SOAP is a protocol definition.
So you must compare RMI with a SOAP implementation.
2) SOAP is just near 1,5 year old and you must
be patient to get a good SOAP implementation
3) In some case, XML parsers are not efficient,
especially on Java implementation. In the most
cases, people use DOM objects wich are pratical
but heavy objects !
More, XML validation has a big cost in term of
time (More with Schema. See 4).
4) One side of SOAP is to specify datatypes
with XML schema. XML Schema becomes a final
recommendation the 02 May 2001, so, today, Schema
validating parsers are just designed to be
reliable and not to be quick !!
When these parsers will be reliable they will
be optimized.
5) RMI is Java world, SOAP is design for
an "idealistic open world" ...
Jean-Louis.
> -----Message d'origine-----
> De : Michael R. Clements [mailto:[EMAIL PROTECTED]]
> Envoy� : mardi 26 juin 2001 22:18
> � : [EMAIL PROTECTED]
> Objet : RE: SOAP Performance (against RMI)
>
>
> It's hard to compare performance because the test results would depend on
> the amount and nature of the data marshaled across the function calls. I
> would suspect that RMI's use of standard Java serialization is more
> efficient than building and parsing XML strings, thus I would
> expect RMI to
> be proportionally faster than SOAP as the amount and complexity
> of the data
> increases.
>
> However, the existing XML parsers are woefully inefficient, which
> would seem
> to indicate that there is room for potentially large performance
> gains. Thus
> while I would expect RMI to maintain a performance advantage over
> SOAP both
> today and in the long term, I would expect the degree of this advantage to
> shrink over time.
>
> In other words, if RMI is 6 times faster than SOAP today, it might only be
> twice as fast a year from now.
>
> But one must consider that if the people in charge of the SOAP standards
> allow the XML format for SOAP to bloat with new features (and complexity),
> this will oppose the gains made from faster XML parsers and could destroy
> any chance of making SOAP more efficient.
>
> Given this possibility, it is impossible to predict how RMI and SOAP will
> continue to compare in performance. We can only hope that the
> SOAP standards
> people keep a tight lid on the XML format and we can realize the
> advantages
> of more efficient parsers.
>
> Regards,
>
> -----Original Message-----
> From: Hansen, Richard [mailto:[EMAIL PROTECTED]]
> Sent: Tuesday, June 26, 2001 12:06
> To: '[EMAIL PROTECTED]'
> Subject: RE: SOAP Performance (against RMI)
>
>
> And SOAP will just be get faster over time. RMI probably won't.
>
> > -----Original Message-----
> > From: SBC [mailto:[EMAIL PROTECTED]]
> > Sent: Tuesday, June 26, 2001 1:56 PM
> > To: [EMAIL PROTECTED]
> > Subject: Re: SOAP Performance (against RMI)
> >
> >
> > I did a simple test & found that RMI is 6 times
> > faster. I used apache 2.2 with wls 6.0. I would not
> > call it a real benchmark, but it gave me some idea.
> >
> > --- Ralf Bierig <[EMAIL PROTECTED]> wrote:
> > > Hi,
> > >
> > > are there any performance measurement materials
> > > about
> > > SOAP against RMI in Web? Did somebody made a
> > > benchmarktest with SOAP (and maybe RMI)?
> > >
> > > I am looking for material to determine, if SOAP is
> > > good enough to fullfil the requirements I need for a
> > > project.
> > >
> > > Greetings
> > > Ralf
> > >
> > > __________________________________________________
> > > Do You Yahoo!?
> > > Get personalized email addresses from Yahoo! Mail
> > > http://personal.mail.yahoo.com/
> >
> >
> > __________________________________________________
> > Do You Yahoo!?
> > Get personalized email addresses from Yahoo! Mail - only $35
> > a year! http://personal.mail.yahoo.com/
> >
>
