i agree with u jim.
lets decide on some test scenarios .
lets write a entity bena first may be cmp or bmp.we'll test both
but can u tell me what all r the performance factors which we r about to
consider.
instead of just putting in words,lets try to implement it practically
Arul
----- Original Message -----
From: "James Cook" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, July 25, 2000 6:50 PM
Subject: Re: ready for fun in comparing app servers with EJB
> Arul,
>
> Despite the two server choices that you have made, I believe that an
> *easily* deployable EJB test suite is needed. I would be happy to work
with
> you to establish some test scenarios. I don't believe that an overall
> benchmark (like TPC-W) has much merit for reasons that William Louth and
> others have stated in this thread. Applications vary too much for such a
> benchmark to be useful.
>
> I do believe that individual benchmarks on marshalling issues, concurrency
> demands, simple transactions, spec compliance, security, etc. are much
more
> telling. Perhaps these individual ratings can be combined with some
> documented weight to form some kind of EJB-SPEC rating.
>
> Regardless, if these tests can be easily distibuted with a proper JUnit
test
> harness, we can leave the running of the benchmarks up to the individual
> developers. It would also be interesting if the results of running these
> benchmark tests can be auto-broadcasted back to a web server and a running
> tally of scores (and particular hardware constraints) can be updated live
on
> the web site. I think this may even be the loophole in the app server's
> restrictions regarding posting benchmark scores, since we aren't running
any
> benchmarks!
>
> jim
>
>
> ----- Original Message -----
> From: "Arul" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Tuesday, July 25, 2000 12:21 AM
> Subject: Re: ready for fun in comparing app servers with EJB
>
>
> > Hello JIM.
> >
> > What i had told is not an anecdotal evidence.
> > just take a look at the chart at the ibm site
> >
> >
>
http://www-4.ibm.com/software/developer/library/java-pollresults/appserver.h
> > tml
> >
> > only this made me to take into consideration Weblogic and Websphere
> >
> > Arul
> > ----- Original Message -----
> > From: "James Cook" <[EMAIL PROTECTED]>
> > To: <[EMAIL PROTECTED]>
> > Sent: Monday, July 24, 2000 6:15 PM
> > Subject: Re: ready for fun in comparing app servers with EJB
> >
> >
> > > ----- Original Message -----
> > > From: "Arul" <[EMAIL PROTECTED]>
> > > > i thought that initially i would consider only two app servers and
as
> > per
> > > > research, the two i mentioned are the mostly used one.So i suggested
> > those
> > > > two servers.
> > >
> > > Just curious Arul, but have you come across some independent source
> > > (gartner, idc, zona, etc.) that shows market share for Appservers? Or
is
> > > this just anecdotal evidence?
> > >
> > > jim
> > >
> > >
> >
>
===========================================================================
> > > 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".
> >
> >
>
===========================================================================
> > 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".
> >
> >
>
>
===========================================================================
> 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".
>
===========================================================================
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".
