[Seaside] HTTP Performance

[Pablo Iaría]

Hello Sven,

Very interesting....but I think the evaluation you did seems to be a
bit unfear, just because seaside is much more than jsp or apache. It
could be more adecuate to compare seaside with struts or some other
web application framework to see the "real" deference (and also test
the developent time/effort).

Best regards,
 Pablo


Wednesday, November 19, 2003, 11:32:24 AM, you wrote:

SVC> Let me start by saying that I like Seaside (and Squeak) a lot. I am
SVC> really amazed at how much design and abstraction there is hidden 
SVC> underneath, and how flexible Smalltalk code really is (and how much fun
SVC> it is to work in such a powerful integrated environment).

SVC> However, since I started evaluating writing and deploying web 
SVC> applications in Squeak (only about two weeks ago), the final HTML
SVC> interface as approached by a browser, felt a little bit slow.

SVC> First, I re-implemented our calculator example in Lisp (minus the
SVC> backtracking support), and it was clear that there was a significant
SVC> speed difference: it was very easy to 'out-click' the seaside based
SVC> calculator, which was a lot harder to do with the Lisp version.

SVC> Then I decided to do a simple benchmark to really compare HTTP 
SVC> performance:

SVC>    http://homepage.mac.com/svc/DW-Bench/index.html

SVC> I compared dynamic and static HTTP performance of JBoss, Lisp and
SVC> Smalltalk.

SVC> The Seaside/Comanche results for Squeak were consistent, but slow. I do
SVC> not know enough about the performance characteristics of Squeak to have
SVC> an educated opinion about these results.

SVC> Does anybody care to run the tests themselves ?
SVC> Does anybody, those with production Seaside/Comanche/Squeak based
SVC> sites, care to comment ?
SVC> What can be done to make things faster ?

SVC> Sven
  

SVC> _______________________________________________
SVC> Seaside mailing list
SVC> Seaside at lists.squeakfoundation.org
SVC> http://lists.squeakfoundation.org/listinfo/seaside