I know you're not paying attention to settings, but I feel I should note that the VW server is set to progressively throttle connections with a default upper bound of 300 (on the assumption that it is better to serve stably than to try and handle as much as possible). So an upper bound of 270 sounds like it could be a throttling artifact. And I believe VW's file I/O is also VM blocking, so it doesn't seem like that has to be the major obstacle.

And, as an aside, I agree with Stef about the end of stream notification as an exception. It's really annoying, and I hope we will be able to get rid of it one of these days.

At 04:22 PM 1/4/2006, David Shaffer wrote:
>Avi Bryant wrote:
>
>>
>> Out of curiosity, how does VW compare?
>>
>
>http://minnow.cc.gatech.edu/squeak/539
>
>VW is relatively stable around 2-3 ms response time for servlets for my
>example.   Note that by moving the polling frequency down to once every
>ms you get Squeak's response times to be relatively equal to VW for
>servlets.  If you use Async file I/O you get file serving performance on
>par with VW as well.
>
>It is also interesting to look at a break down of the time for the first
>response and the total transfer time.  I haven't yet done that detailed
>of an analysis but I have the tools and plan to do it.
>
>I did a fair amount of benchmarking of an earlier WebToolkit version
>(http://www.cincomsmalltalk.com/CincomSmalltalkWiki/Web+Toolkit+Benchmarks-CDS)
>but these are definitely out of date.
>
>David

--
Alan Knight [|], Cincom Smalltalk Development
knight at acm.org
aknight at cincom.com
http://www.cincom.com/smalltalk

"The Static Typing Philosophy: Make it fast. Make it right. Make it run." - Niall Ross