Re: [Webware-discuss] WebKit benchmarks

[Chuck E. <ec...@mi...>]

At 11:43 AM 9/1/00 -0400, Geoff Talvola wrote:
>Hmmm, you have a faster machine than I do, yet your numbers are much 
>worse, and
>I'm using verbose output to boot.  Also, I don't get any exceptions even with
>hundreds of simultaneous requests.  Maybe WinNT is a lot better at socket IO
>and/or threading than Win98?  Not that you'd ever want to deploy anything on
>Win98 anyhow.

I agree with your take on it, that NT probably has better threading and I/O 
than 98.


>Are you running the app server in a DOS window?  If so, be sure to 
>minimize the
>window; keeping it visible takes up a lot of CPU, at least on NT.

I don't think this is a factor if there is no output. Let me try...

No, I get the same thing. It's your console output that's slowing things 
down and even if hidden there's still stuff going on there. But I guess it 
makes sense to leave it on if you would actually deploy with it.

I thought it was interesting that my runs of Async went from 18 req/sec to 
4 when I went for 2 simultaneous requests.

-Chuck