Gunnar R|nning wrote:
>
> Hello,
>
> I deploying a JServ based application for a client under Redhat Linux 6.2
> using java 1.2 with native threads.
>
> Until today I actually thought that JServ had thread pooling code, but
> turning the switch in the configuration on didn't turn on the pooling. So I
> looked into the source and CVS and found that the use of Stefanos thread
> pooling code actually had been removed with the comment(in CVS) that it
> didn't do much to speed things up without doing other improvements as well.
??? I was not aware of this! Was this done thru a votation? Why I was
not contacted for this?
> I didn't fall for that argument without any data to back it up, so I put
> the thread pooling code back in and reran my load testing scripts on the
> application.
>
> The huge effect of the thread pooling code suprises me quite a bit. Under
> load the application has a mean response time off ~640ms with thread
> pooling and ~1040ms without the pooling. The throughput also increased from
> about 45 requests/second without pooling to 60 requests/second with
> pooling.
Yeah, the thread pool *is* significant on almost any circumstances.
> Maybe time to put pooling back into the code for those of us that need it ?
+1
--
Stefano Mazzocchi One must still have chaos in oneself to be
able to give birth to a dancing star.
<[EMAIL PROTECTED]> Friedrich Nietzsche
--------------------------------------------------------------------
Missed us in Orlando? Make it up with ApacheCON Europe in London!
------------------------- http://ApacheCon.Com ---------------------
--
----------------------------------------------------------
To subscribe: [EMAIL PROTECTED]
To unsubscribe: [EMAIL PROTECTED]
Archives and Other: <http://java.apache.org/main/mail.html>
Problems?: [EMAIL PROTECTED]
