Re: IceBreak as an alternative to Apache?

[Thorbjoern Ravn Andersen <ravn@xxxxxxxxxx>]

Den 31/12/10 21.38, Aaron Bartell skrev:
> I am not sure what the max clients is, but I do know that when I have done
> testing with JMeter I hit the TCP layer bottleneck before I hit the OS job
> limitation.  Actually, that was a test for stateless transactions.  I
> basically spun up a couple hundred Apache jobs and hit it hard with JMeter.
>   That's when things started falling apart and I had to call IBM.  They of
> course told me it was my software, so I had to put together a bare bones CGI
> app and prove it wasn't.  Then they got one of the higher level devs on and
> they pretty much determined it was at the TCP layer.  You can adjust the TCP
> layer settings, but they recommended against it.
Am I right in understanding the above to be that there is a limitation 
in the TCP/IP implementation for the i?  Is this still a case for V7R1?  
(in other words could you redo the test :) )

When I read the above the first time, I thought you were overflowing the 
buffer in the networking layer for incoming, but not yet accepted socket 
connections (which I believe is around 20), but now I am not so sure any 
more.  Could you elaborate on what you were seeing?