Re: IceBreak as an alternative to Apache?

[Thorbjoern Ravn Andersen <ravn@xxxxxxxxxx>]

Den 31/12/10 00.12, Niels Liisberg skrev:
> Yep !! and this is the core problem. Now - icebreak starts one process for each session - and keeps it; now successive request for any request in that particular browser session it the routed to that process. This vital - no process invocation is required for each browser-TCP/IP disconnect
I am aware that having one process for each session gives good 
performance, but it nags me a bit about large workloads and you've 
triggered my curiosity.  Do you have any scalability numbers on 
iceBreak?    What is the maximum number of concurrent users that this 
design can handle (does it hit a "you cannot have any more processes, 
sorry" error message eventually, or is it other factors that show?).  
Also what is the raw speed in terms of requests per second and other 
I/O?  What is the largest network pipe you can saturate and with what 
hardware?

It is interesting to use to compare our Java based solution to this.  
And yes, Java is larger and - perhaps - slower, but for those things we 
do, I need the full Java technology stack in the web server :)