directory-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Peter Donald <pe...@realityforge.org>
Subject Re: Serverside NIO: Not ready for Prime Time
Date Thu, 27 Nov 2003 04:50:56 GMT
On Thu, 27 Nov 2003 10:49 am, Alex Karasulu wrote:
> You need statistics and a population of
> clients to determine the net effect in a SEDA based design whether
> stage consolidation really pays off.  Thoughts?

I would go the other way and have as few stages as possible and only break 
down the stages if there is a significant pay off. The basic reason is that 
every stage adds overhead and complexity. In most of the seda-style 
request-response apps I would suggest you start with 5 stages (input, 
decoding, logic, encoding, output). Only add more if you need it or are doing 
significant IO during the logic stage.

-- 
Cheers,

Peter Donald
"All my life I wanted to be someone; I guess I should have been more 
specific."
-- Jane Wagner


Mime
View raw message