Hi,

Here is my usual choices and i did not found reasons to change then yet. Under 
300 hosts, Rush does the job, it easy to install easy to conifgure and there is 
some room for customization.

After 300 hosts things becomes serious and you need to be sure that the system 
will scale gracefully, which means an efficient load balancing scheme and in 
general you need more ressources management features in addition to the process 
management ones. And of course tones of logs to be sure that the thing is 
really efficient. That's really were things like Arsenal and equivalent reach 
their limits. So if you are building a big cluster, have a look at Condor or 
SunGridEngine. My preference goes to Condor but i am sure it is possible to do 
the same things with SGE. In any case both tools are used for years on giant 
computing clusters and you can be sure that it is efficient and scale really 
well.

On the down side, it more work to configure them, but there is actually vendors 
that are used to do that and train your people at the same time. So it might be 
a solution.

Cheers,

Ash



_______________________________________________
Nuke-python mailing list
[email protected], http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python

Reply via email to