Re: [Condor-users] How many submit nodes in your pool ?

[Nick LeRoy <nleroy@xxxxxxxxxxx>]

On Sunday 04 July 2004 10:36 am, Bruce Beckles wrote:
> Hi everyone -
Hello,

> 	We are currently trying to setup a large Condor pool/flock
> (haven't decided which yet), and there is some debate here about submit
> nodes, so I thought it might be useful to find out what other people are
> doing in this regard.  What I would like to know from people with large
> (hundreds or more machines) pools is the following:
>
> - How large is your pool?
>
> - Do you restrict the number of submit nodes, i.e. do you have only a
>   small (maybe only one) number of submit nodes, or can any machine
>   (within reason) submit jobs to the pool/flock?
>
> - If you do restrict the number of submit nodes is this a problem for, or
>   a source of dissatisfaction amongst, your users or do they not
>   particularly mind?

As a Condor developer, I can give you some input.  In general, more submit 
nodes is better than fewer.  Why, you ask?  For running a lot of simultaneous 
jobs, the schedd and shadow processes can become a significant load on the 
submit machine.  For sufficiently "beafy" machines, this number can be pretty 
large (say 1k), but it can still be a bottleneck.

Another advantage to having multiple submit nodes is that if your single 
submit node goes down, no jobs get run, but with multiple submit nodes, only 
the jobs from this dead machine stop.

Just some food for thought.

-Nick

-- 
           <<< Why, oh, why, didn't I take the blue pill? >>>
 /`-_    Nicholas R. LeRoy               The Condor Project
{     }/ http://www.cs.wisc.edu/~nleroy  http://www.cs.wisc.edu/condor
 \    /  nleroy@xxxxxxxxxxx              The University of Wisconsin
 |_*_|   608-265-5761                    Department of Computer Sciences