[erlang-questions] Erlang in a condor cluster

[Andre Nathan]

Hello

Does anyone have experience running erlang using the condor
(http://www.cs.wisc.edu/condor/) cluster manager?

My application consists of a distributed graph (where each vertex is an
erlang process), and to use it I must submit jobs using condor. In the
job submition script,  I would have to start the erlang emulator on each
machine I am given, then from one machine start spawning processes, do
some computation on the graph and then shut down the emulators when the
computation is finished.

What would be the best way to acomplish that? I'm thinking of spawning
an extra process on each machine of the cluster that would call
init:stop() when receiving a 'terminate' message. Is there any better
approach for that?

Thanks in advance,
Andre