Running encoder with lamexec

[johnnysoj]

Hi,  
I was hoping someone could clarify some stuff for me.

My setup works great for the octave examples,

I'm trying to run a video encoder using lamexec on a 4 node Pelican Cluster.
My binaries are installed in /home/user/bin, and my data drive is an nfs mounted hard drive on the origin node.  


I'm running the command lamexec N /home/user/bin/encoder-binary --param1 --param2 --paramn

The command looks like it executes on all the nodes, but it seems like they're not distributing the load across all four nodes.  It looks like each node is running the task individually.

Can pelican HPC split the load of a binary designed to work on a single system?  Or does the application need to be "pelican" aware?

Thanks!

[Michael Creel]

Your code has to be explicitly parallel, it won't be parallelized automatically. As you are running it, you are just running copies of the same thing on the compute nodes.

Examine the octave examples, you'll see how the parallelization is done. The work that is done depends on the mpi rank of the worker the code runs on.

There is some parallel mp3 code available, see http://www.osl.iu.edu/~jsquyres/bladeenc/
I don't know about the quality of the resulting mp3s, but it's a place to get started.