Using MPI with CUDA C

Hi I am planning to use tera grid cluster(…64TeslaCluster) to run my Monte Carlo fortran codes,with CUDA C,using fortran wrapper.I am planning to run
it on multiple nodes,each node having some tesla cards.MPI is used to communicate between processors.But I am confused that,when I run my code,each processor will need a GPU to run it’s part of the CUDA C code,hence each might compete for a particular Tesla card.How can I have control over this competition.Is there any way to do that?

What’s the ratio of CPUs to GPUs in a node?

I have got an reply for this question I will try this suggestion.I have cross posted this question,so sorry about that