you have not specified in your question whether you want to parallelize a single massive MASON simulation
or you would like to rather to run a huge number of single process simulations on a computing cluster.
For the first scenario (a single simulation run that is so huge it cannot fit a single machine) take a look a D-MASON.
D-MASON enables running geographically distributed simulation model on a multi-worker cluster.
However, most likely, in your question you meant running in parallel a very large number of smaller (i.e.. single process/single thread) simulations. This is a very common scenario if you for example need to perform a parameter sweep.
For this scenario I usually use Sun Grid Engine (SGE) - now called Open Grid Scheduler. My university does not own a computing cluster
and hence I run my simulations in the cloud on Amazon Web Services. The tool that makes SGE installation in the Amazon cloud an easily manageable process is called StarCluster. If you decide to use Amazon EC2 Spot instances this approach is extremely cost effective - you can run your computations with cost around 0.01$ per one Xeon vCPU per hour. I usually spin up a cluster of 320 vCPUs and pay $2.90 an hour...
Starcluster requires some dependencies and has several configuration / installation steps.
Since I am doing it many times, some time ago I wrote a script to automate the process, you can get it at:
Hopefully, the above information is enough for the start.
If you have more detailed questions write to me and I will put answer on the MASON forum.