Deploying production VMs running C ++ executable on Azure

I am presented with a program developed in C ++ that implements a computationally intensive algorithm that uses mpi for better results. The executable was tested on a single 16-core VM on Azure and the results were satisfactory. Now we need to test its performance and scalability on 64 or 128 cores. As far as I know, a single VM cannot use more than 16 cores, so I think I need to implement a set of VMs that will do the computation, but I don't know where to start with deployment and communication among VMs. Any advice or ideas would be appreciated.