I am experimenting with repast4py library to build a supply chain simulation model and it involves large amount of data. My guess is to run this at scale I will require at least 20 cores and 200 GB RAM. The cost of a single virtual machine with this configuration is more expensive than say ten 2 core 10 GB RAM VMs.
Question:
Is it possible to run repast4py model in a databricks cluster with multiple smaller worker nodes than run it in a big single VM?