Multi-processing is useful for CPU heavy tasks such zipping/unzipping and it adds extra layer of security.

However, there are fundamental problems with using multi-processing in python and that is Global Interpreter Lock (GIL) that prevents using native python profilers that would help with time-based performance analysis of our software.

It would be better to use asyncio and event loop instead of multiprocessing. The only CPU heavy usage where we need multiprocessing is storage operations but it could be moved to webserver.go and the rest of the services could be moved to asyncio thread.