If we continue looking at memory usage as is the focus of the article, then I suspect that this multi-process model will have the highest memory usage of all th

If we continue looking at memory usage as is the focus of the article, then I suspect that this multi-process model will have the highest memory usage of all the approaches. I’m under the impression that processes are quite heavyweight in resources compared to using threads or async. Also, you may only need the concurrency for a small portion of your code, and the other approaches may be simpler to code and maintain (I’m looking at async).