I’m currently doing some R&D with Ray Core and have had a great experience so far. I’m looking to justify onboarding Ray to my colleagues who are used to using a homebrewed Python API for grid computing that wraps another homebrewed C#/.NET/Redis service (the architecture is 5+ years old and I don’t have any knowledge of it, but surely it can’t compete with this open source product). For example, Ray’s shared object store is a huge value add in performance for our parallel tasks. Additionally, a large open-source community is always seen as a plus, but what am I missing here? What are some other strong arguments to justify ditching the old framework and moving to Ray Core?
- None: Just asking a question out of curiosity