I thought this was common knowledge but I guess not: Google's monopoly famously has over a billion files. No, Git cannot handle it. Their whole software stack is developed around this from the ground up. But they are one of the largest software employers in the world, so quite a few engineers evidently do make do with 200x more than 50 million files.