I wonder what the ratio of repositories to physical machines is these days. I'd also be curious to see this as change over time. I have a hard time with the pr

I wonder what the ratio of repositories to physical machines is these days. I'd also be curious to see this as change over time. I have a hard time with the premise that a mere doubling of git ops would be especially crippling for any particular repository. GitHub runs like ass because it's oversubscribed by a huge factor. Not because git is inherently constraining at scale.