[avinassh]

Isn't count slow on most databases? Either it ends up in table scan or index scan. But if the index is huge, then that will also take a lot of time.

I wonder how do Big Co solve this counting problem.