[chris]

Cool. I'd be interested in seeing the results of the same test using 1 million random myspace id's rather than sequential id's from 100-101M.

[sanj]

Does this does introduce a serious sampling bias? Is 100-101M in the middle of the id range? Or at the head or tail?

[corgan1003]

I did not just look at 100-101M. I looked at 100-110M, ~200k from each interval of one million. I concurrently ran 10 threads looking at intervals of one million. I did not include this in the code because I wanted to keep it as short as possible for the blog entry.