Josh Berkus <josh@agliodbs.com> writes:
>> So what's going on with the empirically derived value of 4?
> It's not empirically derived;
Yes it is. I ran experiments back in the late 90s to derive it.
Check the archives.
Disks have gotten noticeably bigger since then, but I don't think
the ratio of seek time to rotation rate has changed much.
regards, tom lane