Re: CPU costs of random_zipfian in pgbench

> I'm trying to use random_zipfian() for benchmarking of skewed data sets,
> and I ran head-first into an issue with rather excessive CPU costs.
> [...] This happens because generalizedHarmonicNumber() does this:
>
>       for (i = n; i > 1; i--)
>               ans += pow(i, -s);
>
> where n happens to be 1000000000 (range passed to random_zipfian), so
> the loop takes quite a bit of time.

If you find a better formula for the harmonic number, you are welcome
and probably get your name on it:-)