Re: [bloom] Benchmarks with Knuth multiplier-based hash production

From: Ivan Matek (libbooze_at_[hidden])
Date: 2025-06-07 08:23:19

On Sat, Jun 7, 2025 at 9:40 AM Joaquin M López Muñoz via Boost <
boost_at_[hidden]> wrote:

>
> This is the constant suggested by Kostas. A search on the Internet
> suggests this is used in connection with so-called PCG algorithms.
>
> I did a bit of digging, could not find anything interesting in a sense it
proves why it is "good".
Knuth provides no info, beside crediting person who gave him the constant.
This 2010 paper is a bit better, but still not great:
https://icl.utk.edu/users/luszczek/pubs/rarngicl.pdf

*By choosing a = 6364136223846793005 (=3×5×415949×1020018675983), c = 1,
and m = 2^64 we may obtain a sequence with a period 2^64.The same RNG is
used in the current implementation of the High Performance LINPACK (HPL)
benchmark*