Re: Perfect hash function generator

[Joerg Sonnenberger <joerg%britannica.bec.de@localhost>]

On Wed, Jul 15, 2009 at 03:17:17PM +0200, Joerg Sonnenberger wrote:
> On AMD64, which is known to have a very fast multiplication:
> jenkins_vector_hash: 8.66s
> MurmurHash2_64: 5.71s

Actually, scratch that. Stupid compiler. After changing the jenkins hash
to compute in separate variables and not expect to compile to optimise
that, we have:

short strings (the set from terminfo):
jenkins_vector_hash: 3.16s
murmur2 64: 3.92

large strings (random typing until 200 Bytes are reached):
jenkins_vector_hash: 2.12
murmur2 64: 2.51

[difference frequency, the values scale though]

Joerg