Hello Ildar,
> Actually the "bad" one appears in YCSB.
Fine. Then it must be kept, whatever its quality.
> But if we should choose the only one I would stick to murmur too given
> it provides better results while having similar computational
> complexity.
No. Keep both as there is a justification for the bad one. Just make
"hash()" default to a good one.
>> One implementation put constants in defines, the other one uses "const
>> int". [...]
> [...] it looked ugly and hard to read (IMHO), like:
>
> k *= MURMUR2_M;
> k ^= k >> MURMUR2_R;
> k *= MURMUR2_M;
> result ^= k;
> result *= MURMUR2_M;
Yep. The ugliness is significantly linked to the choice of name. With
MM2_MUL and MM2_ROT ISTM that it is more readable:
> k *= MM2_MUL;
> k ^= k >> MM2_ROT;
> k *= MM2_MUL;
> result ^= k;
> result *= MM2_MUL;
> [...] So I'd better leave it the way it is. Actually I was thinking to
> do the same to fnv1a too : )
I think that the implementation style should be homogeneous, so I'd
suggest at least to stick to one style.
I noticed from the source of all human knowledege (aka Wikipedia:-) that
there seems to be a murmur3 successor. Have you considered it? One good
reason to skip it would be that the implementation is long and complex.
I'm not sure about a 8-byte input simplified version.
Just a question: Have you looked at SipHash24?
https://en.wikipedia.org/wiki/SipHash
The interesting point is that it can use a key and seems somehow
cryptographically secure, for a similar cost. However the how to decide
for/control the key is unclear.
--
Fabien.
