4 ms·
Looks good, but has somewhat widely false claims: > SMHasher is a test and benchmark suite for a variety of hash function implementations. It can provide data
by rurban 6y ago
Looks good, but has somewhat widely false claims:
> SMHasher is a test and benchmark suite for a variety of hash function implementations. It can provide data for claims like “our new Foo hash function is faster than the widely used Bar, Baz and Qux hash functions”. However, when comparing Foo to CRC-32, be aware that a SIMD-accelerated CRC-32 implementation can be 47x faster than SMHasher’s simple CRC-32 implementation.
In fact smhasher implements both, the slow soft crc32 he cites, and the fastest crc32_pclmul, which is not just 47x faster, but 5000x faster. Other than wuff's implementation of the crc32_pclmul variant.
Compare https://github.com/rurban/smhasher/#smhasher https://github.com/rurban/smhasher/#smhasher
crc32 392.10 vs crc32_pclmul 1972140.38 MiB/sec
Reminds me a lot on ATS
- Karliss 6y agoThat number doesn't seem realistic. Even for future generations of ram and PCIe the bandwidth is on the order of 50GB/s not 2TB/s .
- nigeltao 6y ago> smhasher implements both Ah, I'm based on aappleby/smasher, which is what the original location (http://code.google.com/p/smhasher http://code.google.com/p/smhasher) redirects to. I didn't realize that you have a more active fork. > not just 47x faster, but 5000x faster It's 47x faster: https://github.com/rurban/smhasher/issues/189 https://github.com/rurban/smhasher/issues/189
- rurban 6y agoThanks a lot for detecting this bug! Nigel's numbers were right, mine were wrong