hashtable-benchmarks
hashtable-benchmarks copied to clipboard
bytell_hash_set's rehash is not expected load factor. My flat emhash can set load factor to 1.0 I have update dynamic_rehash_effect.cc. ``` static void high_load() { size_t maxSize = 1U
The `.bazelrc` file prevents building on ARM platforms (i.e. AWS Graviton) because: https://github.com/google/hashtable-benchmarks/blob/1e2f25fd81d4a15c582ff3a171d0756ff5d98d5f/.bazelrc#L3 It works after removing this line, but it would be nice to build on ARM by default...
Possibly consider adding https://github.com/greg7mdp/parallel-hashmap wdyt?
Currently the benchmark assumes that the rehash point of a given implementation is only determined by the size(), so all of the sets constructed by `GenerateSets` will have the same...
The hot lookup benchmarks repeatedly look for the same key. This allows the processor's branch predictor to learn the correct pattern of branches even for hash table algorithms that have...