parallel-hashmap
parallel-hashmap copied to clipboard
greg7mdp
A family of header-only, very fast and memory-friendly hashmap and btree containers.
Could you provide a way to not download gtest and use the system one?
I find that flat_hash_map will sometimes cause a seg fault when the memory limit of the process is reached and malloc() returns null. The seg fault is on the free()...
It seems custom pointer types are not supported with the parallel hashmap: https://github.com/greg7mdp/parallel-hashmap/blob/master/parallel_hashmap/phmap.h#L843-L846 Would this be difficult to add ? Because I would like to apply the parallel hashmap to...
Thanks for your help earlier today. We have been using your `flat_hash_map` in Merlin HugeCTR for quite some time now. I have implemented AVX2-accelerated groups for parallel-hashmap, which I hope...
Hello, I am currently using your flat_hash_map in an LSH Forest implementation. My main limitation is that the dataset I want to use is quite large (86 million vectors of...
The LLDB pretty-printer sometimes prints junk values. Consider following example: ```cpp #include #include struct ivec2 { int x = 0, y = 0; friend bool operator==(ivec2 a, ivec2 b) {...
Hi @greg7mdp You helped me a few months ago implementing parallel joining of maps: https://github.com/greg7mdp/parallel-hashmap/issues/234 This was working fine, but recently I started having trouble under Mac. As a matter...
phmap mix is a pretty good rehash function. In a multi-stage aggregation process I want to use different mix values. But I don't know how to generate a proper mix...
I want to include parallel-hashmap in a bazel built project, so add these files. @greg7mdp help to review it, thanks in advance.
