[FEA]: [CUCO] Improve cudax HLL
Open
cuco
cudax
- Dominant language
- C++
- Stars
- 2.5k
- Forks
- 486
- Avg merge
- 2d 6h
- Merged PRs (30d)
- 295
Description
#6666 Introduces H.L.L, migrated from cuCollections.
However, there are many more minor issues needs to covered.
This Issue tracks the remaining issues
- [ ] Add extensive testing to test all the APIs for `hyperloglog_impl`
- [ ] Clean up testing file/director structure for cuco in cudax
- [ ] Add benchmarks
- [ ] Add examples
- [ ] Optimize bias and raw estimate global array's to use `__constant__` or `__grid_constant__` for device compilations
- [ ] Rebase with latest update's from cuCollections (https://github.com/NVIDIA/cuCollections/pull/792)
Contributor guide
Assessment
This issue has not been assessed yet.