Cross-architecture parallel algorithms for Julia's CPU and GPU backends. Targets multithreaded CPUs, and GPUs via Intel oneAPI, AMD ROCm, Apple Metal, Nvidia CUDA.
active 2024-11-11 → 2026-07-31 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 628 days from 2024-11-11 to 2026-07-31. Pushes: 139 total, peak 12 in a day. Pull requests: 55 total, peak 4 in a day. Issues: 29 total, peak 2 in a day. Comments: 159 total, peak 12 in a day. Stars: 71 total, peak 2 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| anicusan | 207 | 110 | 28 | 57 |
| christiangnrd | 47 | 13 | 10 | 16 |
| yolhan83 | 22 | 0 | 1 | 19 |
| vchuravy | 18 | 2 | 4 | 10 |
| JuliaRegistrator | 12 | 0 | 0 | 12 |
| maleadt | 12 | 9 | 0 | 2 |
| JuliaTagBot | 9 | 0 | 0 | 9 |
| pxl-th | 7 | 0 | 1 | 4 |
| giordano | 6 | 0 | 0 | 5 |
| albertomercurio | 5 | 0 | 0 | 4 |
| AntonReinhard | 4 | 0 | 0 | 2 |
| dependabot[bot] | 3 | 0 | 2 | 1 |
| 0samuraiE | 3 | 0 | 0 | 2 |
| Moelf | 3 | 0 | 0 | 2 |
| luraess | 3 | 0 | 0 | 2 |
| smillerc | 2 | 0 | 0 | 1 |
| matbesancon | 2 | 0 | 0 | 0 |
| VarLad | 2 | 0 | 0 | 2 |
| dstahlke | 2 | 0 | 1 | 1 |
| mtagliazucchi | 2 | 0 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Pull request#95christiangnrd2026-07-04 16:11
- Issue comment#79christiangnrd2026-06-13 18:58Switch to ParallelTestRunner.
- Issue comment#74AntonReinhard2026-04-07 21:52Accumulate crashes on CUDA with DecoupledLookback
- Issue#75AntonReinhard2026-04-05 12:48Accumulate becomes slow for very large input sizes
- Issue comment#75AntonReinhard2026-04-05 12:48Accumulate becomes slow for very large input sizes
- Pull request#78jariji2026-03-04 23:33
- Issue comment#54ExpandingMan2026-02-28 18:08Support `mapreduce` with multiple source arrays
- Issue#76matbesancon2025-12-20 13:13Define mapreduce with multiple iterators for operations over both values and indices
- Issue#76matbesancon2025-12-20 13:12Define mapreduce with multiple iterators for operations over both values and indices
- Issue#74AntonReinhard2025-12-17 15:24Accumulate crashes on CUDA with DecoupledLookback
- Issue comment#73michel23232025-12-16 16:31Remove oneAPI extension
- Pull request#72christiangnrd2025-12-13 20:10
- Issue comment#71christiangnrd2025-12-13 20:06Support KA 0.10
- Pull request#71christiangnrd2025-12-13 20:03
- Issue comment#63dependabot[bot]2025-11-24 22:26Bump actions/checkout from 4 to 5
- Pull request#70dependabot[bot]2025-11-24 22:26
- Pull request#70dependabot[bot]2025-11-24 22:26
- Issue comment#69VarLad2025-11-13 01:18`AcceleratedKernels.sum` is twice as slow than CUDA.jl's mapreduce implementation
- Issue comment#69VarLad2025-11-12 20:32`AcceleratedKernels.sum` is twice as slow than CUDA.jl's mapreduce implementation
- Issue comment#66albertomercurio2025-09-29 23:01reducing while iterating over the index
- Issue comment#66yolhan832025-09-28 17:34reducing while iterating over the index
- Issue comment#66albertomercurio2025-09-28 17:22reducing while iterating over the index
- Issue comment#66yolhan832025-09-28 14:42reducing while iterating over the index
- Issue comment#66albertomercurio2025-09-21 20:41reducing while iterating over the index
- Issue comment#66yolhan832025-09-21 15:31reducing while iterating over the index
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 71 stars here means stars gained during the window, not the repo's star count.