|
| 1 | +--- |
| 2 | +json_ld: |
| 3 | + '@context': https://schema.org |
| 4 | + '@type': SoftwareApplication |
| 5 | + applicationCategory: DeveloperApplication |
| 6 | + description: "\n The Portable Hardware Locality (hwloc) software package provides\ |
| 7 | + \ a portable\n abstraction (across OS, versions, architectures, ...) of the hierarchical\n\ |
| 8 | + \ topology of modern architectures, including NUMA memory nodes, sockets, shared\n\ |
| 9 | + \ caches, cores and simultaneous multithreading. It also gathers various system\n\ |
| 10 | + \ attributes such as cache and memory information as well as the locality of I/O\n\ |
| 11 | + \ devices such as network interfaces, InfiniBand HCAs or GPUs. It primarily\n\ |
| 12 | + \ aims at helping applications with gathering information about modern computing\n\ |
| 13 | + \ hardware so as to exploit it accordingly and efficiently.\n\n This provides\ |
| 14 | + \ the CUDA and NVML plugins for hwloc\n" |
| 15 | + license: Not confirmed |
| 16 | + name: hwloc-CUDA |
| 17 | + offers: |
| 18 | + '@type': Offer |
| 19 | + price: 0 |
| 20 | + operatingSystem: LINUX |
| 21 | + review: |
| 22 | + '@type': Review |
| 23 | + author: |
| 24 | + '@type': Organization |
| 25 | + name: EESSI |
| 26 | + reviewBody: Application has been successfully made available on all architectures |
| 27 | + supported by EESSI |
| 28 | + reviewRating: |
| 29 | + '@type': Rating |
| 30 | + ratingValue: 5 |
| 31 | + softwareRequirements: See https://www.eessi.io/docs/ for how to make EESSI available |
| 32 | + on your system |
| 33 | + softwareVersion: '[''2.12.1'', ''2.11.2'']' |
| 34 | + url: https://www.open-mpi.org/projects/hwloc/ |
| 35 | +--- |
| 36 | +# hwloc-CUDA |
| 37 | + |
| 38 | + |
| 39 | + |
| 40 | + The Portable Hardware Locality (hwloc) software package provides a portable |
| 41 | + abstraction (across OS, versions, architectures, ...) of the hierarchical |
| 42 | + topology of modern architectures, including NUMA memory nodes, sockets, shared |
| 43 | + caches, cores and simultaneous multithreading. It also gathers various system |
| 44 | + attributes such as cache and memory information as well as the locality of I/O |
| 45 | + devices such as network interfaces, InfiniBand HCAs or GPUs. It primarily |
| 46 | + aims at helping applications with gathering information about modern computing |
| 47 | + hardware so as to exploit it accordingly and efficiently. |
| 48 | + |
| 49 | + This provides the CUDA and NVML plugins for hwloc |
| 50 | + |
| 51 | + |
| 52 | +<small>homepage: </small><span class="software-link">[https://www.open-mpi.org/projects/hwloc/](https://www.open-mpi.org/projects/hwloc/)</span> |
| 53 | + |
| 54 | +## Available installations |
| 55 | + |
| 56 | + |
| 57 | +|hwloc-CUDA version|Supported CPU targets|Supported GPU targets|EESSI version|Module| |
| 58 | +| --- | --- | --- | --- | --- | |
| 59 | +|2.12.1|`generic`: `aarch64`, `x86_64`<br/><span class="software-cpu-arm">Arm</span>: `neoverse_n1`, `neoverse_v1`, `nvidia/grace`<br/><span class="software-cpu-amd">AMD</span>: `zen2`, `zen3`, `zen4`, `zen5`<br/><span class="software-cpu-intel">Intel</span>: `haswell`, `skylake_avx512`, `sapphirerapids`, `icelake`, `cascadelake`<br/>|<span class="software-gpu-nvidia">NVIDIA</span>: `cc100`, `cc120`, `cc70`, `cc80`, `cc90`<br/>|<span class="software-eessi-version-202506">2025.06</span>|`hwloc-CUDA/2.12.1-GCCcore-14.3.0-CUDA-12.9.1`| |
| 60 | +|2.11.2|`generic`: `aarch64`, `x86_64`<br/><span class="software-cpu-arm">Arm</span>: `neoverse_n1`, `neoverse_v1`, `nvidia/grace`<br/><span class="software-cpu-amd">AMD</span>: `zen2`, `zen3`, `zen4`, `zen5`<br/><span class="software-cpu-intel">Intel</span>: `haswell`, `skylake_avx512`, `sapphirerapids`, `icelake`, `cascadelake`<br/>|<span class="software-gpu-nvidia">NVIDIA</span>: `cc100`, `cc120`, `cc70`, `cc80`, `cc90`<br/>|<span class="software-eessi-version-202506">2025.06</span>|`hwloc-CUDA/2.11.2-GCCcore-14.2.0-CUDA-12.8.0`| |
0 commit comments