The NVIDIA Collective Communications Library (NCCL) implements multi-GPU and multi-node collective communication primitives that are performance optimized for NVIDIA GPUs. NCCL provides routines such as all-gather, all-reduce, broadcast, reduce, reduce-scatter, that are optimized to achieve high bandwidth over PCIe and NVLink high-speed interconnect.
| Uploaded | Mon Mar 31 22:50:33 2025 |
| md5 checksum | 347abf1c11f9946fb4ac307842986534 |
| arch | x86_64 |
| build | ha515578_0 |
| depends | cuda-version >=12.4,<13, libgcc-ng >=11.2.0, libstdcxx-ng >=11.2.0 |
| license | BSD-3-Clause |
| license_family | BSD |
| md5 | 347abf1c11f9946fb4ac307842986534 |
| name | nccl |
| platform | linux |
| sha1 | bef3a0e905cdd5e6521db0771d4e35c3da81d736 |
| sha256 | fddf240fcfa8a4a59007ca595155d34fd662099224813e95d1769d150e582341 |
| size | 132270887 |
| subdir | linux-64 |
| timestamp | 1716222100212 |
| version | 2.21.5.1 |