NCCL (pronounced "Nickel") is a stand-alone library of standard collective communication routines, such as all-gather, reduce, broadcast, etc., that have been optimized to achieve high bandwidth over PCIe. NCCL supports an arbitrary number of GPUs installed in a single node and can be used in either single- or multi-process (e.g., MPI) applications.
| Uploaded | Mon Mar 31 22:50:41 2025 |
| md5 checksum | ba67a122cf0a487f63e3a7f52a9fba35 |
| arch | x86_64 |
| build | cuda9.0_0 |
| depends | cudatoolkit 9.0.*, libgcc-ng >=5.4.0, libstdcxx-ng >=5.4.0 |
| license | BSD 3-Clause |
| license_family | BSD |
| md5 | ba67a122cf0a487f63e3a7f52a9fba35 |
| name | nccl |
| platform | linux |
| sha1 | 93d24e6e865b7f493e75319b02cc6295b7beb56b |
| sha256 | e44d28d32d1948ac12e5afa4f4a52422deb45ed5bca71511dc483efb5affd233 |
| size | 1409460 |
| subdir | linux-64 |
| timestamp | 1523280871183 |
| version | 1.3.5 |