Source: nvidia-nccl Section: contrib/libs Priority: optional Maintainer: Debian NVIDIA Maintainers Uploaders: Mo Zhou Standards-Version: 4.6.2 Build-Depends: debhelper-compat (= 13), nvidia-cuda-toolkit-gcc Homepage: https://github.com/NVIDIA/nccl Vcs-Browser: https://salsa.debian.org/nvidia-team/nvidia-nccl Vcs-Git: https://salsa.debian.org/nvidia-team/nvidia-nccl.git Rules-Requires-Root: no Package: libnccl2 Architecture: amd64 arm64 ppc64el Depends: ${misc:Depends}, ${shlibs:Depends} Provides: libnccl.so.2 Description: NVIDIA Optimized primitives for inter-GPU communication NCCL (pronounced "Nickel") is a stand-alone library of standard communication routines for GPUs, implementing all-reduce, all-gather, reduce, broadcast, reduce-scatter, as well as any send/receive based communication pattern. It has been optimized to achieve high bandwidth on platforms using PCIe, NVLink, NVswitch, as well as networking using InfiniBand Verbs or TCP/IP sockets. NCCL supports an arbitrary number of GPUs installed in a single node or across multiple nodes, and can be used in either single- or multi-process (e.g., MPI) applications. . This package contains the shared objects. Package: libnccl-dev Section: contrib/libdevel Architecture: amd64 arm64 ppc64el Depends: libnccl2 (= ${binary:Version}), ${misc:Depends}, ${shlibs:Depends} Provides: libnccl.so Description: NVIDIA Optimized primitives for inter-GPU communication (development) NCCL (pronounced "Nickel") is a stand-alone library of standard communication routines for GPUs, implementing all-reduce, all-gather, reduce, broadcast, reduce-scatter, as well as any send/receive based communication pattern. It has been optimized to achieve high bandwidth on platforms using PCIe, NVLink, NVswitch, as well as networking using InfiniBand Verbs or TCP/IP sockets. NCCL supports an arbitrary number of GPUs installed in a single node or across multiple nodes, and can be used in either single- or multi-process (e.g., MPI) applications. . This package contains the development files.