31 lines
1.3 KiB
Plaintext
31 lines
1.3 KiB
Plaintext
Source: nccl
|
|
Section: universe/libs
|
|
Maintainer: Boris Fomitchev <bfomitchev@nvidia.com>
|
|
Priority: optional
|
|
Build-depends: debhelper(>=9), cuda-core-8-0, cuda-toolkit-8-0,
|
|
cuda-curand-dev-8-0, cuda-cublas-dev-8-0, cuda-cudart-dev-8-0
|
|
Standards-Version: 3.9.5
|
|
|
|
Package: libnccl1
|
|
Section: universe/libs
|
|
Architecture: amd64
|
|
Depends: ${misc:Depends}, ${shlibs:Depends},
|
|
cuda-curand-8-0, cuda-cublas-8-0, cuda-cudart-8-0
|
|
Description: NVIDIA Communication Collectives Library (NCCL) Runtime
|
|
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 up to eight
|
|
GPUs and can be used in either single- or multi-process (e.g., MPI)
|
|
applications.
|
|
|
|
Package: libnccl-dev
|
|
Section: universe/libdevel
|
|
Architecture: amd64
|
|
Depends: ${misc:Depends}, ${shlibs:Depends}, libnccl1 (= ${binary:Version})
|
|
Description: NVIDIA Communication Collectives Library (NCCL) Development Files
|
|
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 up to eight
|
|
GPUs and can be used in either single- or multi-process (e.g., MPI)
|
|
applications.
|