gloo-cuda 0.0~git20230519.597accf-2build1 source package in Ubuntu
Changelog
gloo-cuda (0.0~git20230519.597accf-2build1) noble; urgency=medium * No-change rebuild against libopenmpi3t64 -- Steve Langasek <email address hidden> Wed, 13 Mar 2024 18:25:52 +0000
Upload details
- Uploaded by:
- Steve Langasek
- Uploaded to:
- Noble
- Original maintainer:
- Ubuntu Developers
- Architectures:
- any
- Section:
- misc
- Urgency:
- Medium Urgency
See full publishing history Publishing
| Series | Published | Component | Section |
|---|
Downloads
| File | Size | SHA-256 Checksum |
|---|---|---|
| gloo-cuda_0.0~git20230519.597accf.orig.tar.xz | 180.8 KiB | 68f7bb91c706808d653cb4c9f81537f7c8a732544f76f992bf4e8d7be29c803a |
| gloo-cuda_0.0~git20230519.597accf-2build1.debian.tar.xz | 5.3 KiB | 1dea0f896b860bdf65f6b73f0559cb1b268d59756a215ab270690bae6f32241d |
| gloo-cuda_0.0~git20230519.597accf-2build1.dsc | 2.4 KiB | 034e262e9732aa8015ea941d1def020a9d8bd2a14126624018ed5f284fc55318 |
Available diffs
Binary packages built by this source
- libgloo-cuda-0: Collective communications library (shared object)
Gloo is a collective communications library. It comes with a number of
collective algorithms useful for machine learning applications. These
include a barrier, broadcast, and allreduce.
.
Transport of data between participating machines is abstracted so that
IP can be used at all times, or InifiniBand (or RoCE) when available.
In the latter case, if the InfiniBand transport is used, GPUDirect can
be used to accelerate cross machine GPU-to-GPU memory transfers.
.
Where applicable, algorithms have an implementation that works with system
memory buffers, and one that works with NVIDIA GPU memory buffers. In the
latter case, it is not necessary to copy memory between host and device;
this is taken care of by the algorithm implementations.
.
This package ships the shared object for Gloo.
- libgloo-cuda-0-dbgsym: debug symbols for libgloo-cuda-0
- libgloo-cuda-dev: Collective communications library (development files)
Gloo is a collective communications library. It comes with a number of
collective algorithms useful for machine learning applications. These
include a barrier, broadcast, and allreduce.
.
Transport of data between participating machines is abstracted so that
IP can be used at all times, or InifiniBand (or RoCE) when available.
In the latter case, if the InfiniBand transport is used, GPUDirect can
be used to accelerate cross machine GPU-to-GPU memory transfers.
.
Where applicable, algorithms have an implementation that works with system
memory buffers, and one that works with NVIDIA GPU memory buffers. In the
latter case, it is not necessary to copy memory between host and device;
this is taken care of by the algorithm implementations.
.
This package ships the development files.
