https://github.com/openucx/ucc
Unified Collective Communication Library
https://github.com/openucx/ucc
collectives cuda deep-learning hpc infiniband mpi openshmem pgas pytorch roce sharp
Last synced: 9 months ago
JSON representation
Unified Collective Communication Library
- Host: GitHub
- URL: https://github.com/openucx/ucc
- Owner: openucx
- License: bsd-3-clause
- Created: 2020-07-25T17:18:51.000Z (over 5 years ago)
- Default Branch: master
- Last Pushed: 2025-05-06T08:46:19.000Z (9 months ago)
- Last Synced: 2025-05-06T09:47:47.639Z (9 months ago)
- Topics: collectives, cuda, deep-learning, hpc, infiniband, mpi, openshmem, pgas, pytorch, roce, sharp
- Language: C
- Homepage: https://openucx.github.io/ucc/
- Size: 17.4 MB
- Stars: 251
- Watchers: 24
- Forks: 109
- Open Issues: 79
-
Metadata Files:
- Readme: README.md
- Changelog: NEWS
- Contributing: CONTRIBUTING.md
- License: LICENSE
- Authors: AUTHORS
Awesome Lists containing this project
README
# Unified Collective Communication (UCC)
UCC is a collective communication operations API and library that is flexible, complete, and feature-rich for current and emerging programming models and runtimes.
- [Design Goals](#design-goals)
- [API](https://openucx.github.io/ucc/)
- [Building](#compiling-and-installing)
- [Community](#community)
- [Contributing](#contributing)
- [License](#license)
- [Publication](#publication)
## Design Goals
* Highly scalable and performant collectives for HPC, AI/ML and I/O workloads
* Nonblocking collective operations that cover a variety of programming models
* Flexible resource allocation model
* Support for relaxed ordering model
* Flexible synchronous model
* Repetitive collective operations (init once and invoke multiple times)
* Hardware collectives are a first-class citizen
### UCC Component Architecture
## Contributing
Thanks for your interest in contributing to UCC, please see our technical and
legal guidelines in the [contributing](CONTRIBUTING.md) file.
All contributors have to comply with ["Membership Voluntary
Consensus Standard"](https://ucfconsortium.org/policy/) and ["Export Compliant
Contribution Submissions"](https://ucfconsortium.org/policy/) policies.
## License
UCC is BSD-style licensed, as found in the [LICENSE](LICENSE) file.
## Required packages
* [UCX](https://github.com/openucx/ucx)
* UCC uses utilities provided by UCX's UCS component
* [CUDA](https://developer.nvidia.com/cuda-toolkit) (optional)
* UCC supports CUDA collectives. To compile with CUDA support, install
[NVIDIA CUDA](https://developer.nvidia.com/cuda-downloads) 11.0 or above.
* [HIP](https://rocmdocs.amd.com/en/latest/Programming_Guides/HIP-GUIDE.html) (optional)
* UCC supports AMD GPUs using HIP. Instructions for installing ROCM/HIP can be found at
[AMD ROCM](https://rocmdocs.amd.com/en/latest/Installation_Guide/Installation_new.html).
* Doxygen
* UCC uses Doxygen for generating API documentation
## Compiling and Installing
### Developer's Build
```sh
$ ./autogen.sh
$ ./configure --prefix= --with-ucx=
$ make
```
### Build Documentation
```sh
$ ./autogen.sh
$ ./configure --prefix= --with-docs-only
$ make docs
```
### Open MPI and UCC collectives
#### Compile UCX
```sh
$ git clone https://github.com/openucx/ucx
$ cd ucx
$ ./autogen.sh; ./configure --prefix=; make -j install
```
#### Compile UCC
```sh
$ git clone https://github.com/openucx/ucc
$ cd ucc
$ ./autogen.sh; ./configure --prefix= --with-ucx=; make -j install
```
#### Compile Open MPI
```sh
$ git clone https://github.com/open-mpi/ompi
$ cd ompi
$ ./autogen.pl; ./configure --prefix= --with-ucx= --with-ucc=; make -j install
```
#### Run MPI programs
```sh
$ mpirun -np 2 --mca coll_ucc_enable 1 --mca coll_ucc_priority 100 ./my_mpi_app
```
#### Run OpenSHMEM programs
```sh
$ mpirun -np 2 --mca scoll_ucc_enable 1 --mca scoll_ucc_priority 100 ./my_openshmem_app
```
### SUPPORTED Transports
* UCX/UCP
- InfiniBand, ROCE, Cray Gemini and Aries, Shared Memory
* SHARP
* CUDA
* NCCL
* RCCL
* MLX5
### Publication
To cite UCC in a publication, please use the following BibTex entry:
```
@inproceedings{DBLP:conf/hoti/VenkataPLBALBDS24,
author = {Manjunath Gorentla Venkata and
Valentine Petrov and
Sergey Lebedev and
Devendar Bureddy and
Ferrol Aderholdt and
Joshua Ladd and
Gil Bloch and
Mike Dubman and
Gilad Shainer},
title = {Unified Collective Communication {(UCC):} An Unified Library for CPU,
GPU, and {DPU} Collectives},
booktitle = {{IEEE} Symposium on High-Performance Interconnects, {HOTI} 2024, Albuquerque,
NM, USA, August 21-23, 2024},
pages = {37--46},
publisher = {{IEEE}},
year = {2024},
url = {https://doi.org/10.1109/HOTI63208.2024.00018},
doi = {10.1109/HOTI63208.2024.00018},
timestamp = {Thu, 19 Sep 2024 11:00:54 +0200},
biburl = {https://dblp.org/rec/conf/hoti/VenkataPLBALBDS24.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
```