Software:DeepSpeed
 | |
| Original author(s) | Microsoft Research |
|---|---|
| Developer(s) | Microsoft |
| Initial release | May 18, 2020 |
| Stable release | v0.5.10
/ January 14, 2022 |
| Repository | github |
| Written in | Python, CUDA, C++ |
| Type | Software library |
| License | MIT License |
| Website | deepspeed |
DeepSpeed is an open source deep learning optimization library for PyTorch.[1] The library is designed to reduce computing power and memory use and to train large distributed models with better parallelism on existing computer hardware.[2][3] DeepSpeed is optimized for low latency, high throughput training. It includes the Zero Redundancy Optimizer (ZeRO) for training models with 1 trillion or more parameters.[4] Features include mixed precision training, single-GPU, multi-GPU, and multi-node training as well as custom model parallelism. The DeepSpeed source code is licensed under MIT License and available on GitHub.[5]
The team claimed to achieve up to a 6.2x throughput improvement, 2.8x faster convergence, and 4.6x less communication.[6]
See also
References
- ↑ "Microsoft Updates Windows, Azure Tools with an Eye on The Future". May 22, 2020. https://uk.pcmag.com/news-analysis/127085/microsoft-updates-windows-azure-tools-with-an-eye-on-the-future.
- ↑ Yegulalp, Serdar (February 10, 2020). "Microsoft speeds up PyTorch with DeepSpeed". https://www.infoworld.com/article/3526449/microsoft-speeds-up-pytorch-with-deepspeed.html.
- ↑ "Microsoft unveils "fifth most powerful" supercomputer in the world". https://www.neowin.net/news/microsoft-unveils-fifth-most-powerful-supercomputer-in-the-world/.
- ↑ "Microsoft trains world's largest Transformer language model". February 10, 2020. https://venturebeat.com/2020/02/10/microsoft-trains-worlds-largest-transformer-language-model/.
- ↑ "microsoft/DeepSpeed". July 10, 2020. https://github.com/microsoft/DeepSpeed.
- ↑ "DeepSpeed: Accelerating large-scale model inference and training via system optimizations and compression" (in en-US). 2021-05-24. https://www.microsoft.com/en-us/research/blog/deepspeed-accelerating-large-scale-model-inference-and-training-via-system-optimizations-and-compression/.
Further reading
- Rajbhandari, Samyam; Rasley, Jeff; Ruwase, Olatunji; He, Yuxiong (2019). ZeRO: Memory Optimization Towards Training A Trillion Parameter Models. https://arxiv.org/pdf/1910.02054.pdf.
External links
- AI at Scale - Microsoft Research
- GitHub - microsoft/DeepSpeed
- ZeRO & DeepSpeed: New system optimizations enable training models with over 100 billion parameters - Microsoft Research
 |  |
