Collage: Automated Integration of Deep Learning Backends
Machine Learning & Vision Laboratory, Gwangju Institute of Science and Technology (GIST), Gwangju 61005, South Korea
arXiv:2111.00655 [cs.LG], (1 Nov 2021)
@misc{jeon2021collage,
title={Collage: Automated Integration of Deep Learning Backends},
author={Byungsoo Jeon and Sunghyun Park and Peiyuan Liao and Sheng Xu and Tianqi Chen and Zhihao Jia},
year={2021},
eprint={2111.00655},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
Strong demands for efficient deployment of Deep Learning (DL) applications prompt the rapid development of a rich DL ecosystem. To keep up with its fast advancement, it is crucial for DL frameworks to efficiently integrate a variety of optimized libraries and runtimes as their backends and generate the fastest possible executable by using them properly. However, current DL frameworks require significant manual effort to integrate diverse backends and often fail to deliver high performance. In this paper, we propose Collage, an automatic framework for integrating DL backends. Collage provides a backend registration interface that allows users to precisely specify the capability of various backends. By leveraging the specifications of available backends, Collage searches for an optimized backend placement for a given workload and execution environment. Our evaluation shows that Collage automatically integrates multiple backends together without manual intervention, and outperforms existing frameworks by 1.21x, 1.39x, 1.40x on two different NVIDIA GPUs and an Intel CPU respectively.
November 7, 2021 by hgpu
