AutoSys: The Design and Operation of Learning-Augmented Systems


Chieh-Jan Mike Liang, Hui Xue, Mao Yang, and Lidong Zhou, Microsoft Research; Lifei Zhu, Peking University and Microsoft Research; Zhao Lucis Li and Zibo Wang, University of Science and Technology of China and Microsoft Research; Qi Chen and Quanlu Zhang, Microsoft Research; Chuanjie Liu, Microsoft Bing Platform; Wenjun Dai, Microsoft Bing Ads


Although machine learning (ML) and deep learning (DL) provide new possibilities into optimizing system design and performance, taking advantage of this paradigm shift requires more than implementing existing ML/DL algorithms. This paper reports our years of experience in designing and operating several production learning-augmented systems at Microsoft. AutoSys is a framework that unifies the development process, and it addresses common design considerations including ad-hoc and nondeterministic jobs, learning-induced system failures, and programming extensibility. Furthermore, this paper demonstrates the benefits of adopting AutoSys with measurements from one production system, Web Search. Finally, we share long-term lessons stemmed from unforeseen implications that have surfaced over the years of operating learning-augmented systems.

Open Access Media

USENIX is committed to Open Access to the research presented at our events. Papers and proceedings are freely available to everyone once the event begins. Any video, audio, and/or slides that are posted after the event are also free and open to everyone. Support USENIX and our commitment to Open Access.

@inproceedings {254402,
author = {Chieh-Jan Mike Liang and Hui Xue and Mao Yang and Lidong Zhou and Lifei Zhu and Zhao Lucis Li and Zibo Wang and Qi Chen and Quanlu Zhang and Chuanjie Liu and Wenjun Dai},
title = {{AutoSys}: The Design and Operation of {Learning-Augmented} Systems},
booktitle = {2020 USENIX Annual Technical Conference (USENIX ATC 20)},
year = {2020},
isbn = {978-1-939133-14-4},
pages = {323--336},
url = {},
publisher = {USENIX Association},
month = jul

Presentation Video