Deep Learning Inference Service at Microsoft


Jonathan Soifer, Jason Li, Mingqin Li, Jeffrey Zhu, Yingnan Li, Yuxiong He, Elton Zheng, Adi Oltean, Maya Mosyak, Chris Barnes, Thomas Liu, and Junhua Wang, Microsoft


This paper introduces the Deep Learning Inference Service, an online production service at Microsoft for ultra-low-latency deep neural network model inference. We present the system architecture and deep dive into core concepts such as intelligent model placement, heterogeneous resource management, resource isolation, and efficient routing. We also present production scale and performance numbers.

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 {232993,
author = {Jonathan Soifer and Jason Li and Mingqin Li and Jeffrey Zhu and Yingnan Li and Yuxiong He and Elton Zheng and Adi Oltean and Maya Mosyak and Chris Barnes and Thomas Liu and Junhua Wang},
title = {Deep Learning Inference Service at Microsoft},
booktitle = {2019 USENIX Conference on Operational Machine Learning (OpML 19)},
year = {2019},
isbn = {978-1-939133-00-7},
address = {Santa Clara, CA},
pages = {15--17},
url = {},
publisher = {USENIX Association},
month = may