@inproceedings{c979043e7c404aca85eae9f72f72ab92,
title = "Xenos: Dataflow-Centric Optimization to Accelerate Model Inference on Edge Devices",
abstract = "In this paper, we propose Xenos, a high-performance edge platform for model inference. Unlike the prior works which mainly focus one operator-centric optimization, Xenos can automatically conduct dataflow-centric optimization on the computation graph and accelerate inference in two dimensions. Vertically, Xenos develops operator linking technique to improve data locality by restructuring the inter-operator dataflow. Horizontally, Xenos develops DSP-aware operator split technique to enable higher parallelism across multiple DSP units. Our evaluation proves the effectiveness of Xenos {\textquoteright} vertical and horizontal dataflow optimization, which reduce the inference time by 15.0\%{\textendash}84.9\% and 17.9\%{\textendash}89.9\%, respectively.",
keywords = "DSP-aware, data locality, dataflow-centric, edge-based inference",
author = "Runhua Zhang and Hongxu Jiang and Fangzheng Tian and Jinkun Geng and Xiaobin Li and Yuhang Ma and Chenhui Zhu and Dong Dong and Xin Li and Haojie Wang",
note = "Publisher Copyright: {\textcopyright} 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.; 28th International Conference on Database Systems for Advanced Applications, DASFAA 2023 ; Conference date: 17-04-2023 Through 20-04-2023",
year = "2023",
doi = "10.1007/978-3-031-30637-2\_35",
language = "英语",
isbn = "9783031306365",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Science and Business Media Deutschland GmbH",
pages = "535--545",
editor = "Xin Wang and Sapino, \{Maria Luisa\} and Wook-Shin Han and \{El Abbadi\}, Amr and Gill Dobbie and Zhiyong Feng and Yingxiao Shao and Hongzhi Yin",
booktitle = "Database Systems for Advanced Applications - 28th International Conference, DASFAA 2023, Proceedings",
address = "德国",
}