中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 计算技术研究所
  3. 中国科学院计算技术研究所
  4. 中国科学院计算技术研究所期刊论文
  5. 英文
DNA: A General Dynamic Neural Network Accelerator

文献类型：期刊论文

作者Liu, Lian1,2; Yu, Jinxin1,2; Wang, Mengdi1,2; Li, Xiaowei1,2; Han, Yinhe1,2; Wang, Ying1,2
刊名IEEE TRANSACTIONS ON COMPUTERS
出版日期2025-09-01
卷号74期号:9页码:3210-3222
关键词Dynamic scheduling Artificial neural networks DNA Processor scheduling Loading Prefetching Runtime Costs Switches Optimization Dynamic NN NPU design accelerator
ISSN号0018-9340
DOI10.1109/TC.2025.3587617
英文摘要Due to the demonstrated superiority, dynamic neural networks (NNs), which adapt their network structures to different inputs, have been recognized as an optimized alternative to conventional static NNs. However, researchers have not explored the implications of dynamic NN on neural processing unit (NPU) architecture design. Consequently, we analyze the characteristics and inefficient sources of executing dynamic NNs on existing hardware. From our analysis, existing NPUs, designed for static NNs, cannot effectively handle the execution of dynamic operator and agent-dependent data loading in dynamic NNs. To this end, we present DNA, an efficient accelerator optimized to deal with the challenges of running general dynamic NNs. Firstly, to improve the execution efficiency of dynamic operators, we propose a transverter-based online scheduling strategy to rapidly generate efficient scheduling for each dynamic operator. Secondly, to mitigate hardware idleness caused by the non-deterministic and agent-dependent data access patterns in dynamic NNs, we propose a novel predictor-based prefetching strategy that achieves effective data preloading with negligible cost. We implemented our accelerator, DNA, by integrating an additional online scheduler into a typical many-core baseline accelerator. According to our evaluation of various dynamic NNs, DNA achieves 3.48x speedup and 3.03 x energy savings over the baseline accelerator.
资助项目National Natural Science Foundation of China[62025404] ; National Natural Science Foundation of China[62222411] ; National Key R&D Program of China[2023YFB4404400]
WOS研究方向Computer Science ; Engineering
语种英语
WOS记录号WOS:001561348800003
出版者IEEE COMPUTER SOC
源URL[http://119.78.100.204/handle/2XEOYT63/41710]  
专题中国科学院计算技术研究所期刊论文_英文
通讯作者Wang, Ying
作者单位1.Chinese Acad Sci, Inst Comp Technol, Beijing 100190, Peoples R China
2.Univ Chinese Acad Sci, Dept Comp Sci, Beijing 100190, Peoples R China
推荐引用方式
GB/T 7714		 Liu, Lian,Yu, Jinxin,Wang, Mengdi,et al. DNA: A General Dynamic Neural Network Accelerator[J]. IEEE TRANSACTIONS ON COMPUTERS,2025,74(9):3210-3222.
APA Liu, Lian,Yu, Jinxin,Wang, Mengdi,Li, Xiaowei,Han, Yinhe,&Wang, Ying.(2025).DNA: A General Dynamic Neural Network Accelerator.IEEE TRANSACTIONS ON COMPUTERS,74(9),3210-3222.
MLA Liu, Lian,et al."DNA: A General Dynamic Neural Network Accelerator".IEEE TRANSACTIONS ON COMPUTERS 74.9(2025):3210-3222.

入库方式： OAI收割

来源：计算技术研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。