Optimus: An Operator Fusion Framework for Deep Neural Networks
文献类型:期刊论文
作者 | Cai, Xuyi3,4; Wang, Ying1,2; Zhang, Lei4 |
刊名 | ACM TRANSACTIONS ON EMBEDDED COMPUTING SYSTEMS |
出版日期 | 2023 |
卷号 | 22期号:1页码:26 |
ISSN号 | 1539-9087 |
关键词 | Neural network embedded processor memory layer fusion |
DOI | 10.1145/3520142 |
英文摘要 | The reduction of neural parameters and operations for the applications on embedded and IoT platforms in current deep neural network (DNN) architectures has received increasing attention. Relatively, the intermediate feature maps of such lightweight neural networks begin to grow and usually outsize the on-chip memory as the new bottleneck, which introduces considerable power-consuming off-chip memory accesses. To reduce the feature-induced memory accesses, operator fusion has been proposed to parallelize the execution of multiple convolutional layers and shown significant reduction of off-chip memory accesses. However, how to fuse the neural operators is still a challenging issue that heavily depends on both the neural network (NN) topology and the specific DNN accelerator configuration. In this work, we observed prior operator fusion approaches fail to guarantee memory-level optimality as they search in the constrained operator fusion design space. Considering the complexity of the NN topologies and the constrained resources of the DNN accelerators, we develop a novel operator fusion framework, Optimus. Optimus includes an accurate memory cost model dedicated to the scheduler to evaluate the potential operator-fusion schemes and a directed acyclic graph-based operator fusion algorithm for both off-line and on-line workload deployment scenarios, which altogether generates high-efficiency operator-fusion solutions for arbitrary networkmodels running on DNN accelerators. The experimental results show that Optimus reduces 17-75% off-chip memory accesses and obtains 1.86x-3.66x energy efficiency on state-of-the-art DNN workloads when compared to the baselines and brings significant power-efficiency boost to the DNN accelerators of different architectures and dataflows. |
WOS研究方向 | Computer Science |
语种 | 英语 |
出版者 | ASSOC COMPUTING MACHINERY |
WOS记录号 | WOS:000908419900001 |
源URL | [http://119.78.100.204/handle/2XEOYT63/20029] |
专题 | 中国科学院计算技术研究所期刊论文 |
通讯作者 | Wang, Ying |
作者单位 | 1.Chinese Acad Sci, Inst Comp Technol, State Key Lab Comp Architecture, Beijing, Peoples R China 2.Chinese Acad Sci, Zhejiang Lab, Beijing, Peoples R China 3.Univ Chinese Acad Sci, Beijing, Peoples R China 4.Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China |
推荐引用方式 GB/T 7714 | Cai, Xuyi,Wang, Ying,Zhang, Lei. Optimus: An Operator Fusion Framework for Deep Neural Networks[J]. ACM TRANSACTIONS ON EMBEDDED COMPUTING SYSTEMS,2023,22(1):26. |
APA | Cai, Xuyi,Wang, Ying,&Zhang, Lei.(2023).Optimus: An Operator Fusion Framework for Deep Neural Networks.ACM TRANSACTIONS ON EMBEDDED COMPUTING SYSTEMS,22(1),26. |
MLA | Cai, Xuyi,et al."Optimus: An Operator Fusion Framework for Deep Neural Networks".ACM TRANSACTIONS ON EMBEDDED COMPUTING SYSTEMS 22.1(2023):26. |
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。