Mathematical Framework for Optimizing Crossbar Allocation for ReRAM-based CNN Accelerators
文献类型：期刊论文
| 作者 | Li, Wanqian1,2; Han, Yinhe2; Chen, Xiaoming2 |
| 刊名 | ACM TRANSACTIONS ON DESIGN AUTOMATION OF ELECTRONIC SYSTEMS
 |
| 出版日期 | 2024 |
| 卷号 | 29期号:1页码:24 |
| 关键词 | ReRAM crossbars CNN accelerator mathematical framework crossbar allocation |
| ISSN号 | 1084-4309 |
| DOI | 10.1145/3631523 |
| 英文摘要 | The resistive random-access memory (ReRAM) has widely been used to accelerate convolutional neural networks (CNNs) thanks to its analog in-memory computing capability. ReRAM crossbars not only store layers' weights, but also perform in-situ matrix-vector multiplications which are core operations of CNNs. To boost the performance of ReRAM-based CNN accelerators, crossbars can be duplicated to explore more intra-layer parallelism. The crossbar allocation scheme can significantly influence both the computing throughput and bandwidth requirements of ReRAM-based CNN accelerators. Under the resource constraints (i.e., crossbars and memory bandwidths), how to find the optimal number of crossbars for each layer to maximize the inference performance for an entire CNN is an unsolved problem. In this work, we find the optimal crossbar allocation scheme by mathematically modeling the problem as a constrained optimization problem and solving it with a dynamic programming based solver. Experiments demonstrate that ourmodel for CNN inference time is almost precise, and the proposed framework can obtain solutions with near-optimal inference time. We also emphasize that communication (i.e., data access) is an important factor and must also be considered when determining the optimal crossbar allocation scheme. |
| 资助项目 | National Key R&D Program of China[2018YFA0701500] ; National Natural Science Foundation of China[62122076] ; National Natural Science Foundation of China[61834006] ; National Natural Science Foundation of China[62025404] ; Key Research Program of Frontier Sciences, CAS[ZDBS-LY-JSC012] ; Strategic Priority Research Program of CAS[XDB44000000] ; Innovation Funding of ICT, CAS[E261040] |
| WOS研究方向 | Computer Science |
| 语种 | 英语 |
| WOS记录号 | WOS:001168397800021 |
| 出版者 | ASSOC COMPUTING MACHINERY |
| 源URL | [http://119.78.100.204/handle/2XEOYT63/38804]  |
| 专题 | 中国科学院计算技术研究所期刊论文_英文 |
| 通讯作者 | Chen, Xiaoming |
| 作者单位 | 1.Univ Chinese Acad Sci, 6 Kexueyuan S Rd, Beijing 100190, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, 6 Kexueyuan S Rd, Beijing 100190, Peoples R China |
| 推荐引用方式 GB/T 7714 | Li, Wanqian,Han, Yinhe,Chen, Xiaoming. Mathematical Framework for Optimizing Crossbar Allocation for ReRAM-based CNN Accelerators[J]. ACM TRANSACTIONS ON DESIGN AUTOMATION OF ELECTRONIC SYSTEMS,2024,29(1):24. |
| APA | Li, Wanqian,Han, Yinhe,&Chen, Xiaoming.(2024).Mathematical Framework for Optimizing Crossbar Allocation for ReRAM-based CNN Accelerators.ACM TRANSACTIONS ON DESIGN AUTOMATION OF ELECTRONIC SYSTEMS,29(1),24. |
| MLA | Li, Wanqian,et al."Mathematical Framework for Optimizing Crossbar Allocation for ReRAM-based CNN Accelerators".ACM TRANSACTIONS ON DESIGN AUTOMATION OF ELECTRONIC SYSTEMS 29.1(2024):24. |
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。