A Fine-Grained Modeling Approach for Systolic Array-Based Accelerator
The systolic array provides extremely high efficiency for running matrix multiplication and is one of the mainstream architectures of today’s deep learning accelerators. In order to develop efficient accelerators, people usually employ simulators to make design trade-offs. However, current simulator...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-09-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/11/18/2928 |
| _version_ | 1797489069504069632 |
|---|---|
| author | Yuhang Li Mei Wen Jiawei Fei Junzhong Shen Yasong Cao |
| author_facet | Yuhang Li Mei Wen Jiawei Fei Junzhong Shen Yasong Cao |
| author_sort | Yuhang Li |
| collection | DOAJ |
| description | The systolic array provides extremely high efficiency for running matrix multiplication and is one of the mainstream architectures of today’s deep learning accelerators. In order to develop efficient accelerators, people usually employ simulators to make design trade-offs. However, current simulators suffer from coarse-grained modeling methods and ideal assumptions, which limits their ability to describe structural characteristics of systolic arrays. In addition, they do not support the exploration of microarchitecture. This paper presents FG-SIM, a fine-grained modeling approach for evaluating systolic array accelerators by using an event-driven method. FG-SIM can obtain accurate results and provide the best mapping scheme for different workloads due to its fine-grained modeling technique and deny of ideal assumption. Experimental results show that FG-SIM plays a significant role in design trade-offs and outperforms state-of-the-art simulators, with an accuracy of more than 95%. |
| first_indexed | 2024-03-10T00:11:16Z |
| format | Article |
| id | doaj.art-8d532de7a3634395a1c0baa31a52243c |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-10T00:11:16Z |
| publishDate | 2022-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-8d532de7a3634395a1c0baa31a52243c2023-11-23T15:59:00ZengMDPI AGElectronics2079-92922022-09-011118292810.3390/electronics11182928A Fine-Grained Modeling Approach for Systolic Array-Based AcceleratorYuhang Li0Mei Wen1Jiawei Fei2Junzhong Shen3Yasong Cao4School of Computer Science, National University of Defense Technology, Changsha 410073, ChinaSchool of Computer Science, National University of Defense Technology, Changsha 410073, ChinaSchool of Computer Science, National University of Defense Technology, Changsha 410073, ChinaSchool of Computer Science, National University of Defense Technology, Changsha 410073, ChinaSchool of Computer Science, National University of Defense Technology, Changsha 410073, ChinaThe systolic array provides extremely high efficiency for running matrix multiplication and is one of the mainstream architectures of today’s deep learning accelerators. In order to develop efficient accelerators, people usually employ simulators to make design trade-offs. However, current simulators suffer from coarse-grained modeling methods and ideal assumptions, which limits their ability to describe structural characteristics of systolic arrays. In addition, they do not support the exploration of microarchitecture. This paper presents FG-SIM, a fine-grained modeling approach for evaluating systolic array accelerators by using an event-driven method. FG-SIM can obtain accurate results and provide the best mapping scheme for different workloads due to its fine-grained modeling technique and deny of ideal assumption. Experimental results show that FG-SIM plays a significant role in design trade-offs and outperforms state-of-the-art simulators, with an accuracy of more than 95%.https://www.mdpi.com/2079-9292/11/18/2928modelingsystolic arrayaccelerator |
| spellingShingle | Yuhang Li Mei Wen Jiawei Fei Junzhong Shen Yasong Cao A Fine-Grained Modeling Approach for Systolic Array-Based Accelerator Electronics modeling systolic array accelerator |
| title | A Fine-Grained Modeling Approach for Systolic Array-Based Accelerator |
| title_full | A Fine-Grained Modeling Approach for Systolic Array-Based Accelerator |
| title_fullStr | A Fine-Grained Modeling Approach for Systolic Array-Based Accelerator |
| title_full_unstemmed | A Fine-Grained Modeling Approach for Systolic Array-Based Accelerator |
| title_short | A Fine-Grained Modeling Approach for Systolic Array-Based Accelerator |
| title_sort | fine grained modeling approach for systolic array based accelerator |
| topic | modeling systolic array accelerator |
| url | https://www.mdpi.com/2079-9292/11/18/2928 |
| work_keys_str_mv | AT yuhangli afinegrainedmodelingapproachforsystolicarraybasedaccelerator AT meiwen afinegrainedmodelingapproachforsystolicarraybasedaccelerator AT jiaweifei afinegrainedmodelingapproachforsystolicarraybasedaccelerator AT junzhongshen afinegrainedmodelingapproachforsystolicarraybasedaccelerator AT yasongcao afinegrainedmodelingapproachforsystolicarraybasedaccelerator AT yuhangli finegrainedmodelingapproachforsystolicarraybasedaccelerator AT meiwen finegrainedmodelingapproachforsystolicarraybasedaccelerator AT jiaweifei finegrainedmodelingapproachforsystolicarraybasedaccelerator AT junzhongshen finegrainedmodelingapproachforsystolicarraybasedaccelerator AT yasongcao finegrainedmodelingapproachforsystolicarraybasedaccelerator |