SynDCIM: A Performance-Aware Digital Computing-in-Memory Compiler with Multi-Spec-Oriented Subcircuit Synthesis
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866916551991492608 |
|---|---|
| author | Shao, Kunming Tian, Fengshi Wang, Xiaomeng Zheng, Jiakun Chen, Jia He, Jingyu Wu, Hui Chen, Jinbo Guan, Xihao Deng, Yi Tu, Fengbin Yang, Jie Sawan, Mohamad Cheng, Tim Kwang-Ting Tsui, Chi-Ying |
| author_facet | Shao, Kunming Tian, Fengshi Wang, Xiaomeng Zheng, Jiakun Chen, Jia He, Jingyu Wu, Hui Chen, Jinbo Guan, Xihao Deng, Yi Tu, Fengbin Yang, Jie Sawan, Mohamad Cheng, Tim Kwang-Ting Tsui, Chi-Ying |
| contents | Digital Computing-in-Memory (DCIM) is an innovative technology that integrates multiply-accumulation (MAC) logic directly into memory arrays to enhance the performance of modern AI computing. However, the need for customized memory cells and logic components currently necessitates significant manual effort in DCIM design. Existing tools for facilitating DCIM macro designs struggle to optimize subcircuit synthesis to meet user-defined performance criteria, thereby limiting the potential system-level acceleration that DCIM can offer. To address these challenges and enable agile design of DCIM macros with optimal architectures, we present SynDCIM, a performance-aware DCIM compiler that employs multi-spec-oriented subcircuit synthesis. SynDCIM features an automated performance-to-layout generation process that aligns with user-defined performance expectations. This is supported by a scalable subcircuit library and a multi-spec-oriented searching algorithm for effective subcircuit synthesis. The effectiveness of SynDCIM is demonstrated through extensive experiments and validated with a test chip fabricated in a 40nm CMOS process. Testing results reveal that designs generated by SynDCIM exhibit competitive performance when compared to state-of-the-art manually designed DCIM macros. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_16806 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | SynDCIM: A Performance-Aware Digital Computing-in-Memory Compiler with Multi-Spec-Oriented Subcircuit Synthesis Shao, Kunming Tian, Fengshi Wang, Xiaomeng Zheng, Jiakun Chen, Jia He, Jingyu Wu, Hui Chen, Jinbo Guan, Xihao Deng, Yi Tu, Fengbin Yang, Jie Sawan, Mohamad Cheng, Tim Kwang-Ting Tsui, Chi-Ying Hardware Architecture Digital Computing-in-Memory (DCIM) is an innovative technology that integrates multiply-accumulation (MAC) logic directly into memory arrays to enhance the performance of modern AI computing. However, the need for customized memory cells and logic components currently necessitates significant manual effort in DCIM design. Existing tools for facilitating DCIM macro designs struggle to optimize subcircuit synthesis to meet user-defined performance criteria, thereby limiting the potential system-level acceleration that DCIM can offer. To address these challenges and enable agile design of DCIM macros with optimal architectures, we present SynDCIM, a performance-aware DCIM compiler that employs multi-spec-oriented subcircuit synthesis. SynDCIM features an automated performance-to-layout generation process that aligns with user-defined performance expectations. This is supported by a scalable subcircuit library and a multi-spec-oriented searching algorithm for effective subcircuit synthesis. The effectiveness of SynDCIM is demonstrated through extensive experiments and validated with a test chip fabricated in a 40nm CMOS process. Testing results reveal that designs generated by SynDCIM exhibit competitive performance when compared to state-of-the-art manually designed DCIM macros. |
| title | SynDCIM: A Performance-Aware Digital Computing-in-Memory Compiler with Multi-Spec-Oriented Subcircuit Synthesis |
| topic | Hardware Architecture |
| url | https://arxiv.org/abs/2411.16806 |