Guardado en:
| Autores principales: | , , , , , , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2404.05528 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866913305538330624 |
|---|---|
| author | Wang, Min Hou, Zhengyi Wang, Chenyi Yan, Zhengjie Li, Shixing Du, Ao Cai, Wenlong Li, Jinhao Zhang, Hongchao Cao, Kaihua Shi, Kewen Wang, Bi Zhao, Yuanfu Xiang, Qingyi Wang, Zhaohao Zhao, Weisheng |
| author_facet | Wang, Min Hou, Zhengyi Wang, Chenyi Yan, Zhengjie Li, Shixing Du, Ao Cai, Wenlong Li, Jinhao Zhang, Hongchao Cao, Kaihua Shi, Kewen Wang, Bi Zhao, Yuanfu Xiang, Qingyi Wang, Zhaohao Zhao, Weisheng |
| contents | We demonstrate approximate storage based on NAND-like spin-orbit torque (SOT) MRAM, through "device-modeling-architecture" explorations. We experimentally achieve down to 1E-5 level selectivity. Selectivity and low-power solutions are established by numerical calculation workflow. System-level power consumption is evaluated in the 512 KB last-level cache according to 5 quality levels. Error-tolerant applications, such as image processing, alleviate the demand for selectivity down to the 5E-2 level, leading to 54% ~ 61% energy-saving. Our proposal paves the novel and suitable path for high-density and low-power NAND-like SOT-MRAM. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_05528 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | NAND-like SOT-MRAM-based Approximate Storage for Error-Tolerant Applications Wang, Min Hou, Zhengyi Wang, Chenyi Yan, Zhengjie Li, Shixing Du, Ao Cai, Wenlong Li, Jinhao Zhang, Hongchao Cao, Kaihua Shi, Kewen Wang, Bi Zhao, Yuanfu Xiang, Qingyi Wang, Zhaohao Zhao, Weisheng Applied Physics We demonstrate approximate storage based on NAND-like spin-orbit torque (SOT) MRAM, through "device-modeling-architecture" explorations. We experimentally achieve down to 1E-5 level selectivity. Selectivity and low-power solutions are established by numerical calculation workflow. System-level power consumption is evaluated in the 512 KB last-level cache according to 5 quality levels. Error-tolerant applications, such as image processing, alleviate the demand for selectivity down to the 5E-2 level, leading to 54% ~ 61% energy-saving. Our proposal paves the novel and suitable path for high-density and low-power NAND-like SOT-MRAM. |
| title | NAND-like SOT-MRAM-based Approximate Storage for Error-Tolerant Applications |
| topic | Applied Physics |
| url | https://arxiv.org/abs/2404.05528 |