Stateful implication logic based on perpendicular magnetic tunnel junctions

被引：16

作者：

Cai, Wenlong ^{[1
]}

Wang, Mengxing ^{[1
]}

Cao, Kaihua ^{[1
]}

Yang, Huaiwen ^{[1
,2
]}

Peng, Shouzhong ^{[1
]}

Li, Huisong ^{[1
]}

Zhao, Weisheng ^{[1
,2
]}

机构：

[1] Beihang Univ, Fert Beijing Inst, Sch Integrated Circuit Sci & Engn, MIIT Key Lab Spintron, Beijing 100191, Peoples R China

[2] Beihang Univ, Qingdao Res Inst, Beihang Goertek Joint Microelect Inst, Qingdao 266104, Peoples R China

来源：

SCIENCE CHINA-INFORMATION SCIENCES | 2022年 / 65卷 / 02期

基金：

中国国家自然科学基金;

关键词：

perpendicular magnetic anisotropy magnetic tunnel junction; implication; spin transfer torque; spintronic implication logic gate; processing-in-memory; SPIN;

D O I：

10.1007/s11432-020-3189-x

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

As the conventional von Neumann architecture meets critical limitations of data transfer bandwidth and energy consumption, perpendicular magnetic anisotropy magnetic tunnel junction based processing-in-memory paradigm attracts extensive attention as a promising substitute thanks to its non-volatility, low-power switching, fast access and infinite endurance. In this work, we propose and experimentally demonstrate a new spintronic implication logic gate that consists of two parallel perpendicular magnetic anisotropy magnetic tunnel junctions with different diameters. Material implication and furthermore NAND logic functions are implemented by all electrically-controlled operations. The reliability of this structure is verified, especially in sub-20 nm node, which shows great potential for large-density processing-in-memory applications.

引用

页数：7

共 29 条

[1] A Scalable Processing-in-Memory Accelerator for Parallel Graph Processing [J].