journal article·2023-02-01Open access

Hierarchical cache configuration based on hybrid SOT- and STT-MRAM

Shaopu Han, Qiguang Wang, Yanfeng Jiang

AIP Advances· Volume 13· Issue 2

Abstract

With the rapid growth of big data information and the continuous iteration progress of CPU architecture, the implementation of a new memory-based cache architecture is urgent and challenging. In the paper, a CPU cache architecture system based on MRAM is built. Firstly, the performance of SRAM, STT-MRAM and SOT-MRAM as caches from 8 kb to 32 Mb is evaluated. Secondly, by summarizing the performance of SRAM and MRAM in different cache levels, a new quad-core CPU cache architecture design scheme with SOT-MRAM as the first level of cache and STT-MRAM as the second level of cache is determined. Thirdly, the built cache system is simulated. A non-inclusive strategy is proposed to replace the inclusive strategy in order to solve the problem of high dynamic energy of STT-MRAM at the second level. The idea of having a quad-core CPU dynamically share the second-level cache is proposed in the paper. Finally, the caching system in the paper is compared with the other previous works, showing up to 60.78% energy consumption advantage and 33.22% leakage power advantage. The proposed MRAM-based CPU cache system and the corresponding cache strategy have potential application with the benefits of low power and less area.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Advanced Memory and Neural Computing · Parallel Computing and Optimization Techniques · Advanced Data Storage Technologies

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

Department of Electrical Engineering, School of Internet of Things (IoT) Engineering, Jiangnan University 1 , Wuxi, Jiangsu 214122, China

Institute of Advanced Technology, Jiangnan University 2 , Wuxi, Jiangsu 214122, China

Funding

National Natural Science Foundation of China

References (24)

Show source-provided references
1. New non-volatile logic based on spin-MTJ10.1002/pssa.200778135 ↗
2. Spin-transfer torque magnetic random access memory (STT-MRAM)10.1145/2463585.2463589 ↗
3. A 1-Mbit MRAM based on 1T1MTJ bit cell integrated with copper interconnects10.1109/jssc.2003.810048 ↗
4. Shape-anisotropy-controlled magnetoresistive response in magnetic tunnel junctions10.1063/1.118933 ↗
5. Multilevel spin-orbit torque MRAMs10.1109/ted.2014.2377721 ↗
6. Ultra-fast and high-reliability SOT-MRAM: From cache replacement to normally-off computing10.1109/TMSCS.2015.2509963 ↗
7. Study of two writing schemes for a magnetic tunnel junction based on spin orbit torque10.5281/zenodo.1086683 ↗
8. Microarchitectural exploration of STT-MRAM last-level cache parameters for energy-efficient devices10.1145/3490391 ↗
9. Write back energy optimization for STT-MRAM-based last-level cache with data pattern characterization10.1145/3381860 ↗
10. Comparative analysis of STT and SOT based MRAMs for last level caches10.1016/j.jmmm.2022.169161 ↗
11. Evaluation of hybrid memory technologies using SOT-MRAM for on-chip cache hierarchy10.1109/tcad.2015.2391254 ↗
12. System level exploration of a STT-MRAM based level 1 data-cache
13. Comparative analysis of spintronic memories for low power on-chip caches10.1142/S2010324720500277 ↗
14. Statistical traffic properties and model inference for shared cache interface in multi-core CPUs10.1109/access.2016.2603169 ↗
15. SRAM- and STT-RAM-based hybrid, shared last-level cache for on-chip CPU–GPU heterogeneous architectures10.1007/s11227-018-2389-3 ↗
16. HALLS: An energy-efficient highly adaptable last level STT-RAM cache for multicore systems10.1109/TC.2019.2918153 ↗
17. TA-LRW: A replacement policy for error rate reduction in STT-MRAM caches10.1109/tc.2018.2875439 ↗
18. Proactively invalidating dead blocks to enable fast writes in STT-MRAM caches10.1109/access.2022.3158493 ↗
19. ROCKY: A robust hybrid on-chip memory kit for the processors with STT-MRAM cache technology10.1109/tc.2020.3040152 ↗
20. CMOS latch bit-cell array for low-power SRAM design10.1587/elex.7.1145 ↗
21. NvSim: A circuit-level performance, energy, and area model for emerging nonvolatile memory10.1109/tcad.2012.2185930 ↗
22. Impact of die-to-die and within-die parameter variations on the clock frequency and throughput of multi-core processors10.1109/access.2016.2603169 ↗
23. The gem5 simulator10.1145/2024716.2024718 ↗
24. SPEC CPU2006 benchmark descriptions10.1145/1186736.1186737 ↗