Efficient cache replacement policy for minimizing error rate in L2-STT-MRAM caches
In recent times, various challenges have been encountered in the design and development of SRAM cache which consequently has led to a design where memory cell technologies are converted into on-chip embedded caches. The current research statistics for cache designing reveals that Spin Torque Transfe...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English English |
| Published: |
Inderscience Publishers
2018
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/66902/ http://irep.iium.edu.my/66902/ http://irep.iium.edu.my/66902/ http://irep.iium.edu.my/66902/30/66902%20Efficient%20cache%20replacement%20policy%20for%20minimising.pdf http://irep.iium.edu.my/66902/31/66902%20Efficient%20cache%20replacement%20policy%20for%20minimising%20SCOPUS.pdf |
| Summary: | In recent times, various challenges have been encountered in the design and development of SRAM cache which consequently has led to a design where memory cell technologies are converted into on-chip embedded caches. The current research statistics for cache designing reveals that Spin Torque Transfer Magnetic RAMs, preferably termed as STTMRAMs, have become one of the most promising technologies in the field of memory chip design, gaining a lot of attention from researchers due to its dynamic direct map and data access policies for reducing the average cost i.e. both time and energy optimization. Despite having efficient main memory access capability, L2-stochastic STT-MRAMs suffer due to high Write Error Rate (WER) caused by switching storage elements viz. Magnetic Tunnel Junction (MTJs) stochastically in write operations. It can be seen that cache replacement algorithms play a significant part in minimizing the Error Rate (ER) induced by write operations. The proposed study is intended to conceptualize an efficient cache replacement policy namely Minimum Error Rate (MER) along with Hamming Distance (HD) Computation to reduce the WER of L2-STTMRAM caches with acceptable overheads. The performance analysis of the proposed algorithm ensures its effectiveness in reducing the WER and cost overheads as compared to the conventional LRU technique implemented on SRAM cells. |
|---|