Low-End Memory Subsystem Optimization Process
AUTHORS
Jungseok Cho,EE, Sunchon National University, South Korea
Jeongdu Lee,EE, Sunchon National University, South Korea
Doosan Cho,EE, Sunchon National University, South Korea
ABSTRACT
Traditional memories can be categorized to its characteristics like shortcomings advantages. DRAM and SRAM are volatile memories. SRAMs are faster than DRAMs, but they are used in cache memories because they have low integration and high energy consumption compared to DRAMs. On the other hand, DRAM has a slower processing speed than SRAM, but has high integration and low energy consumption. Since the flash memory is a nonvolatile memory, it is used in portable devices. However, it is slower than DRAM and consumes much energy. Non-volatile memory is a new memory that is attracting attention as the next generation memory. Non-volatile memory is a memory that has both higher integration density than DRAM, and low energy consumption, high-speed operation of SRAM, and non-volatile nature of flash memory, which is advantage of conventional memory. For this reason, it is expected to replace DRAM, flash memory, and hard disk in the future. However, Non-volatile memory is not enough to replace DRAM. In case of using non-volatile memory, higher energy consumption may consume than using DRAM only. To solve this problem, it is necessary to study hybrid memories using DRAM and non-volatile memory together. Hybrid memory can be a solution to the slow processing speed and high energy consumption of the negative effect from non-volatile memory technique.
KEYWORDS
Energy consumption, Hybrid memory, Embedded system, Nonvolatile memory, Low-end memory subsystemr.
REFERENCES
[1] P. Zhou, B. Zhao, J. Yang, and Y.-T. Zhang. “A Durable and Energy Efficient Main Memory Using Phase Change Memory Technology”. Computer Architecture News, pp.14-23 (2009)
[2] B. Lee, P. Zhou, J. Yang, Y. Zhang, B. Zhao, E. Ipek, O. Mutlu, D. Burger. “Phase-Change Technology and the Future of Main Memory”. IEEE Micro, vol.30, no.1 (2010), pp.143. [DOI: 10.1109/MM.2010.24](CrossRef)(Google Scholar)
[3] J. Hu, C. Xue,W. Tseng, Q. Zhuge and E. H.-M. Sha. “Minimizing Write Activities to Non-volatile Memory via Scheduling and Recomputation”. Proc. 8th IEEE Symposium on Application Specific Processors, (2010), pp.7-12. [DOI: 10.1109/SASP.2010.5521139](CrossRef)(Google Scholar)
[4] J. Hu, C. Xue, W. Tseng, Y. He, M. Qiu and E. H.-M. Sha. “Reducing Write Activities on Non-volatile Memories in Embedded CMPs via Data Migration and Recomputation”. Proceedings 47th IEEE/ACM Design Automation Conference, (2010), pp.350-355. [DOI: 10.1145/1837274.1837363](CrossRef)(Google Scholar)
[5] L. Shi, C. Xue, J. Hu, W. Tseng and E. H.-M. Sha. “Write Activity Reduction on Flash Main Memory via Smart Victim Cache”. Proceedings ACM/IEEE 20th Great Lakes Symposium on VLSI, (2010), pp.91-94. [DOI: 10.1145/1785481.1785503](CrossRef)(Google Scholar)
[6] C. H. Gebotys. “Low Energy Memory and Register Allocation Using Network Flow”. Proceedings of the 34th Annual Design Automation Conference, (1997), pp.435-440
[7] Hyungmin Cho , Bernhard Egger , Jaejin Lee , Heonshik Shin. “Dynamic data scratchpad memory management for a memory subsystem with an MMU”. ACM LCTES, (2007), pp.13-15. [DOI: 10.1145/1273444.1254804](CrossRef)(Google Scholar)
[8] S. Kang and A. G. Dean. “Leveraging both data cache and scratchpad memory through synergetic data allocation”. IEEE Real-Time and Embedded Technology and Applications Symposium, (2012) [DOI: 10.1109/RTAS.2012.22](CrossRef)(Google Scholar)