하이브리드 하드디스크를 위한 효율적인 데이터 블록 교체 및 재배치 기법

An Efficient Data Block Replacement and Rearrangement Technique for Hybrid Hard Disk Drive

  • 발행 : 2010.01.15

초록

최근 낸드 플래시 메모리가 하드디스크 수준으로 읽기 성능이 향상되고, 전력소비가 훨씬 적음에 따라, 플래시메모리와 하드디스크를 같이 사용하는 하이브리드 하드디스크와 같은 이기종 저장장치들이 출시되고 있다. 하지만 낸드 플래시 메모리의 쓰기 및 삭제 속도가 기존 자기디스크의 쓰기 성능에 비해 매우 느릴 뿐 아니라, 사용자 층에서 쓰기 요청이 집중될 경우 CPU, 메인 메모리에 심각한 오버헤드를 발생시킨다. 본 논문에서는 비휘발성 캐시의 역할을 하는 낸드 플래시 메모리의 성능을 향상시키기 위해 읽기의 참조 빈도는 낮고, 쓰기의 갱신 빈도가 높은 데이터 블록들을 교체하는 LFU(Least Frequently Used)-Hot 기법을 제시하고, 교체 될 데이터 블록들을 재배치하여 자기디스크로 플러싱하는 기법을 제시한다. 실험 결과, 본 논문에서 제안하는 LFU-Hot 블록 교체 기법과 멀티존 기반의 데이터 블록 재배치기법 실행시간이 기존 LRU, LFU 블록 교체 기법들보다 입출력 성능 면에서 최대 38% 빠르고, 비휘발성 캐시의 수명을 약 40% 이상 향상 시킴을 증명하였다.

Recently heterogeneous storage system such as hybrid hard disk drive (H-HDD) combining flash memory and magnetic disk is launched, according as the read performance of NAND flash memory is enhanced as similar to that of hard disk drive (HDD) and the power consumption of NAND flash memory is reduced less than that of HDD. However, the read and write operations of NAND flash memory are slower than those of rotational disk. Besides, serious overheads are incurred on CPU and main memory in the case that intensive write requests to flash memory are repeatedly occurred. In this paper, we propose the Least Frequently Used-Hot scheme that replaces the data blocks whose reference frequency of read operation is low and update frequency of write operation is high, and the data flushing scheme that rearranges the data blocks into the multi-zone of the rotation disk. Experimental results show that the execution time of the proposed method is 38% faster than those of conventional LRU and LFU block replacement schemes in I/O performance aspect and the proposed method increases the life span of Non-Volatile Cache 40% higher than those of conventional LRU, LFU, FIFO block replacement schemes.

키워드

참고문헌

  1. Windsor W. Hsu, Alan Jay Smith, and Honesty C. Young, "The automatic improvement of locality in storage systems," ACM Transactions of Computer Systems, no.23(4), pp.424-473, 2005. https://doi.org/10.1145/1113574.1113577
  2. Timothy Bisson, Scott A. Brandt, and Darrell D.E. Long, "NVCache: Increasing the effectiveness of disk spin-down algorithms with caching," In Proc. of the 14th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, pp.422-432, Sep. 2006.
  3. Kwanghee Park, Jun-Sik Yang, Joon-Hyuk Chang, and Deok-Hwan Kim, "Anticipatory I/O Management for Clustered Flash Translation Layer in NAND Flash Memory," ETRI Journal., vol.30, no.6, pp.790-798, Dec. 2008. https://doi.org/10.4218/etrij.08.0108.0145
  4. Jun-Sik Yang, Young-Wook Ko, Chan-Gun Lee, and Deok-Hwan Kim, "Design and Implementation of Hybrid Hard Disk I/O System based on n-Block Prefetching for Low Power Consumption and High I/O Performance," Journal of the KIISE : Computer Systems and Theory, vol.36, no.6, pp. 451-462, Dec. 2009. (in Korean)
  5. Luis Useche, Jorge Guerra, Medha Bhadkamar, Mauricio Alarcon and Raju Rangaswami, "EXCES: EXternal Caching in Energy Saving Storage Systems," In Proc. of the IEEE International Symposium on High-Performance Computer Architecture, pp.89-100, Feb. 2008.
  6. Avi Silberschatz, Peter Baer Galvin, and Greg Gagne, "Operating System Concepts," John Wiley & Sons, Inc, 2008.
  7. Timothy Bisson, Scott A. Brandt, "Flushing policies for NVCache enabled hard disks," In Proc. of the 24th IEEE Conference on Mass Storage Systems and Technologies, pp.299-304, Sep. 2007.
  8. Donghee Lee, Sam H. Noh, Sang Lyul Min, Yookun Cho, "LRFU: A Block Replacement Policy which exploits Infinite History of References," Journal of the KIISE(A), vol.24, no.7, pp.632-641, July 1997. (in Korean)
  9. Seon-yeong Park, Dawoon Jung, Jeong-uk Kang, Jin-soo Kim, and Joonwon Lee, "CFLRU: A Replacement Algorithm for Flash Memory," In Proc. of the 2006 International Conference on Compilers, Architecture and Synthesis for Embedded systems (CASE), pp.234-241, Oct. 2006.
  10. Yun-Seok Yoo, Hyejeong Lee, Yeonseung Ryu, and Hyokyung Bahn, "Page Replacement Algorithms for NAND Flash Memory Storages," In Proc. of the 2007 International Conference Computational Science and Its Applications (ICCSA), pp.201-212, Oct. 2007.
  11. Heeseung Jo, Jeong-Uk Kang, Seon-Yeong Park, Jin-Soo Kim, Joonwon Lee, "FAB: flash-aware buffer management policy for portable media players," IEEE Transactions on Consumer Electronics, vol.52, Issue 2, pp.485-493.
  12. Hyojun Kim and Seongjun Ahn, "BPLRU: A Buffer Management Scheme for Improving Random Writes in Flash Storage," The 6th USENIX File And Storage Technologies (FAST'08), San Jose, California, USA. Feb. 26-29, 2008.
  13. Fran Wang, Y. Deng, N. Helian, S. Wu, V. Khare, C. Liao, and A. Parker, "Evolutionary storage: speeding up a magnetic disk by clustered frequent data," IEEE Transactions on Magnetics, vol.43, Issue 6, pp.2295-2297, June 2007. https://doi.org/10.1109/TMAG.2007.892331
  14. Shahram Ghandeharizadeh, Douglas J. Ierardi, Dongho Kim, and Roger Zimmermann, "Placement of data in multi-zone disk drives," In Proc. of the 2nd International Baltic Workshop on Database and Information Systems, June 1996.
  15. Byung-gu Yu and Seon-ho Kim, "An efficient zoning technique for multi-dimensional access methods," In Proc. of the VLDB Workshop on Trends in Enterprise Application Architecture, Aug. 2005.
  16. Ruston Panabaker, "Hybrid hard disk and Ready-Drive technology: Improving performance and power for Windows Vista mobile PCs," Microsoft Windows Hardware Engineering Conference, 2006.
  17. IOZone Filesystem Benchmark, http://www.iozone.org
  18. Network Appliance, "Postmark: a new file system benchmark," Technical Report TR-3022, October 1997.