The Transactions of the Korea Information Processing Society (한국정보처리학회논문지)

Korea Information Processing Society (한국정보처리학회)
권호 표지

A Disk Allocation Scheme for High-Performance Parallel File System

고성능 병렬화일 시스템을 위한 디스크 할당 방법

  • 박기현 (계명대학교 컴퓨터 및 전자공학부)
  • Published : 2000.09.01
Download PDF
⟨ Previous Next ⟩

Abstract

In recent years, much attention has been focused on improving I/O devices' processing speed which is essential in such large data processing areas as multimedia data processing. And studies on high-performance parallel file systems are considered to be one of such efforts. In this paper, an efficient disk allocation scheme is proposed for high-performance parallel file systems. In other words, the concept of a parallel disk file's parallelism is defined using data declustering characteristic of a given parallel file. With the concept, an efficient disk allocation scheme is proposed which calculates the appropriate degree of data declustering on disks for each parallel file in order to obtain the maximum throughput when more than one parallel file is used at the same time. Since, calculation for obtaining the maximum throughput is too complex as the number of parallel files increases, an approximate disk allocation algorithm is also proposed in this paper. The approximate algorithm is very simple and especially provides very good results when I/O workload is high. In addition, it has shown that the approximate algorithm provides the optimal disk allocation for the maximum throughput when the arrival rate of I/O requests is infinite.

최근, 멀티미디어 정보처리와 같은 대규모 데이터 처리에 필수적인 입출력 시스템의 성능을 높이기 위하여 많은 관심이 집중되고 있으며, 고성능 병렬화일 시스템에 관한 연구도 이런 노력에 속한다. 본 연구에서는 고성능 병렬화일 시스템을 위한 효율적인 디스크 할당 방법을 제안한다. 즉, 병렬화일의 자료 분산(data declustering)특성을 이용하여 병렬화일에 대한 병렬도 개념을 정의하고, 이를 기반으로 여러 병렬화일들이 동시에 처리되는 경우에, 최대의 작업처리량(throughput)을 얻기 위한 각 병렬화일에 적합한 디스크상의 자료 분산 정도를 계산하는 방법을 제안한다. 또한 동시에 처리되는 병렬화일들이 많이 늘어날수록, 최대의 작업처리량을 얻기 위한 계산이 너무 복잡해지므로, 효율적인 근사 디스크 할당 알고리즘도 아울러 제안한다. 제안된 근사 알고리즘은 계산이 간단하고, 특히 입출력 작업부하(workload)가 높은 환경에서는 매우 효율적임을 보여준다. 또한 입출력 요구들의 도착 비율이 무한대일 경우, 근사 알고리즘만을 이용하여도 최대 작업처리량을 위한 최적 디스크 할당을 얻을 수 있음을 증명하였다.

Keywords

References

  1. P. M. Chen, et. al., 'An evaluation of redundant arrays of disks using an Amdahl 5890,' Proc. of SIGMETRICS Conf. on Measurement and Modell-ing of Computer Systems, pp.74-85, May 1990 https://doi.org/10.1145/98457.98509
  2. J. G. Choi and K. H. Park, 'Performance comparisons of processor allocation schemes using simulation techniques,' Journal of the Korea Society for Simulation, Vol.3, No.1, pp.43-54, July 1994
  3. S. D. Conte and C. Boor, Elementary Numerical Analysis : An Algorithmic Approach, 3rd. ed., McGraw-Hill, 1980
  4. P. F. Corbett, 'Overview of the Vesta parallel file system,' Computer Architecture News, Vol.21., No.5, pp.7-14, 1993 https://doi.org/10.1145/165660.165662
  5. P. C. Dibble, M. L. Scott and C. S. Eliss, 'Bridge : a high-performance file system for parallel processing,' 8th Int. Conf. on Dist. Comput. Syst., pp.154-161, 1988 https://doi.org/10.1109/DCS.1988.12513
  6. L. W. Dowdy and M. R. Leuze, 'On modeling par-titioned multiprocessor systems,' Int. Journal of High Speed Computing, Vol.5, Sept. 1993
  7. D. L. Eager, J. Zahorjan, and E. D. Lazowska, 'Speedup versus efficiency in parallel systems,' IEEE Trans. on Computers, Vol.38, pp.408-423, Mar. 1989 https://doi.org/10.1109/12.21127
  8. G. R. Ganger, et. al., 'Disk arrays : high-Performance, high-reliability storage subsystems,' IEEE COMPUTER, Vol.27, No.3, pp.30-36, Mar. 1994 https://doi.org/10.1109/2.268882
  9. R. Karedla, J. S. Love, and B. G. Wherry, 'Caching strategies to improve disk system performance,' IEEE COMPUTER, Vol.27, No.3, pp.38-46, Mar. 1994 https://doi.org/10.1109/2.268884
  10. R. H. Katz, et. al., 'Disk system architectures for high performance computing,' Proc. of the IEEE, Vol.77, No.12, pp.1842-1858, Dec. 1989 https://doi.org/10.1109/5.48827
  11. S. Kenneth and M. H. Garcia, 'Disk striping,' Proc. of the IEEE Data Engineering Conf, Los Angeles, CA, pp.336-342, Feb. 1986
  12. M. Y. Kim, 'Synchronized disk interleaving,' IEEE Trans. on Computers, C-35(11), pp.978-988, Nov. 1986
  13. L. Kleinrock, Queueing Systems Vol.1 : Theory, John Wiley & Sons, New York, 1975
  14. E. K Lee, 'Software and performance issues in the implementation of a RAID prototype,' Tech Report UCB/CSD 93, Mar. 1993
  15. S. J. LoVerso, M. Isman, and A. Nanopoulos, 'sfs : a parallel file system for the CM-5,' Proc. of 1993 Summer USENIX, pp.291-304, 1993
  16. E. L. Miller and R. H. Katz, 'RAMA : A File System for Massively-Parallel Computers,' 12th IEEE Symp. on Mass Storage Systems, pp.163-198, 1993 https://doi.org/10.1109/MASS.1993.289764
  17. N. Nieuweiaar and D. Kotz, 'The Galley Parallel File System,' ACM ICS '96, Philadelphia, USA, pp.374-381, 1996
  18. K. H. Park, Dynamic processor partitioning for multiporgrammed multiprocessor systems. Ph.D. Thesis, Vanderbilt Univ., Nashville, TN, 1990
  19. D. A. Patterson, G. Gibson and R. H. Katz, 'Introduction to redundant arrays of inexpensive disks (RAID),' Proc. of Spring COMCON, pp.112-117, Feb. 1989 https://doi.org/10.1109/CMPCON.1989.301912
  20. P. Pierce, 'A concurrent file system for a highly parallel mass storage subsystem,' 4th Conf. on Hypercubes, Concurrent Computers and Appliartions, pp.161-166, Mar. 1989
  21. T. W. Pratt, et. al., 'A comparison of the architecture and performance of two parallel file systems,' 4th Conf. on Hypercubes, Concurrent Computers and Applications, pp.161-166, 1989
  22. A. N. Reddy and P. Banerjee, 'An evaluation of multiple-disk I/O systems,' IEEE Trans. on COMPUTERS, Vol.38, No.12, pp.1680-1690, Dec. 1989 https://doi.org/10.1109/12.40846
  23. A. N. Reddy and J. C. Wyllie, 'I/O issues in a multimedia system,' IEEE COMPUTER, Vol.27, No.3, pp.69-74, Mar. 1994 https://doi.org/10.1109/2.268888
  24. J. Rosario and A. N. Choudhary, 'High-performance I/O for massively parallel computers : problems and prospects,' IEEE COMPUTER, Vol.27, No.3, pp.59-68, Mar. 1994 https://doi.org/10.1109/2.268887
  25. K. Salem and G. Garcia-Molina, 'Disk striping,' Proc. of 2nd IEEE Data Engr. Conf, pp.336-342, Feb. 1986