전자공학회논문지CI (Journal of the Institute of Electronics Engineers of Korea CI)

대한전자공학회 (The Institute of Electronics and Information Engineers)
권호 표지

분산 이기종 컴퓨팅 시스템에서 효율적인 리스트 스케줄링 알고리즘

An Efficient List Scheduling Algorithm in Distributed Heterogeneous Computing System

  • 발행 : 2009.05.25
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

이기종 컴퓨팅 환경에서 방향성 비순환 그래프(directed acyclic graph DAG)의 효율적인 스케줄링은 시스템의 성능을 높게 만드는데 매우 중요한 역할을 한다. 이기종의 컴퓨팅 환경에서 DAG로 표현되는 프로그램의 최적 스케줄링 방법을 찾는 것은 잘 알려진 '정해진 시간 내에 해결하기 어려운 문제(NP-complete)' 이다. 본 논문은 분산 이기종 컴퓨팅 시스템에서 병렬로 실행 가능한 프로그램을 위한 새로운 리스트 스케줄링 알고리즘인 HRPS(Heterogeneous Rank-Path Scheduling)를 제안하였다. HRPS의 가장 궁극적인 목적은 프로그램의 실행시간을 최소화하는 것이다. 알고리즘의 성능을 위해 DAG 입력 그래프를 이용하여 기존에 제안되어진 CPOP, HCPT, FLB 알고리즘과 스케줄의 길이를 비교한 결과 성능 향상의 결과를 얻을 수 있었다.

Efficient DAG scheduling is critical for achieving high performance in heterogeneous computing environments. Finding an optimal solution to the problem of scheduling an application modeled by a directed acyclic graph(DAG) onto a set of heterogeneous machines is known to be an NP-complete problem. In this paper we propose a new list scheduling algorithm, called the Heterogeneous Rank-Path Scheduling(HRPS) algorithm, to exploit all of a program's available parallelism in distributed heterogeneous computing system. The primary goal of HRPS is to minimize the schedule length of applications. The performance of the algorithm has been observed by its application to some practical DAGs, and by comparing it with other existing scheduling algorithm such as CPOP, HCPT and FLB in term of the schedule length. The comparison studies show that HRPS significantly outperform CPOP, HCPT and FLB in schedule length.

키워드

참고문헌

  1. T.Braun, H.J. Siegel, N. Beck, L.L. Boloni, M.Maheswaran, A.I.Reuther, J.P Robertson, M.D. Theys, B.Yao, D.Hengsen, and R.F. Freund, 'A Comparison Study of Static Mapping Heuristics for a Classes of Meta-Tasks on Heterogeneous Computing Systems,' Proc, Heterogeneous Computing Workshop, pp.15-29, 1999
  2. Oliver Sinnen, 'Task Scheduling For Parallel Systems,' Wiley, pp.7-35, 2007
  3. J.D Ullman, 'NP-Complete Scheduling Problems,' J.Computer and Systems Sciences, vol. 10, pp. 384-393, 1975 https://doi.org/10.1016/S0022-0000(75)80008-0
  4. Y.K. Kwok and I. Ahmad. 'Static scheduling algorithms for allocating directed task graphs to multiprocessors,' ACM Comput. Surv., 31(4). pp. 406–471, 1999 https://doi.org/10.1145/344588.344618
  5. Y.K Kwok and I. Ahmad. 'Dynamic critical-path scheduling: an effective technique for allocating task graphs to multiprocessors.' IEEE Trans. on Parallel and Distributed Systems, pp. 506-521, 1996
  6. H. Togcuglou, S, Hariri and M.Y. Wu, 'Performance Effective and Low-Complexity Task Scheduling for Heterogeneous Computing,' IEEE Trans. On Parallel and Distributed Systems, vol 13, No.3, Feb. 2002 https://doi.org/10.1109/71.993206
  7. T. Hagras and J. Janecek, 'A Simple Scheduling Heuristic for Heterogeneous Computing. Environments,' IEEE Proceedings of Second International Symposium on Parallel and. Distributed Computing (ISPDC'03), pp. 104–110, October 2003 https://doi.org/10.1109/ISPDC.2003.1267650
  8. A.Radulescu, and A.van Gemund, 'Performance Effective and Low-complexity Task Scheduling for Heterogeneous Computing,' 9th Heterogeneous Computing Workshop, pp.229- 238. 2000 https://doi.org/10.1109/71.993206
  9. T.Hagras, J.Janecek 'A High Performance, Low Complexity Algorithm for Compile-Time Task Scheduling in Heterogeneous Systems,' Parallel and Computing, 31, 653-670, 2005 https://doi.org/10.1016/j.parco.2005.04.002
  10. S.Darbha, D.P.Agrawal. 'Optimal Scheduling Algorithm for Distributed-Memory Machines,' IEEE Trans. Parallel and Distributed Systems, 9(1), 97-95, Jan. 1998 https://doi.org/10.1109/71.663861
  11. Lan Zhou, Sun Shixin, 'Scheduling algorithm based on critical tasks in heterogeneous environments,' Journal of Systems Engineering and Electronics, Vol. 19, No. 2, pp.398-404, 2008 https://doi.org/10.1016/S1004-4132(08)60099-7
  12. Kafil, M. and I. Ahmed, 'Optimal task assignment in heterogeneous distributed computing systems,' IEEE Concurrency, 6: 42-51, 1998 https://doi.org/10.1109/4434.708255
  13. Ranaweera, A. and D.P. Agrawal, 'A task duplication based algorithm for heterogeneous systems,' Proc. IPDPS, pp: 445-450, 2000
  14. Cristina Boeres, Jos'e Viterbo Filho and Vinod E. F. Rebello, 'A cluster-based strategy for scheduling task on heterogeneous processors,' Proc. 16th Symp. on Computer Architecture and High Performance Computing (SBAC-PAD), 2004 https://doi.org/10.1109/CAHPC.2004.1
  15. Y.K. Kwok, I. Ahmad, Benchmarking and comparison of the task graph scheduling algorithms, J. Parallel Distrib. Comput. 59(3). pp. 381–422,1999 https://doi.org/10.1006/jpdc.1999.1578
  16. Takao Tobita and Hironory kasahara, 'A Standdart Task Graph Set for Fair Evaluation of Multiprocessor Scheduling Algorithms,' Journal of Scheduling, 5, pp. 379-394, 2002 https://doi.org/10.1002/jos.116
  17. http://www.kasahara.elec.waseda.ac.jp.