한국정보과학회논문지:시스템및이론 (Journal of KIISE:Computer Systems and Theory)

한국정보과학회 (Korean Institute of Information Scientists and Engineers)
권호 표지

그리드 컴퓨팅의 다중 큐 하이브리드 작업스케줄링 기법

Multi-queue Hybrid Job Scheduling Mechanism in Grid Computing

  • 강창훈 (극동정보대학 방송영상미디어과) ;
  • 최창열 (숭실대학교 정보미디어기술연구소) ;
  • 박기진 (아주대학교 공과대학 산업정보시스템공학부) ;
  • 김성수 (아주대학교 정보통신전문대학원)
  • 발행 : 2007.08.15
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

그리드(Grid) 컴퓨팅은 지리적으로 분산된 컴퓨팅 자원들을 네트워크로 연동시켜 서로 공유 될 수 있도록 해주는 서비스이다. 본 논문에서는 그리드 컴퓨팅 시스템을 구성하는 전체 노드를 대상으로 작업을 분배하는 메타 스케줄링 정책과 특정 한 개의 노드 내에서 작업을 분배하는 작업 스케줄링을 동시에 고려하는 하이브리드 스케줄링 기법을 제안한다. 그리드 컴퓨팅 노드로 제출된 작업을 필요 프로세서 수와 예상 작업수행 시간에 따라 구분하여 우선순위가 높은 작업은 작업 큐(Job Queue)로, 우선순위가 낮은 작업과 원거리 작업은 백필 큐(Backfill Queue)로 할당시킴으로써, 그리드 컴퓨팅 시스템의 성능을 높이는 방법을 연구한다. 다양한 실험을 통하여 제안된 기법들의 성능을 평가하며, 그리드 컴퓨팅 시스템의 이용률이 높아지고, 작업 지연시간이 감소됨을 확인한다.

Grid computing is a service that share geographically distributed computing resources through high speed network. In this paper, we propose hybrid scheduling scheme which considers not only meta-scheduling scheme to distribute the job between the nodes of grid computing system but also the job scheduling to distribute the job within the local nodes. According to the number of processors needed and expected execution time, the job with high priority is allocated to job queue while the one with low priority and remote job are allocated to backfill queue. We evaluate the proposing scheme through the various experiments and the results show that the utilization of grid computing system increases and the job slowdown decreases.

키워드

참고문헌

  1. I. Foster, et al., 'Grid Services for Distributed System Integration,' Computer, Vol. 35, No.6, pp. 37-46, 2002
  2. K. Krauter, et al., 'A Taxonomy and Survey of Grid Resource Management Systems for Distributed Computing,' Software Practice and Experience Journal, Vol. 32, No.2, pp. 135-164, Feb. 2002 https://doi.org/10.1002/spe.432
  3. I. Foster, et al., 'Globus: A Metacomputing Infrastructure Toolkit,' The International Journal of Supercomputer Applications and Performance Computing, Vol. 11, No.2, pp. 115-128, Oct. 1997 https://doi.org/10.1177/109434209701100205
  4. http://www.legion.org
  5. B. Lawson, et al., 'Multiple-queue Backfilling Scheduling with Priorities and Reservations for Parallel Systems,' The 8th International Workshop, JSSPP 2002 Edinburgh, Scotland, UK, pp. 72-87, July 2002
  6. H. Shan, et al., 'Job Superscheduler Architecture and Performance in Computational Grid Environments,' In SC2003 Conference, 2003
  7. A. Mualem, et al., 'Utilization, Predictability, Workloads and User Run time Estimates in Scheduling the IBM SP2 with Backfilling,' IEEE Trans. Parallel and Distributed System, Vol. 12, No.6, pp. 529-543, June 2001 https://doi.org/10.1109/71.932708
  8. O. H. Ibarra, et al., 'Heuristic Algorithm for Scheduling Independent Tasks on Nonidentical Processors,' Journal of ACM, Vol. 24, No.2, pp. 280-289, Apr. 1977 https://doi.org/10.1145/322003.322011
  9. V. Harnscher, et. al., 'Evaluation of Job-Scheduling Strategies for Grid Computing,' The 1st IEEE/ ACM International Workshop on Grid Computing (Grid 2000) at the 7th International Conference on High Performance Computing (HiPC-2000), LNCS 1971, pp. 191-202, 2000
  10. V. Subramani, et al., 'Distributed Job Scheduling on Computational Grids Using Multiple Simultaneous Requests,' The 11th IEEE International Symposium on High Performance Distributed Computing (HPDC-11 2002), pp. 359-368, July 2002
  11. Q. Wang, et al., 'De-centralized Job Scheduling on Computational Grids Using Distributed Backfilling,' Grid and Cooperative Computing - GCC 2004: Third International Conference, LNCS 3251, pp. 285-292, Oct. 2004
  12. K. Li, 'Job Scheduling for Grid Computing on Metacomputers,' The 19th International Parallel and Distributed Processing Symposium (IPDPS 2005), Abstracts Proceedings, Apr. 2005
  13. B. Bode, et al., 'The Portable Batch Scheduler and the MauiScheduler on Linux Clusters,' in Proceedings of the 4th Annual Linux Showcase and Conference, Atlanta, Georgia, Oct. 2000
  14. D. G. Feitelson, et al., 'Theory and Practice in Parallel Job Scheduling,' Job Scheduling Strategies for Parallel Processing, IPPS'97 Workshop, Geneva, Switzerland, LNCS 1291, pp. 1-34, Apr. 5, 1997
  15. R. McCann, et al., 'A Dynamic Processor Allocation Policy for Multiprogrammed Sharedrnemory Multiprocessors,' ACM Trans. on Computer System, Vol. 11, No.2, pp. 146-178, May 1993 https://doi.org/10.1145/151244.151246
  16. D. G. Feitelson, et al., 'Improved Utilization and Responsiveness with Gang Scheduling,' Job Scheduling Strategies for Parallel Processing, IPPS'97 Workshop, Geneva, Switzerland, LNCS 1291, pp. 238-261, Apr. 1997
  17. D. Zotkin, et al., 'Job-Length Estimation and Performance in Backfilling Schedulers,' The 8th IEEE International Symposium on High Performance Distributed Computing (HPDC'99), Aug. 1999
  18. S. Srinivasan, et al., 'Characterization of Backfilling Strategies for Parallel Jobs Scheduling,' 31st International Conference on Parallel Processing Workshops (ICPP 2002 Workshops), pp. 514-522, Aug. 2002
  19. D. Talby, et al., 'Supporting Priorities and Improving Utilization of the IBM SP2 Scheduler Using Slack-based Backfilling,' In Proceedings of the Intemational Parallel Processing Symposium, pp. 513-517, Apr. 1999
  20. W. A. Ward Jr., et al., 'Scheduling Jobs on Parallel Systems Using a Relaxed Backfill Strategy,' 8th International Workshop on Job Scheduling Strategies for Parallel Processing (JSSPP'2002), pp. 88-102, July 2002
  21. D. Feitelson, 'Logs of Real Parallel Workloads from Production Systems,' http://www.cs.huji.ac.il/labs/parallel/workload/logs.html