Load Balancing of Heterogeneous Workstation Cluster based on Relative Load Index

상대적 부하 색인을 기반으로 한 이기종 워크스테이션 클러스터의 부하 균형

  • 지병준 (한림정보산업대학 전산정보처리과) ;
  • 이광모 (한림대학교 컴퓨터공학과)
  • Published : 2002.04.01

Abstract

The clustering environment with heterogeneous workstations provides the cost effectiveness and usability for executing applications in parallel. Load balancing is considered a necessary feature for a cluster of heterogeneous workstations to minimize the turnaround time. Previously, static load balancing that assigns a predetermined weight for the processing capability of each workstation, or dynamic approaches which execute a benchmark program to get relative processing capability of each workstation were proposed. The execution of the benchmark program, which has nothing to do with the application being executed, consumes the computation time and the overall turnaround time is delayed. In this paper, we present efficient methods for task distribution and task migration, based on the relative load index. We designed and implemented a load balancing system for the clustering environment with heterogeneous workstations. Turnaround times of our methods and the round-robin approach, as well as the load balancing method using a benchmark program, were compared. The experimental results show that our methods outperform all the other methods that we compared.

References

  1. http://www.specbench.org/osg/cpu2000/results/cpu2000.html
  2. 지병준, 이광모, '병렬 가상 컴퓨터에서 작업 부하를 고려한 타스크 관리자의 설계 및 구현', 정보과학회병렬처리연구회 학술발표논문집, 제7권 2호, 1996.5
  3. Jeremy Casas, Jonathan Walpole, 'MPVM: A Migration Transparent Version of PVM', Tech report, Dept of Computer Science and Engineering, Oregon Graduate Institute of Science & Technology, Feb 1995
  4. J. Casas, D. Clark, P Galbiati, R. Konuru, S. Otto, 'MIST:PVM with Transparent Migration and Checkpoint', Dept of computer Science and Engineering Oregon Graduate Institute, 1995
  5. Leen Dikken, Peter Sloot, 'Dynamic PVM, Dynamic Load Balancing on Parallel System', High Performance Computing and Networking, Vol. 797, pp. 273-277, 1994 https://doi.org/10.1007/3-540-57981-8_130
  6. Chao-Wei Ou, Sanjay Ranka, 'Parallel Incremental Graph Partitioning', IEEE Transactions on Parallel and Distributed Systems, 8(8), August 1997 https://doi.org/10.1109/71.605773
  7. S. Zhou, 'A Trace Driven Simulation Study of Dynamic Load Balancing', IEEE Trans. Software Engineering, Vol. 14, No. 9, pp. 1327-1341, Sept 1988 https://doi.org/10.1109/32.6176
  8. O. Kremien, J. Kramer, J. Magee, 'Scalable, Adaptive Load Sharing for Distributed System', IEEE Parallel and Distributed Technology, Vol. 1, No. 3, pp. 62-70, Aug 1993 https://doi.org/10.1109/88.242447
  9. T.S. Hsu, J.C. Lee, D.R. Lopez, 'Task Allocation on a Network of Processors', IEEE Trans. on Computers, Vol. 49, No. 12, Dec 2000 https://doi.org/10.1109/12.895858
  10. Olivier DALLE, 'LoadBuilder:a tool for generating and modeling workloads in distributed workstation environment', Proceeding of 9th ISCA, 1996
  11. Domenico Ferrari, 'A Study of Load Indices for Load Balancing Schemes', Report No. UCB/CSD 86/262, October 1985
  12. Virginia Mary Lo, 'Heuristic Algorithms for Task Assignment in Distributed System', IEEE Trans on Computers, 37(11), November 1988 https://doi.org/10.1109/12.8704
  13. F. Meisgen, E. Speckenmeyer, 'Dynamic Load Balancing on Clusters of Heterogenous Workstations', Dept of Computer Science University of Cologne, Tech Report No. 97-261, 1997
  14. J. C. Fabero, I. Martin, A. Bautista, S. Molina, 'Dynamic Load Balancing in a Heterogeneous Environment under PVM', IEEE Proceedings of PDP'96, 1996 https://doi.org/10.1109/EMPDP.1996.500614
  15. Khaled Al-Saqabi, Steve W. Otto, Jonathan Walpole, 'Gang Scheduling in Heterogeneous Distributed System', Oregon Graduate Institute, Tech Report, 1994