论文信息 - Research of Process Migration Mechanism Based on Checkpoint in Computational Grid

Research of Process Migration Mechanism Based on Checkpoint in Computational Grid

Checkpoint mechanism is very important for long-running and large-scale scientific project computing tasks. So it just meet the requirements of fault-tolerant in grid system. This paper imports GridCPR into Grid Environment. When a job was failed, it can be started from the latest checkpoint instead of the begin to provide with the function of fault tolerance in grid platform and enhance the efficiency of computing tasks.

Zhen Li | Wenxing Wang | Zhuguo Li | Wenxing Wang

[1] Dejan S. Milojicic,et al. Experiences with load distribution on top of the mach microkernel , 1993 .

[2] George G. Robertson,et al. Accent: A communication oriented network operating system kernel , 1981, SOSP.

[3] Miron Livny,et al. Checkpoint and Migration of UNIX Processes in the Condor Distributed Processing System , 1997 .

[4] Miron Livny,et al. Managing Checkpoints for Parallel Programs , 1996, JSSPP.

[5] Rajkumar Buyya,et al. High Performance Cluster Computing: Architectures and Systems , 1999 .

[6] Hua Zhong,et al. CRAK: Linux Checkpoint/Restart As a Kernel Module , 1996 .

[7] Peter M. A. Sloot,et al. The implementation of dynamite: an environment for migrating PVM tasks , 2000, OPSR.

[8] Li Xiao,et al. Improving distributed workload performance by sharing both CPU and memory resources , 2000, Proceedings 20th IEEE International Conference on Distributed Computing Systems.