Transparent adaptive library-based checkpointing for master-worker style parallelism

Permanent URL: http://hdl.handle.net/2047/d20000618
Title:
Transparent adaptive library-based checkpointing for master-worker style parallelism
Creator:
Cooperman, Gene D. (Author)
Ansel, Jason (Author)
Ma, Xiaoqin (Author)
Copyright date:
2006
Type of resource:
Text
Genre:
Articles
Format:
electronic
Digital origin:
born digital
Abstract/Description:
We present a transparent, system-level checkpointing solution for master-worker parallelism that automatically adapts, upon restart, to the number of processor nodes available. This is important, since nodes in a cluster fail. It also allows one to adapt to using multiple cluster partitions and multiple resources from the Computational Grid, as they become available. Checkpointing a master-worker computation has the additional advantage of needing to checkpoint only the master process. This is both fast and more economical of disk space. This has been demonstrated by checkpointing Geant4, a million line C++ program. Our solution has been implemented in the context of TOP-C (Task Oriented Parallel C/C++), a free, open-source parallel package, although it can easily be ported to additional master-worker packages.
Comments:
Originally published in Cluster Computing and the Grid, 2006. CCGRID 06. Sixth IEEE International Symposium on, vol.1, 9 pp., 16-19 May 2006. doi: 10.1109/CCGRID.2006.106
Subjects and keywords:
Fault-tolerant computing
checkpointing
cluster computing
Geant4
C++
TOP-C
Computer Sciences
Permanent Link:
http://hdl.handle.net/2047/d20000618

Downloads