Abstract
Networks of workstations (NOWs) offer a cost-effective platform for high-performance, long-running parallel computations. However, these computations must be able to tolerate the changing and often faulty nature of NOW environments. We present high-performance implementations of several fault-tolerant algorithms for distributed scientific computing. The fault-tolerance is based on diskless checkpointing, a paradigm that uses processor redundancy rather than stable storage as the fault-tolerant medium. These algorithms are able to run on clusters of workstations that change over time due to failure, load, or availability. As long as there are at leastnprocessors in the cluster, and failures occur singly, the computation will complete in an efficient manner. We discuss the details of how the algorithms are tuned for fault-tolerance and present the performance results on a PVM network of Sun workstations connected by a fast, switched ethernet.
Original language | English |
---|---|
Pages (from-to) | 125-138 |
Number of pages | 14 |
Journal | Journal of Parallel and Distributed Computing |
Volume | 43 |
Issue number | 2 |
DOIs | |
State | Published - Jun 15 1997 |
Funding
James Plank was supported by National Science Foundation Grant CCR-9409496 and the ORAU Junior Faculty Enhancement Award. Jack Dongarra is supported by the Defense Advanced Research Projects Agency under Contract DAAL03-91-C-0047, administered by the Army Research Office, by the Office of Scientific Computing, U.S. Department of Energy, under Contract DE-AC05-84OR21400, and by the National Science Foundation Science and Technology Center Cooperative Agreement CCR-8809615.