Algorithm-Based Diskless Checkpointing for Fault Tolerant Matrix Operations

This paper is an exploration of diskless checkpointing for distributed scienti c computations. With the widespread use of the \Network Of Workstation" (NOW) platform for distributed computing, long-running scienti c computations need to tolerate the changing and often faulty nature of NOW environments. We present high-performance implementations of several algorithms for distributed scienti c computing, including Cholesky factorization, LU factorization, QR factorization, and preconditioned conjugate gradient. These implementations are able to run on PVM networks of at least N processors, and can complete with low overhead as long as any N processors remain functional. 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, and on the IBM SP2.