Skip to content Skip to main navigation Report an accessibility issue

EECS Publication

Algorithmic Based Fault Tolerance Applied to High Performance Computing

George Bosilca, Remi Delmas, Jack Dongarra, and Julien Langou

We present a new approach to fault tolerance for High Performance Computing system. Our approach is based on a careful adaptation of the Algorithmic Based Fault Tolerance technique (Huang and Abraham, 1984) to the need of parallel distributed computation. We obtain a strongly scalable mechanism for fault tolerance. We can also detect and correct errors (bit-flip) on the fly of a computation. To assess the viability of our approach, we have developed a fault tolerant matrixmatrix multiplication subroutine and we propose some models to predict its running time. Our parallel fault-tolerant matrix-matrix multiplication scores 1.4 TFLOPS on 484 processors (cluster jacquard.nersc.gov) and returns a correct result while one process failure has happened. This represents 65% of the machine peak efficiency and less than 12% overhead with respect to the fastest failure-free implementation. We predict (and have observed) that, as we increase the processor count, the overhead of the fault tolerance drops significantly.

Published  2008-06-19 04:00:00  as  ut-cs-08-620 (ID:95)

ut-cs-08-620.pdf

« Back to Listing