84 citations · 124 across the 4 of their papers we have counts for
4 papers
Tolerating Silent Data Corruption in Opaque Preconditioners
James Elliott, Mark Hoemmen, Frank Mueller
We demonstrate algorithm-based fault tolerance for silent, transient data corruption in "black-box" preconditioners. We consider both additive Schwarz domain decomposition with an…
Resilience in Numerical Methods: A Position on Fault Models and Methodologies
James Elliott, Mark Hoemmen, Frank Mueller
Future extreme-scale computer systems may expose silent data corruption (SDC) to applications, in order to save energy or increase performance. However, resilience research struggl…
Exploiting Data Representation for Fault Tolerance
James Elliott, Mark Hoemmen, Frank Mueller
We explore the link between data representation and soft errors in dot products. We present an analytic model for the absolute error introduced should a soft error corrupt a bit in…
Evaluating the Impact of SDC on the GMRES Iterative Solver
James Elliott, Mark Hoemmen, Frank Mueller
Increasing parallelism and transistor density, along with increasingly tighter energy and peak power constraints, may force exposure of occasionally incorrect computation or storag…