Fault-Tolerant Iterative Methods via Selective Reliability

@inproceedings{Hoemmen2011FaultTolerantIM,
  title={Fault-Tolerant Iterative Methods via Selective Reliability},
  author={Mark Hoemmen},
  year={2011}
}
Current iterative methods for solving linear equations assume reliability of data (no “bit flips”) and arithmetic (correct up to rounding error). If faults occur, the solver usually either aborts, or computes the wrong answer without indication. System reliability guarantees consume energy or reduces performance. As processor counts continue to grow, these costs will become unbearable. Instead, we show that if the system lets applications apply reliability selectively, we can develop iterations… CONTINUE READING
Highly Influential
This paper has highly influenced 14 other papers. REVIEW HIGHLY INFLUENTIAL CITATIONS
Highly Cited
This paper has 136 citations. REVIEW CITATIONS

Citations

Publications citing this paper.
Showing 1-10 of 71 extracted citations

136 Citations

02040'12'14'16'18
Citations per Year
Semantic Scholar estimates that this publication has 136 citations based on the available data.

See our FAQ for additional information.

References

Publications referenced by this paper.
Showing 1-10 of 34 references

Updating a rank-revealing ULV decomposition

  • G. W. Stewart
  • SIAM J. Matrix Anal. Appl., 14(2):494–499, April
  • 1993
Highly Influential
3 Excerpts

Soft errors in linear solvers as integrated components of a simulation

  • V. E. Howle
  • Presented at the Copper Mountain Conference on…
  • 2010
2 Excerpts

Similar Papers

Loading similar papers…