We propose a fault tolerance approach for sparse matrix operations that detects and implicitly locates errors in the results for efficient local correction. This approach reduces the runtime overhead for fault tolerance and provides high error coverage. Existing algorithm-based fault tolerance approaches for sparse matrix operations detect and correct errors, but they often rely on expensive error localization steps. General checkpointing schemes can induce large recovery cost for high error rates. For sparse matrix-vector multiplications, experimental results show an average reduction in runtime overhead of 43.8%, while the error coverage is on average improved by 52.2% compared to related work. The practical applicability is demonstrated in a case study using the iterative Preconditioned Conjugate Gradient solver. When scaling the error rate by four orders of magnitude, the average runtime overhead increases only by 31.3% compared to low error rates.
%0 Conference Paper
%1 SchoeBKW2016
%A Schöll, Alexander
%A Braun, Claus
%A Kochte, Michael A.
%A Wunderlich, Hans-Joachim
%B Proceedings of the 46th Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN'16)
%D 2016
%K ABFT SimTech algebra error-detection fault-tolerance linear localization myown online sparse
%P 251--262
%R http://dx.doi.org/10.1109/DSN.2016.31
%T Efficient Algorithm-Based Fault Tolerance for Sparse Matrix Operations
%X We propose a fault tolerance approach for sparse matrix operations that detects and implicitly locates errors in the results for efficient local correction. This approach reduces the runtime overhead for fault tolerance and provides high error coverage. Existing algorithm-based fault tolerance approaches for sparse matrix operations detect and correct errors, but they often rely on expensive error localization steps. General checkpointing schemes can induce large recovery cost for high error rates. For sparse matrix-vector multiplications, experimental results show an average reduction in runtime overhead of 43.8%, while the error coverage is on average improved by 52.2% compared to related work. The practical applicability is demonstrated in a case study using the iterative Preconditioned Conjugate Gradient solver. When scaling the error rate by four orders of magnitude, the average runtime overhead increases only by 31.3% compared to low error rates.
@inproceedings{SchoeBKW2016,
abstract = {We propose a fault tolerance approach for sparse matrix operations that detects and implicitly locates errors in the results for efficient local correction. This approach reduces the runtime overhead for fault tolerance and provides high error coverage. Existing algorithm-based fault tolerance approaches for sparse matrix operations detect and correct errors, but they often rely on expensive error localization steps. General checkpointing schemes can induce large recovery cost for high error rates. For sparse matrix-vector multiplications, experimental results show an average reduction in runtime overhead of 43.8%, while the error coverage is on average improved by 52.2% compared to related work. The practical applicability is demonstrated in a case study using the iterative Preconditioned Conjugate Gradient solver. When scaling the error rate by four orders of magnitude, the average runtime overhead increases only by 31.3% compared to low error rates.},
added-at = {2018-03-19T16:15:07.000+0100},
author = {Schöll, Alexander and Braun, Claus and Kochte, Michael A. and Wunderlich, Hans-Joachim},
biburl = {https://puma.ub.uni-stuttgart.de/bibtex/2973313fa8f49996637b903eadac4fc19/clausbraun},
booktitle = {Proceedings of the 46th Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN'16)},
doi = {http://dx.doi.org/10.1109/DSN.2016.31},
file = {http://www.iti.uni-stuttgart.de/fileadmin/rami/files/publications/2016/DSN_SchoeBKW2016.pdf},
interhash = {3e8ca3c5bdad028a00edd6b200f98b53},
intrahash = {973313fa8f49996637b903eadac4fc19},
keywords = {ABFT SimTech algebra error-detection fault-tolerance linear localization myown online sparse},
pages = {251--262},
timestamp = {2018-03-19T15:26:03.000+0100},
title = {{Efficient Algorithm-Based Fault Tolerance for Sparse Matrix Operations}},
year = 2016
}
