Iterative solvers like the Preconditioned Conjugate Gradient (PCG) method are widely-used in compute-intensive domains including science and engineering that often impose tight accuracy demands on computational results. At the same time, the error resilience of such solvers may change in the course of the iterations, which requires careful adaption of the induced approximation errors to reduce the energy demand while avoiding unacceptable results. A novel adaptive method is presented that enables iterative Preconditioned Conjugate Gradient (PCG) solvers on Approximate Computing hardware with high energy efficiency while still providing correct results. The method controls the underlying precision at runtime using a highly efficient fault tolerance technique that monitors the induced error and the quality of intermediate computational results.
%0 Conference Paper
%1 SchoeBW2017
%A Schöll, Alexander
%A Braun, Claus
%A Wunderlich, Hans-Joachim
%B Proceedings of the 23rd IEEE International Symposium on On-Line Testing and Robust System Design (IOLTS'17)
%D 2017
%K AxC SimTech approximate computing energy-efficiency fault monitoring myown quality tolerance
%P 237--239
%R http://dx.doi.org/10.1109/IOLTS.2017.8046244
%T Energy-efficient and Error-resilient Iterative Solvers for Approximate Computing
%X Iterative solvers like the Preconditioned Conjugate Gradient (PCG) method are widely-used in compute-intensive domains including science and engineering that often impose tight accuracy demands on computational results. At the same time, the error resilience of such solvers may change in the course of the iterations, which requires careful adaption of the induced approximation errors to reduce the energy demand while avoiding unacceptable results. A novel adaptive method is presented that enables iterative Preconditioned Conjugate Gradient (PCG) solvers on Approximate Computing hardware with high energy efficiency while still providing correct results. The method controls the underlying precision at runtime using a highly efficient fault tolerance technique that monitors the induced error and the quality of intermediate computational results.
@inproceedings{SchoeBW2017,
abstract = {Iterative solvers like the Preconditioned Conjugate Gradient (PCG) method are widely-used in compute-intensive domains including science and engineering that often impose tight accuracy demands on computational results. At the same time, the error resilience of such solvers may change in the course of the iterations, which requires careful adaption of the induced approximation errors to reduce the energy demand while avoiding unacceptable results. A novel adaptive method is presented that enables iterative Preconditioned Conjugate Gradient (PCG) solvers on Approximate Computing hardware with high energy efficiency while still providing correct results. The method controls the underlying precision at runtime using a highly efficient fault tolerance technique that monitors the induced error and the quality of intermediate computational results.},
added-at = {2018-03-19T16:15:07.000+0100},
author = {Schöll, Alexander and Braun, Claus and Wunderlich, Hans-Joachim},
biburl = {https://puma.ub.uni-stuttgart.de/bibtex/2bfd0a364cc8901abde747841b8f60a69/clausbraun},
booktitle = {Proceedings of the 23rd IEEE International Symposium on On-Line Testing and Robust System Design (IOLTS'17)},
doi = {http://dx.doi.org/10.1109/IOLTS.2017.8046244},
file = {http://www.iti.uni-stuttgart.de/fileadmin/rami/files/publications/2017/IOLTS_SchoeBW2017.pdf},
interhash = {70d88e8ae6518962c2cc6a2b24f8fbd6},
intrahash = {bfd0a364cc8901abde747841b8f60a69},
keywords = {AxC SimTech approximate computing energy-efficiency fault monitoring myown quality tolerance},
pages = {237--239},
timestamp = {2018-03-19T15:28:37.000+0100},
title = {{Energy-efficient and Error-resilient Iterative Solvers for Approximate Computing}},
year = 2017
}
