An Efficient and Experimentally Tuned Software-Based Hardening Strategy for Matrix Multiplication on GPUs
Authors:
- 30 April 2013
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Nuclear Science
- Vol. 60 (4) , 2797-2804
Abstract
Neutron radiation experiment results on matrix multiplication on graphic processing units (GPUs) show that multiple errors are detected at the output in more than 50% of the cases. In the presence of multiple errors, the available hardening strategies may become ineffective or inefficient. Analyzing radiation-induced error distributions, we developed an optimized and experimentally tuned software-based hardening strategy for GPUs. With fault-injection simulations, we compare the performance and correcting capabilities of the proposed technique with the available ones.Keywords
This publication has 11 references indexed in Scilit:
- Neutron-Induced Soft Errors in Graphic Processing Units2012
- The impact of new technology on soft error rates2011
- Soft errors — Past history and recent discoveries2010
- ABC-SysBio—approximate Bayesian computation in Python with GPU supportBioinformatics, 2010
- A New Hardware/Software Platform and a New 1/E Neutron Source for Soft Error Studies: Testing FPGAs at the ISIS FacilityIEEE Transactions on Nuclear Science, 2007
- Defect and error tolerance in the presence of massive numbers of defectsIEEE Design & Test of Computers, 2004
- Linear algebra operators for GPU implementation of numerical algorithmsACM Transactions on Graphics, 2003
- Single event upset at ground levelIEEE Transactions on Nuclear Science, 1996
- Algorithm-Based Fault Tolerance for Matrix OperationsIEEE Transactions on Computers, 1984
- Fast probabilistic algorithms1979