Understanding GPU errors on large-scale HPC systems and the implications for system design and operation
Top Cited Papers
- 1 February 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15300897,p. 331-342
Abstract
Increase in graphics hardware performance and improvements in programmability has enabled GPUs to evolve from a graphics-specific accelerator to a general-purpose computing device. Titan, the world's second fastest supercomputer for open science in 2014, consists of more dum 18,000 GPUs that scientists from various domains such as astrophysics, fusion, climate, and combustion use routinely to run large-scale simulations. Unfortunately, while the performance efficiency of GPUs is well understood, their resilience characteristics in a large-scale computing system have not been fully evaluated. We present a detailed study to provide a thorough understanding of GPU errors on a large-scale GPU-enabled system. Our data was collected from the Titan supercomputer at the Oak Ridge Leadership Computing Facility and a GPU cluster at the Los Alamos National Laboratory. We also present results from our extensive neutron-beam tests, conducted at Los Alamos Neutron Science Center (LANSCE) and at ISIS (Rutherford Appleron Laboratories, UK), to measure the resilience of different generations of GPUs. We present several findings from our field data and neutron-beam experiments, and discuss the implications of our results for future GPU architects, current and future HPC computing facilities, and researchers focusing on GPU resilience.Keywords
This publication has 30 references indexed in Scilit:
- Addressing failures in exascale computingThe International Journal of High Performance Computing Applications, 2014
- On the evaluation of soft-errors detection techniques for GPGPUs2013
- Feng shui of supercomputer memory2013
- Experimental evaluation of thread distribution effects on multiple output errors in GPUs2013
- An Efficient and Experimentally Tuned Software-Based Hardening Strategy for Matrix Multiplication on GPUsIEEE Transactions on Nuclear Science, 2013
- Neutron radiation test of graphic processing units2012
- Matrix Multiplication on GPUs with On-Line Fault Tolerance2011
- DRAM errors in the wild2009
- A Large-Scale Study of Failures in High-Performance Computing SystemsIEEE Transactions on Dependable and Secure Computing, 2009
- Failure data analysis of a large-scale heterogeneous server environment2004