Understanding GPU errors on large-scale HPC systems and the implications for system design and operation

Devesh Tiwari, Saurabh Gupta, James Rogers, Don Maxwell, Paolo Rech, Sudharshan Vazhkudai, Daniel Oliveira, Dave Londo, Nathan Debardeleben, Philippe Navaux, Luigi Carro, Arthur Bland

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

140 Scopus citations

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.

Original languageEnglish
Title of host publication2015 IEEE 21st International Symposium on High Performance Computer Architecture, HPCA 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages331-342
Number of pages12
ISBN (Electronic)9781479989300
DOIs
StatePublished - Mar 6 2015
Event2015 21st IEEE International Symposium on High Performance Computer Architecture, HPCA 2015 - Burlingame, United States
Duration: Feb 7 2015Feb 11 2015

Publication series

Name2015 IEEE 21st International Symposium on High Performance Computer Architecture, HPCA 2015

Conference

Conference2015 21st IEEE International Symposium on High Performance Computer Architecture, HPCA 2015
Country/TerritoryUnited States
CityBurlingame
Period02/7/1502/11/15

Fingerprint

Dive into the research topics of 'Understanding GPU errors on large-scale HPC systems and the implications for system design and operation'. Together they form a unique fingerprint.

Cite this