Sustainable GPU computing at scale

Justin Y. Shi, Moussa Taifi, Abdallah Khreishah, Jie Wu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

15 Scopus citations

Abstract

General purpose GPU (GPGPU) computing has produced the fastest running supercomputers in the world. For continued sustainable progress, GPU computing at scale also need to address two open issues: a) how increase applications mean time between failures (MTBF) as we increase supercomputer's component counts, and b) how to minimize unnecessary energy consumption. Since energy consumption is defined by the number of components used, we consider a sustainable high performance computing (HPC) application can allow better performance and reliability at the same time when adding computing or communication components. This paper reports a two-tier semantic statistical multiplexing framework for sustainable HPC at scale. The idea is to leverage the powers of statistic multiplexing to tame the nagging HPC scalability challenges. We include the theoretical model, sustainability analysis and computational experiments with automatic system level multiple CPU/GPU failure containment. Our results show that assuming three times slowdown of the statistical multiplexing layer, for an application using 1024 processors with 35% checkpoint overhead, the two-tier framework will produce sustained time and energy savings for MTBF less than 6 hours. With 5% checkpoint overhead, 1.5 hour MTBF would be the break even point. These results suggest the practical feasibility for the proposed two-tier framework.

Original languageEnglish (US)
Title of host publicationProc. - 14th IEEE Int. Conf. on Computational Science and Engineering, CSE 2011 and 11th Int. Symp.on Pervasive Systems, Algorithms, and Networks, I-SPAN 2011 and 10th IEEE Int. Conf. IUCC 2011
Pages263-272
Number of pages10
DOIs
StatePublished - 2011
Externally publishedYes
Event14th IEEE Int. Conf. on Computational Science and Engineering, CSE 2011, the 11th International Symposium on Pervasive Systems, Algorithms, and Networks, I-SPAN 2011, and the 10th IEEE Int. Conf. on Ubiquitous Computing and Communications, IUCC 2011 - Dalian, Liaoning, China
Duration: Aug 24 2011Aug 26 2011

Publication series

NameProc. - 14th IEEE Int. Conf. on Computational Science and Engineering, CSE 2011 and 11th Int. Symp. on Pervasive Systems, Algorithms, and Networks, I-SPA 2011 and 10th IEEE Int. Conf. on IUCC 2011

Other

Other14th IEEE Int. Conf. on Computational Science and Engineering, CSE 2011, the 11th International Symposium on Pervasive Systems, Algorithms, and Networks, I-SPAN 2011, and the 10th IEEE Int. Conf. on Ubiquitous Computing and Communications, IUCC 2011
Country/TerritoryChina
CityDalian, Liaoning
Period8/24/118/26/11

All Science Journal Classification (ASJC) codes

  • Computational Theory and Mathematics
  • Computer Networks and Communications
  • Computer Science Applications

Keywords

  • Data parallel processing
  • Fault tolerant GPU computing
  • Semantic statistical multiplexing
  • Tuple switching network

Fingerprint

Dive into the research topics of 'Sustainable GPU computing at scale'. Together they form a unique fingerprint.

Cite this