HPC interconnection networks: The key to exascale computing

Jeffrey S. Vetter, Vinod Tipparaju, Weikuan Yu, Philip C. Roth

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

6 Scopus citations

Abstract

Scientists from many domains desire to address problems within the next decade that, by all estimates, require computer systems that can achieve sustained exaflop computing rates (i.e., 1×1018 floating point operations per second) with real-world applications. Simply scaling existing designs is insufficient: analysis of current technological trends suggests that only a few architectural components are on track to reach the performance levels needed for exascale computing. The network connecting computer system nodes presents a particularly difficult challenge because of the prevalence of a wide variety of communication patterns and collective communication operations in algorithms used in scientific applications and their tendency to be the most significant limit to application scalability. Researchers at Oak Ridge National Laboratory and elsewhere are actively working to overcome these network-related scalability barriers using advanced hardware and software design, alternative network topologies, and performance prediction using modeling and simulation.

Original languageEnglish
Title of host publicationHigh Speed and Large Scale Scientific Computing
PublisherIOS Press BV
Pages95-106
Number of pages12
ISBN (Print)9781607500735
DOIs
StatePublished - 2009

Publication series

NameAdvances in Parallel Computing
Volume18
ISSN (Print)0927-5452

Keywords

  • Exascale
  • high-performance computing
  • interconnection networks

Fingerprint

Dive into the research topics of 'HPC interconnection networks: The key to exascale computing'. Together they form a unique fingerprint.

Cite this