Tuning Remote GPU Virtualization for InfiniBand Networks

Carlos Reaño, Federico Silla

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

In the past few years, a tendency towards using InfiniBand networks to interconnect high performance computing clusters can be observed. Thus, most of the supercomputers appearing in the TOP500 list either use Ethernet or InfiniBand interconnects. Regarding the latter, the complexity of the InfiniBand programming API (i.e., InfiniBand Verbs) makes it difficult for applications to get the maximum performance of these networks. In this paper we expose how we have tuned a remote GPU virtualization framework whose communications module is implemented using InfiniBand Verbs. The net result is a noticeable increase in the performance of this framework, significantly reducing the gap between remote and local GPUs.
Original languageEnglish
Pages (from-to)4520-4545
Number of pages26
JournalThe Journal of Supercomputing
Volume72
Early online date23 May 2016
DOIs
Publication statusPublished - Dec 2016

Fingerprint Dive into the research topics of 'Tuning Remote GPU Virtualization for InfiniBand Networks'. Together they form a unique fingerprint.

  • Cite this