Tuning remote GPU virtualization for InfiniBand networks

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 InfiniBa...

Full description

Bibliographic Details
Authors: Reaño González, Carlos, Silla, Federico|||0000-0002-6435-1200
Format: article
Publication Date:2016
Country:España
Institution:Universitat Politècnica de València (UPV)
Repository:RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Language:English
OAI Identifier:oai:riunet.upv.es:10251/83128
Online Access:https://riunet.upv.es/handle/10251/83128
Access Level:Open access
Keyword:HPC
InfiniBand
CUDA
Remote GPU virtualization
Networks
Performance
Tuning
ARQUITECTURA Y TECNOLOGIA DE COMPUTADORES
Description
Summary: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.