Exploring I/O virtualization data paths for MPI applications in a cluster of VMs: A networking perspective

DSpace/Manakin Repository

Show simple item record

dc.contributor.author Nanos, A en
dc.contributor.author Goumas, G en
dc.contributor.author Koziris, N en
dc.date.accessioned 2014-03-01T02:53:16Z
dc.date.available 2014-03-01T02:53:16Z
dc.date.issued 2011 en
dc.identifier.issn 03029743 en
dc.identifier.uri http://hdl.handle.net/123456789/36206
dc.subject.other Data paths en
dc.subject.other Evaluating method en
dc.subject.other High-performance computing en
dc.subject.other MPI applications en
dc.subject.other Network behaviors en
dc.subject.other Network configuration en
dc.subject.other Network devices en
dc.subject.other Network load en
dc.subject.other Network virtualization en
dc.subject.other Optimized solutions en
dc.subject.other Performance bottlenecks en
dc.subject.other Scientific applications en
dc.subject.other Virtualizations en
dc.subject.other Virtualized environment en
dc.subject.other Behavioral research en
dc.subject.other Cluster computing en
dc.subject.other Computer software selection and evaluation en
dc.subject.other Message passing en
dc.subject.other Network performance en
dc.subject.other Optimization en
dc.subject.other Parallel architectures en
dc.subject.other Virtual reality en
dc.subject.other Benchmarking en
dc.title Exploring I/O virtualization data paths for MPI applications in a cluster of VMs: A networking perspective en
heal.type conferenceItem en
heal.identifier.primary 10.1007/978-3-642-21878-1_82 en
heal.identifier.secondary http://dx.doi.org/10.1007/978-3-642-21878-1_82 en
heal.publicationDate 2011 en
heal.abstract Nowadays, seeking optimized data paths that can increase I/O throughput in Virtualized environments is an intriguing task, especially in a high-performance computing context. This study endeavors to address this issue by evaluating methods for optimized network device access using scientific applications and micro-benchmarks. We examine the network performance bottlenecks that appear in a Cluster of Xen VMs using both generic and intelligent network adapters. We study the network behavior of MPI applications. Our goal is to: (a) explore the implications of alternative data paths between applications and network hardware and (b) specify optimized solutions for scientific applications that put pressure on network devices. To monitor the network load and the applications' total throughput we build a custom testbed using different network configurations. We use the Xen bridge mechanism and I/O Virtualization techniques and examine the trade-offs. Preliminary results show that a combination of these techniques is essential to overcome network virtualization overheads and achieve near-native performance. © 2011 Springer-Verlag Berlin Heidelberg. en
heal.journalName Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) en
dc.identifier.doi 10.1007/978-3-642-21878-1_82 en
dc.identifier.volume 6586 LNCS en
dc.identifier.spage 665 en
dc.identifier.epage 671 en

Files in this item

Files Size Format View

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record