Fast Parallel I/O on Cluster Computers

Simple item page
dc.creatorDuessel, Thomas
dc.creatorEicker, Norbert
dc.creatorIsaila, Florin
dc.creatorLippert, Thomas
dc.creatorMoschny, Thomas
dc.creatorNeff, Hartmut
dc.creatorSchilling, Klaus
dc.creatorTichy, Walter
dc.date2003-03-19
dc.date.accessioned2026-07-07T03:19:31Z
dc.date.available2026-07-07T03:19:31Z
dc.descriptionToday's cluster computers suffer from slow I/O, which slows down I/O-intensive applications. We show that fast disk I/O can be achieved by operating a parallel file system over fast networks such as Myrinet or Gigabit Ethernet. In this paper, we demonstrate how the ParaStation3 communication system helps speed-up the performance of parallel I/O on clusters using the open source parallel virtual file system (PVFS) as testbed and production system. We will describe the set-up of PVFS on the Alpha-Linux-Cluster-Engine (ALiCE) located at Wuppertal University, Germany. Benchmarks on ALiCE achieve write-performances of up to 1 GB/s from a 32-processor compute-partition to a 32-processor PVFS I/O-partition, outperforming known benchmark results for PVFS on the same network by more than a factor of 2. Read-performance from buffer-cache reaches up to 2.2 GB/s. Our benchmarks are giant, I/O-intensive eigenmode problems from lattice quantum chromodynamics, demonstrating stability and performance of PVFS over Parastation in large-scale production runs.
dc.description22 pages, 10 figures
dc.identifierhttps://arxiv.org/abs/cs/0303016
dc.identifierhttp://arxiv.org/abs/cs/0303016
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/31493
dc.subjectDistributed, Parallel, and Cluster Computing
dc.subjectHardware Architecture
dc.subjectHigh Energy Physics - Lattice
dc.subjectB.4.3; C.1.2; C.2.2; D.4.3
dc.titleFast Parallel I/O on Cluster Computers
dc.typetext

Files

Collections

arXiv