000860393 001__ 860393 000860393 005__ 20200914103555.0 000860393 0247_ $$2doi$$a10.1016/S0927-5452(04)80071-6 000860393 0247_ $$2ISSN$$a0927-5452 000860393 0247_ $$2ISSN$$a1879-808X 000860393 037__ $$aFZJ-2019-01168 000860393 082__ $$a510 000860393 1001_ $$0P:(DE-Juel1)132090$$aEicker, N.$$b0$$ufzj 000860393 1112_ $$aParallel Computing 2003$$cDresden$$d2003-09-02 - 2003-09-05$$gParCo 2003$$wGermany 000860393 245__ $$aFast parallel I/O on parastation clusters 000860393 260__ $$aAmsterdam [u.a.]$$bElsevier$$c2004 000860393 29510 $$aParallel Computing - Software Technology, Algorithms, Architectures and Applications / Eicker, N. ; : Elsevier, 2004, ; ISSN: 09275452 ; ISBN: 9780444516893 ; doi:10.1016/S0927-5452(04)80071-6 000860393 300__ $$a559 - 567 000860393 3367_ $$2ORCID$$aCONFERENCE_PAPER 000860393 3367_ $$033$$2EndNote$$aConference Paper 000860393 3367_ $$2BibTeX$$aINPROCEEDINGS 000860393 3367_ $$2DRIVER$$aconferenceObject 000860393 3367_ $$2DataCite$$aOutput Types/Conference Paper 000860393 3367_ $$0PUB:(DE-HGF)8$$2PUB:(DE-HGF)$$aContribution to a conference proceedings$$bcontrib$$mcontrib$$s1600072536_28885 000860393 3367_ $$0PUB:(DE-HGF)7$$2PUB:(DE-HGF)$$aContribution to a book$$mcontb 000860393 4900_ $$aAdvances in Parallel Computing$$v13 000860393 520__ $$aWe have demonstrated that the ParaStation3 communication system speeds up the performance of parallel I/O on cluster computers such as ALiCE. I/O benchmarks with PVFS using Parastation over Myrinet achieve a throughput for write operations of up to 1 GB/s from a 32-processor compute partition, given a 32-processor PVFS I/O partition. These results outperform known benchmark results for PVFS on 1.28 Gbit Myrinet by more than a factor of 2, a fact that is mainly due to the superior communication features of ParaStation. Read performance from buffer cache reaches up to 2.2 GB/s, while reading from hard disk saturates at the cumulative hard disk performance. The I/O performance achieved with PVFS using ParaStation enables us to carry out extremely data-intensive eigenmode computations on ALiCE in the application field of lattice quantum chromodynamics. For the future, we plan to utilize the I/O system for storing and processing mass data in high energy physics data analysis on clusters. 000860393 588__ $$aDataset connected to CrossRef Book Series 000860393 7001_ $$0P:(DE-HGF)0$$aIsaila, F.$$b1 000860393 7001_ $$0P:(DE-Juel1)132179$$aLippert, T.$$b2$$ufzj 000860393 7001_ $$0P:(DE-HGF)0$$aMoschny, T.$$b3 000860393 7001_ $$0P:(DE-HGF)0$$aTichy, W. F.$$b4 000860393 773__ $$a10.1016/S0927-5452(04)80071-6 000860393 909CO $$ooai:juser.fz-juelich.de:860393$$pextern4vita 000860393 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)132090$$aForschungszentrum Jülich$$b0$$kFZJ 000860393 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)132179$$aForschungszentrum Jülich$$b2$$kFZJ 000860393 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS 000860393 980__ $$acontrib 000860393 980__ $$aEDITORS 000860393 980__ $$acontb 000860393 980__ $$aI:(DE-Juel1)JSC-20090406 000860393 980__ $$aI:(DE-Juel1)NIC-20090406 000860393 9801_ $$aEXTERN4VITA