Hide metadata

dc.date.accessioned2013-03-12T07:58:59Z
dc.date.available2013-03-12T07:58:59Z
dc.date.issued2010en_US
dc.date.submitted2011-04-13en_US
dc.identifier.citationVrba, Zeljko, , , , , Halvorsen, Pål, , , , , Griwodz, Carsten, , , , , Beskow, Paul, , , , , Espeland, Håvard, , , , , Johansen, Dag, , , , , . The Nornir run-time system for parallel programs using Kahn process networks on multi-core machines A flexible alternative to MapReduce. Journal of Supercomputing. 2010en_US
dc.identifier.urihttp://hdl.handle.net/10852/8849
dc.description.abstractEven though shared-memory concurrency is a paradigm frequently used for developing parallel applications on small- and middle-sized machines, experience has shown that it is hard to use. This is largely caused by synchronization primitives which are low-level, inherently non-deterministic, and, consequently, non-intuitive to use. In this paper, we present the Nornir run-time system. Nornir is comparable to well-known frameworks such as MapReduce and Dryad that are recognized for their efficiency and simplicity. Unlike these frameworks, Nornir also supports process structures containing branches and cycles. Nornir is based on the formalism of Kahn process networks, which is a shared-nothing, message-passing model of concurrency. We deem this model a simple and deterministic alternative to shared-memory concurrency. Experiments with real and synthetic benchmarks on up to 8 CPUs show that performance in most cases scales almost linearly with the number of CPUs, when not limited by data dependencies. We also show that the modeling flexibility allows Nornir to outperform its MapReduce counterparts using well-known benchmarks. This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.eng
dc.language.isoengen_US
dc.titleThe Nornir run-time system for parallel programs using Kahn process networks on multi-core machines A flexible alternative to MapReduceen_US
dc.typeJournal articleen_US
dc.date.updated2011-08-22en_US
dc.creator.authorVrba, Zeljkoen_US
dc.creator.authorHalvorsen, Pålen_US
dc.creator.authorGriwodz, Carstenen_US
dc.creator.authorBeskow, Paulen_US
dc.creator.authorEspeland, Håvarden_US
dc.creator.authorJohansen, Dagen_US
dc.subject.nsiVDP::420en_US
cristin.unitcode150500en_US
cristin.unitnameInformatikken_US
dc.identifier.cristin532486en_US
dc.identifier.bibliographiccitationinfo:ofi/fmt:kev:mtx:ctx&ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.jtitle=Journal of Supercomputing&rft.date=2010en_US
dc.identifier.jtitleJournal of Supercomputing
dc.identifier.volume63
dc.identifier.issue1
dc.identifier.startpage191
dc.identifier.endpage217
dc.identifier.doihttp://dx.doi.org/10.1007/s11227-010-0503-2
dc.identifier.urnURN:NBN:no-28948en_US
dc.type.documentTidsskriftartikkelen_US
dc.identifier.duo115585en_US
dc.type.peerreviewedPeer revieweden_US
dc.identifier.fulltextFulltext https://www.duo.uio.no/bitstream/handle/10852/8849/1/fulltext7.pdf
dc.type.versionPublishedVersion


Files in this item

Appears in the following Collection

Hide metadata