Logo des Repositoriums
 

Modelling and Predicting Memory Behaviour in Parallel Systems with Network Links? Palladio-based Experiment Report

Kurzanzeige
dc.contributor.authorGruber, Philipp
dc.contributor.authorFrank, Markus
dc.contributor.editorKelter, Udo
dc.date.accessioned2023-02-27T13:59:29Z
dc.date.available2023-02-27T13:59:29Z
dc.date.issued2019
dc.description.abstractThis work improves the capabilities of Palladio to predict the performance of parallel software in multicore environments. In previous work, we could show that the accuracy of the Palladio simulations is not sufficient for multicore systems. We assume that one reason for this is the memory bandwidth behaviour, which is not included in the Palladio Component Model and can become a bottleneck in parallel software. We present an approach to model the memory bandwidth behaviour by the means of an already existing network link concept. We can show that by using network link as a memory model we can improve our predictions up to 26% points using 16 cores on a machine and can receive an accuracy of 90% for our use case.en
dc.identifier.pissn0720-8928
dc.identifier.urihttps://dl.gi.de/handle/20.500.12116/40493
dc.language.isoen
dc.publisherGesellschaft für Informatik e.V.
dc.relation.ispartofSoftwaretechnik-Trends Band 39, Heft 4
dc.relation.ispartofseriesSoftwaretechnik-Trends
dc.subjectperformance
dc.subjectparallel software
dc.subjectbandwidth
dc.subjectmemory model
dc.subjectprediction
dc.titleModelling and Predicting Memory Behaviour in Parallel Systems with Network Links? Palladio-based Experiment Reporten
dc.typeText/Conference Paper
gi.citation.endPage19
gi.citation.publisherPlaceBonn
gi.citation.startPage17
gi.conference.date5.-6. November 2019
gi.conference.locationWürzburg
gi.conference.sessiontitle10th Symposium on Software Performance (SSP)

Dateien

Originalbündel
1 - 1 von 1
Vorschaubild
Name:
SSP2019_Gruber.pdf
Größe:
223.05 KB
Format:
Adobe Portable Document Format
Herunterladen

Sammlungen

Softwaretechnik-Trends 39(4) - 2019