A novel multi-level integrated roofline model approach for performance characterization

dc.contributor.authorTuomas Koskela
dc.contributor.authorZakhar Matveev
dc.contributor.authorCharlene Yang
dc.contributor.authorAdetokunbo Adedoyin
dc.contributor.authorRoman Belenov
dc.contributor.authorPhilippe Thierry
dc.contributor.authorZhengji Zhao
dc.contributor.authorRahulkumar Gayatri
dc.contributor.authorHongzhang Shan
dc.contributor.authorLeonid Oliker
dc.contributor.authorJack Deslippe
dc.contributor.authorRon Green
dc.contributor.authorSamuel Williams
dc.contributor.organizationfi=avaruustutkimuslaboratorio|en=Space Research Laboratory|
dc.contributor.organization-code2606702
dc.converis.publication-id32081756
dc.converis.urlhttps://research.utu.fi/converis/portal/Publication/32081756
dc.date.accessioned2022-10-27T11:58:59Z
dc.date.available2022-10-27T11:58:59Z
dc.description.abstractWith energy-efficient architectures, including accelerators and many-core processors, gaining traction, application developers face the challenge of optimizing their applications for multiple hardware features including many-core parallelism, wide processing vector-units and on-chip high-bandwidth memory. In this paper, we discuss the development and utilization of a new application performance tool based on an extension of the classical roofline-model for simultaneously profiling multiple levels in the cache-memory hierarchy. This tool presents a powerful visual aid for the developer and can be used to frame the many-dimensional optimization problem in a tractable way. We show case studies of real scientific applications that have gained insights from the Integrated Roofline Model.<br />
dc.format.pagerange226
dc.format.pagerange245
dc.identifier.eisbn978-3-319-92040-5
dc.identifier.isbn978-3-319-92039-9
dc.identifier.issn0302-9743
dc.identifier.jour-issn0302-9743
dc.identifier.olddbid173288
dc.identifier.oldhandle10024/156382
dc.identifier.urihttps://www.utupub.fi/handle/11111/31313
dc.identifier.urnURN:NBN:fi-fe2021042719358
dc.language.isoen
dc.okm.affiliatedauthorKoskela, Tuomas
dc.okm.discipline113 Computer and information sciencesen_GB
dc.okm.discipline113 Tietojenkäsittely ja informaatiotieteetfi_FI
dc.okm.internationalcopublicationinternational co-publication
dc.okm.internationalityInternational publication
dc.okm.typeA4 Conference Article
dc.publisher.countryUnited Statesen_GB
dc.publisher.countryYhdysvallat (USA)fi_FI
dc.publisher.country-codeUS
dc.relation.conferenceInternational Conference on High Performance Computing
dc.relation.doi10.1007/978-3-319-92040-5_12
dc.relation.ispartofjournalLecture Notes in Computer Science
dc.relation.ispartofseriesLecture Notes in Computer Science
dc.relation.volume10876
dc.source.identifierhttps://www.utupub.fi/handle/10024/156382
dc.titleA novel multi-level integrated roofline model approach for performance characterization
dc.title.bookHigh Performance Computing
dc.year.issued2018

Tiedostot

Näytetään 1 - 1 / 1
Ladataan...
Name:
multi-level-integrated (2).pdf
Size:
988.27 KB
Format:
Adobe Portable Document Format
Description:
Final draft