Show simple item record

dc.contributor.authorLiu, Qixiao
dc.contributor.authorMoreto Planas, Miquel
dc.contributor.authorAbella Ferrer, Jaume
dc.contributor.authorCazorla Almeida, Francisco Javier
dc.contributor.authorValero Cortés, Mateo
dc.contributor.otherUniversitat Politècnica de Catalunya. Departament d'Arquitectura de Computadors
dc.date.accessioned2014-10-15T07:24:39Z
dc.date.created2014
dc.date.issued2014
dc.identifier.citationLiu, Q. [et al.]. DReAM: Per-task DRAM energy metering in multicore systems. A: International Conference on Parallel and Distributed Computing. "Euro-Par 2014: Parallel Processing 20th International Conference: Porto, Portugal, August 25-29, 2014: proceedings". Porto: Springer, 2014, p. 111-123.
dc.identifier.isbn978-3-319-09872-2
dc.identifier.urihttp://hdl.handle.net/2117/24369
dc.description.abstractInteraction across applications in DRAM memory impacts its energy consumption. This paper makes the case for accurate per-task DRAM energy metering in multicores, which opens new paths to energy/performance optimizations, such as per-task energy-aware task scheduling and energy-aware billing in datacenters. In particular, the contributions of this paper are (i) an ideal per-task energy metering model for DRAM memories; (ii) DReAM, an accurate, yet low cost, implementation of the ideal model (less than 5% accuracy error when 16 tasks share memory); and (iii) a comparison with standard methods (even distribution and access-count based) proving that DReAM is more accurate than these other methods.
dc.description.sponsorshipThis work has been partially supported by the Spanish Ministry of Science and Innovation under grant TIN2012-34557, the HiPEAC Network of Excellence, by the European Research Council under the European Union’s 7th FP, ERC Grant Agreement n. 321253, and by a joint study agreement between IBM and BSC (number W1361154). Qixiao Liu has also been funded by the Chinese Scholarship Council under grant 2010608015.
dc.format.extent13 p.
dc.language.isoeng
dc.publisherSpringer
dc.subjectÀrees temàtiques de la UPC::Informàtica::Arquitectura de computadors
dc.subject.lcshMultiprocessors
dc.subject.otherDRAM chips
dc.subject.otherEnergy consumption
dc.subject.otherMultiprocessing systems
dc.subject.otherPower aware computing
dc.titleDReAM: Per-task DRAM energy metering in multicore systems
dc.typeConference report
dc.subject.lemacMultiprocessadors
dc.contributor.groupUniversitat Politècnica de Catalunya. CAP - Grup de Computació d'Altes Prestacions
dc.identifier.doi10.1007/978-3-319-09873-9-10
dc.relation.publisherversionhttp://link.springer.com/chapter/10.1007%2F978-3-319-09873-9_10
dc.rights.accessOpen Access
drac.iddocument15228632
dc.description.versionPostprint (published version)
dc.relation.projectidinfo:eu-repo/grantAgreement/EC/FP7/321253/EU/Riding on Moore's Law/ROMOL
dc.date.lift10000-01-01
upcommons.citation.authorLiu, Q.; Moretó, M.; Abella, J.; Cazorla, F.; Valero, M.
upcommons.citation.contributorInternational Conference on Parallel and Distributed Computing
upcommons.citation.pubplacePorto
upcommons.citation.publishedtrue
upcommons.citation.publicationNameEuro-Par 2014: Parallel Processing 20th International Conference: Porto, Portugal, August 25-29, 2014: proceedings
upcommons.citation.startingPage111
upcommons.citation.endingPage123


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record

All rights reserved. This work is protected by the corresponding intellectual and industrial property rights. Without prejudice to any existing legal exemptions, reproduction, distribution, public communication or transformation of this work are prohibited without permission of the copyright holder