Publication:
Techniques for Autotuning Algorithms on Heterogenous Platforms

Simple item page
dc.affiliation.dptoUC3M. Departamento de Informáticaes
dc.affiliation.grupoinvUC3M. Grupo de Investigación: Arquitectura de Computadores, Comunicaciones y Sistemases
dc.contributor.authorDiéguez, Adrián P.
dc.contributor.authorAmor, Margarita
dc.contributor.authorDoallo, Ramón
dc.contributor.editorCarretero Pérez, Jesús
dc.contributor.editorGarcía Blas, Javier
dc.contributor.editorPetcu, Dana
dc.date.accessioned2016-04-29T08:26:32Z
dc.date.available2016-04-29T08:26:32Z
dc.date.issued2016-02
dc.descriptionProceedings of the First PhD Symposium on Sustainable Ultrascale Computing Systems (NESUS PhD 2016) Timisoara, Romania. February 8-11, 2016.en
dc.description.abstractCurrent GPUs (Graphic Processing Units) can obtain high computational performance in scientific applications. Nevertheless, programmers have to use suitable parallel algorithms for these architectures and have to consider optimization techniques in the implementation in order to achieve that performance. This thesis is focused on designing and implementing parallel prefix algorithms into GPU architectures with little effort. For that, we have developed a very optimized library called BPLG (Tuning Butterfly Processing Library for GPUs) and based on a set of building blocks that enable to easily design well-known algorithms such as FFT, tridiagonal systems solvers, scan operator, sorting or signal processing. This library is designed under a tuning methodology based on two-stages indentified as GPU resource analysis and operator string manipulation. Specifically, this strategy is focused on a set of parallel prefix algorithms that can be represented according to a set of common permutations of the digits of each of its element indices [4], denoted as Index-Digit (ID) algorithms. So far, the proposed methodology has obtained very good results with respect to state-of-art libraries, as CUFFT, CUSPARSE, CUDPP or ModernGPU.en
dc.description.sponsorshipEuropean Cooperation in Science and Technology. COSTen
dc.format.extent4
dc.format.mimetypeapplication/pdf
dc.identifier.bibliographicCitationCarretero Pérez, Jesús; et.al. (eds.). (2016). Proceedings of the First PhD Symposium on Sustainable UltrascaleComputing Systems (NESUS PhD 2016). Timisoara, Romania. Universidad Carlos III de Madrid, ARCOS. Pp. 25-28.en
dc.identifier.isbn978-84-608-6309-0
dc.identifier.publicationfirstpage25
dc.identifier.publicationlastpage28
dc.identifier.publicationtitleProceedings of the First PhD Symposium on Sustainable UltrascaleComputing Systems (NESUS PhD 2016)en
dc.identifier.urihttps://hdl.handle.net/10016/22888
dc.language.isoengen
dc.relation.eventdateFebruary 8-11, 2016en
dc.relation.eventnumber1
dc.relation.eventplaceTimisoara, Romaniaen
dc.relation.eventtitlePhD Symposium on Sustainable Ultrascale Computing Systems (NESUS PhD 2016)en
dc.rightsAtribución-NoComercial-SinDerivadas 3.0 España
dc.rights.accessRightsopen access
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/3.0/es/
dc.subject.ecienciaInformáticaes
dc.subject.otherCUDAen
dc.subject.otherParallel prefix algorithmsen
dc.subject.otherGPUen
dc.subject.otherID-algorithmsen
dc.subject.otherTuningen
dc.titleTechniques for Autotuning Algorithms on Heterogenous Platformsen
dc.typeconference paper*
dc.type.hasVersionVoR*
dspace.entity.typePublication
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
techniques_perez_nesus_2016.pdf
Size:
288.96 KB
Format:
Adobe Portable Document Format
Download
Collections
First PhD Symposium on Sustainable Ultrascale Computing Systems (NESUS PhD 2016)