Vis enkel innførsel

dc.contributor.advisorJenssen, Robert
dc.contributor.advisorKampffmeyer, Michael C.
dc.contributor.advisorBianchi, Filippo M.
dc.contributor.advisorSalberg, Arnt-Børre
dc.contributor.authorJohansen, Thomas A. Haugland
dc.date.accessioned2017-03-07T07:52:15Z
dc.date.available2017-03-07T07:52:15Z
dc.date.issued2016-12-08
dc.description.abstractThe key objectives in this thesis are; the study of GPU-accelerated eigenvalue decomposition in an effort to uncover both benefits and pitfalls, and then to investigate and facilitate a future GPU implementation of the symmetric QR algorithm with permutations. With the current trend of having ever larger datasets both in terms of features and observations, we propose that GPU computation can help ameliorate the temporal penalties incurred by eigendecomposing large matrices. We successfully show the benefits of performing eigendecomposition on GPUs, and also highlight some problems with current GPU implementations. While implementing the QR algorithm on GPU, we discovered that the GPU-based QR decomposition does not explicitly form the orthogonal matrix needed as part of the QR algorithm. Therefore, we propose a novel GPU algorithm for “implicitly” computing the orthogonal matrix Q from the Householder vectors given by the QR decomposition. To illustrate the benefits of our methods, we show that the kernel entropy component analysis algorithm on GPU is two orders of magnitude faster than an equivalent CPU implementation.en_US
dc.identifier.urihttps://hdl.handle.net/10037/10450
dc.language.isoengen_US
dc.publisherUiT Norges arktiske universiteten_US
dc.publisherUiT The Arctic University of Norwayen_US
dc.rights.accessRightsopenAccessen_US
dc.rights.holderCopyright 2016 The Author(s)
dc.rights.urihttps://creativecommons.org/licenses/by-nc-sa/3.0en_US
dc.rightsAttribution-NonCommercial-ShareAlike 3.0 Unported (CC BY-NC-SA 3.0)en_US
dc.subject.courseIDFYS-3941
dc.subjectVDP::Mathematics and natural science: 400::Mathematics: 410::Statistics: 412en_US
dc.subjectVDP::Matematikk og Naturvitenskap: 400::Matematikk: 410::Statistikk: 412en_US
dc.subjectVDP::Mathematics and natural science: 400::Mathematics: 410::Applied mathematics: 413en_US
dc.subjectVDP::Matematikk og Naturvitenskap: 400::Matematikk: 410::Anvendt matematikk: 413en_US
dc.subjectVDP::Mathematics and natural science: 400::Information and communication science: 420::Mathematical modeling and numerical methods: 427en_US
dc.subjectVDP::Matematikk og Naturvitenskap: 400::Informasjons- og kommunikasjonsvitenskap: 420::Matematisk modellering og numeriske metoder: 427en_US
dc.subjectVDP::Mathematics and natural science: 400::Information and communication science: 420::Algorithms and computability theory: 422en_US
dc.subjectVDP::Matematikk og Naturvitenskap: 400::Informasjons- og kommunikasjonsvitenskap: 420::Algoritmer og beregnbarhetsteori: 422en_US
dc.titleOn the improvement and acceleration of eigenvalue decomposition in spectral methods using GPUsen_US
dc.typeMaster thesisen_US
dc.typeMastergradsoppgaveen_US


Tilhørende fil(er)

Thumbnail
Thumbnail

Denne innførselen finnes i følgende samling(er)

Vis enkel innførsel

Attribution-NonCommercial-ShareAlike 3.0 Unported (CC BY-NC-SA 3.0)
Med mindre det står noe annet, er denne innførselens lisens beskrevet som Attribution-NonCommercial-ShareAlike 3.0 Unported (CC BY-NC-SA 3.0)