Show simple item record

dc.contributor.authorDutta, Bishwajiten
dc.contributor.authorAdhinarayanan, Vigneshen
dc.contributor.authorFeng, Wu-chunen
dc.date.accessioned2018-02-02T21:25:02Zen
dc.date.available2018-02-02T21:25:02Zen
dc.date.issued2018-02-02en
dc.identifier.urihttp://hdl.handle.net/10919/81997en
dc.description.abstractA software-based approach to achieve high performance within a power budget often involves dynamic voltage and frequency scaling (DVFS). Consequently, accurately predicting the power consumption of an application at different DVFS levels (or more generally, different processor configurations) is paramount for the energy-efficient functioning of a high-performance computing (HPC) system. The increasing prevalence of graphics processing units (GPUs) in HPC systems presents new multi-dimensional challenges in power management, and machine learning presents an unique opportunity to improve the software-based power management of these HPC systems. As such, we explore the problem of predicting the power consumption of a GPU at different DVFS states via machine learning. Specifically, we perform statistically rigorous experiments to quantify eight machine-learning techniques (i.e., ZeroR, simple linear regression (SLR), KNN, bagging, random forest, sequential minimal optimization regression (SMOreg), decision tree, and neural networks) to predict GPU power consumption at different frequencies. Based on these results, we propose a hybrid ensemble technique that incorporates SMOreg, SLR, and decision tree, which, in turn, reduces the mean absolute error (MAE) to 3.5%.en
dc.language.isoen_USen
dc.publisherDepartment of Computer Science, Virginia Polytechnic Institute & State Universityen
dc.relation.ispartofComputer Science Technical Reportsen
dc.rightsIn Copyrighten
dc.rights.urihttp://rightsstatements.org/vocab/InC/1.0/en
dc.subjectHigh Performance Computingen
dc.subjectParallel and Distributed Computingen
dc.subjectData Mining / Machine Learningen
dc.subjectAlgorithmsen
dc.subjectComputational Science and Engineeringen
dc.subjectModeling and Simulationen
dc.titleGPU Power Prediction via Ensemble Machine Learning for DVFS Space Explorationen
dc.typeTechnical reporten
dc.contributor.departmentComputer Scienceen
dc.identifier.trnumberTR-18-01en
dc.type.dcmitypeTexten


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record