Show simple item record

dc.contributor.authorLi, Boen_US
dc.date.accessioned2019-01-29T09:00:21Z
dc.date.available2019-01-29T09:00:21Z
dc.date.issued2019-01-28
dc.identifier.othervt_gsexam:17717en_US
dc.identifier.urihttp://hdl.handle.net/10919/87064
dc.description.abstractEmergent systems in high-performance computing (HPC) expect maximal efficiency to achieve the goal of power budget under 20-40 megawatts for 1 exaflop set by the Department of Energy. To optimize efficiency, emergent systems provide multiple power-performance control techniques to throttle different system components and scale of concurrency. In this dissertation, we focus on three throttling techniques: CPU dynamic voltage and frequency scaling (DVFS), dynamic memory throttling (DMT), and dynamic concurrency throttling (DCT). We first conduct an empirical analysis of the performance and energy trade-offs of different architectures under the throttling techniques. We show the impact on performance and energy consumption on Intel x86 systems with accelerators of Intel Xeon Phi and a Nvidia general-purpose graphics processing unit (GPGPU). We show the trade-offs and potentials for improving efficiency. Furthermore, we propose a parallel performance model for coordinating DVFS, DMT, and DCT simultaneously. We present a multivariate linear regression-based approach to approximate the impact of DVFS, DMT, and DCT on performance for performance prediction. Validation using 19 HPC applications/kernels on two architectures (i.e., Intel x86 and IBM BG/Q) shows up to 7% and 17% prediction error correspondingly. Thereafter, we develop the metrics for capturing the performance impact of DVFS, DMT, and DCT. We apply the artificial neural network model to approximate the nonlinear effects on performance impact and present a runtime control strategy accordingly for power capping. Our validation using 37 HPC applications/kernels shows up to a 20% performance improvement under a given power budget compared with the Intel RAPL-based method.en_US
dc.format.mediumETDen_US
dc.publisherVirginia Techen_US
dc.rightsThis item is protected by copyright and/or related rights. Some uses of this item may be deemed fair and permitted by law even without permission from the rights holder(s), or the rights holder(s) may have licensed the work for use under certain conditions. For other uses you need to obtain permission from the rights holder(s).en_US
dc.subjectParallel Performance Modelingen_US
dc.subjectDynamic Voltage and Frequency Scalingen_US
dc.subjectDynamic Memory Throttlingen_US
dc.subjectDynamic Concurrency Throttlingen_US
dc.subjectShared-Memory Systemsen_US
dc.titleModeling and Runtime Systems for Coordinated Power-Performance Managementen_US
dc.typeDissertationen_US
dc.contributor.departmentComputer Scienceen_US
dc.description.degreePh. D.en_US
thesis.degree.namePh. D.en_US
thesis.degree.leveldoctoralen_US
thesis.degree.grantorVirginia Polytechnic Institute and State Universityen_US
thesis.degree.disciplineComputer Science and Applicationsen_US
dc.contributor.committeechairCameron, Kirk W.en_US
dc.contributor.committeememberButt, Alien_US
dc.contributor.committeememberLee, Dongyoonen_US
dc.contributor.committeememberLeon, Edgar A.en_US
dc.contributor.committeememberBack, Godmar Volkeren_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record