{"id":{"repo_id":"vt","oai_identifier":"oai:vtechworks.lib.vt.edu:10919/87064"},"canonical_url":"https://search.dev.ndltd.org/etd/vt/oai:vtechworks.lib.vt.edu:10919/87064","repository":{"repo_id":"vt","name":"Virginia Tech","base_url":"https://vtechworks.lib.vt.edu/oai/request"},"display":{"title":"Modeling and Runtime Systems for Coordinated Power-Performance Management","abstract":"Emergent 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.","abstract_html":"Emergent 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.","abstract_has_math":false,"creators":["Li, Bo"],"institution":"Virginia Tech","degree_name":"Ph. D.","degree_level":"doctoral","degree_discipline":"Computer Science and Applications","degree_department":"Computer Science","school":null,"contributors":[],"advisors":[],"committee_chairs":["Cameron, Kirk W."],"committee_members":["Butt, Ali R.","Lee, Dongyoon","Leon, Edgar A.","Back, Godmar V."],"year":2019,"date_issued":"2019-01-28","date_published":"2019-01-28","updated_at":"2026-07-22T22:19:29Z","subjects":["Parallel Performance Modeling","Dynamic Voltage and Frequency Scaling","Dynamic Memory Throttling","Dynamic Concurrency Throttling","Shared-Memory Systems"],"languages":[],"rights":["In Copyright"],"rights_urls":["http://rightsstatements.org/vocab/InC/1.0/"],"identifier_entries":[{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["vt_gsexam:17717"],"render_values":[{"text":"vt_gsexam:17717","href":null,"code":true}]}]},"links":{"outbound_url":"http://hdl.handle.net/10919/87064","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.committeechair","label":"Committee Chair","values":["Cameron, Kirk W."]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Butt, Ali R.","Lee, Dongyoon","Leon, Edgar A.","Back, Godmar V."]},{"key":"dc:contributor.department","label":"Department","values":["Computer Science"]},{"key":"dc:creator","label":"Author","values":["Li, Bo"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2019-01-29T09:00:21Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2019-01-29T09:00:21Z"]},{"key":"dc:date.issued","label":"Date","values":["2019-01-28"]},{"key":"dc:publisher","label":"Institution","values":["Virginia Tech"]},{"key":"dc:type","label":"Dc Type","values":["Dissertation"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science and Applications"]},{"key":"thesis:degree_level","label":"Degree Level","values":["doctoral"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Ph. D."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Virginia Polytechnic Institute and State University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Parallel Performance Modeling","Dynamic Voltage and Frequency Scaling","Dynamic Memory Throttling","Dynamic Concurrency Throttling","Shared-Memory Systems"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights","label":"Dc Rights","values":["In Copyright"]},{"key":"dc:rights.uri","label":"Rights URI","values":["http://rightsstatements.org/vocab/InC/1.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["vt_gsexam:17717"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["http://hdl.handle.net/10919/87064"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Emergent 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."]},{"key":"dc:description.abstractgeneral","label":"General Abstract","values":["System efficiency on high-performance computing (HPC) systems is the key to achieving the goal of power budget for exascale supercomputers. Techniques for adjusting the performance of different system components can help accomplish this goal by dynamically controlling system performance according to application behaviors. In this dissertation, we focus on three techniques: adjusting CPU performance, memory performance, and the number of threads for running parallel applications. First, we profile the performance and energy consumption of different HPC applications on both Intel systems with accelerators and IBM BG/Q systems. We explore the trade-offs of performance and energy under these techniques and provide optimization insights. Furthermore, we propose a parallel performance model that can accurately capture the impact of these techniques on performance in terms of job completion time. We present an approximation approach for performance prediction. The approximation has up to 7% and 17% prediction error on Intel x86 and IBM BG/Q systems respectively under 19 HPC applications. Thereafter, we apply the performance model in a runtime system design for improving performance under a given power budget. Our runtime strategy achieves up to 20% performance improvement to the baseline method."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["Ph. D."]},{"key":"dc:format.medium","label":"Dc Format Medium","values":["ETD"]},{"key":"dc:title","label":"Title","values":["Modeling and Runtime Systems for Coordinated Power-Performance Management"]}]}],"canonical_facts":{"dc:contributor.committeechair":["Cameron, Kirk W."],"dc:contributor.committeemember":["Butt, Ali R.","Lee, Dongyoon","Leon, Edgar A.","Back, Godmar V."],"dc:contributor.department":["Computer Science"],"dc:creator":["Li, Bo"],"dc:date.accessioned":["2019-01-29T09:00:21Z"],"dc:date.available":["2019-01-29T09:00:21Z"],"dc:date.issued":["2019-01-28"],"dc:description.abstract":["Emergent 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."],"dc:description.abstractgeneral":["System efficiency on high-performance computing (HPC) systems is the key to achieving the goal of power budget for exascale supercomputers. Techniques for adjusting the performance of different system components can help accomplish this goal by dynamically controlling system performance according to application behaviors. In this dissertation, we focus on three techniques: adjusting CPU performance, memory performance, and the number of threads for running parallel applications. First, we profile the performance and energy consumption of different HPC applications on both Intel systems with accelerators and IBM BG/Q systems. We explore the trade-offs of performance and energy under these techniques and provide optimization insights. Furthermore, we propose a parallel performance model that can accurately capture the impact of these techniques on performance in terms of job completion time. We present an approximation approach for performance prediction. The approximation has up to 7% and 17% prediction error on Intel x86 and IBM BG/Q systems respectively under 19 HPC applications. Thereafter, we apply the performance model in a runtime system design for improving performance under a given power budget. Our runtime strategy achieves up to 20% performance improvement to the baseline method."],"dc:description.degree":["Ph. D."],"dc:format.medium":["ETD"],"dc:identifier.other":["vt_gsexam:17717"],"dc:identifier.uri":["http://hdl.handle.net/10919/87064"],"dc:publisher":["Virginia Tech"],"dc:rights":["In Copyright"],"dc:rights.uri":["http://rightsstatements.org/vocab/InC/1.0/"],"dc:subject":["Parallel Performance Modeling","Dynamic Voltage and Frequency Scaling","Dynamic Memory Throttling","Dynamic Concurrency Throttling","Shared-Memory Systems"],"dc:title":["Modeling and Runtime Systems for Coordinated Power-Performance Management"],"dc:type":["Dissertation"],"thesis:degree_discipline":["Computer Science and Applications"],"thesis:degree_level":["doctoral"],"thesis:degree_name":["Ph. D."],"thesis:institution_name":["Virginia Polytechnic Institute and State University"]},"updated_at":"2026-07-22T22:19:29Z"}