Development of a Computational Steering Framework for High Performance Computing Environments on Blue Gene/P Systems

Handle URI:
http://hdl.handle.net/10754/267252
Title:
Development of a Computational Steering Framework for High Performance Computing Environments on Blue Gene/P Systems
Authors:
Danani, Bob K.
Abstract:
Computational steering has revolutionized the traditional workflow in high performance computing (HPC) applications. The standard workflow that consists of preparation of an application’s input, running of a simulation, and visualization of simulation results in a post-processing step is now transformed into a real-time interactive workflow that significantly reduces development and testing time. Computational steering provides the capability to direct or re-direct the progress of a simulation application at run-time. It allows modification of application-defined control parameters at run-time using various user-steering applications. In this project, we propose a computational steering framework for HPC environments that provides an innovative solution and easy-to-use platform, which allows users to connect and interact with running application(s) in real-time. This framework uses RealityGrid as the underlying steering library and adds several enhancements to the library to enable steering support for Blue Gene systems. Included in the scope of this project is the development of a scalable and efficient steering relay server that supports many-to-many connectivity between multiple steered applications and multiple steering clients. Steered applications can range from intermediate simulation and physical modeling applications to complex computational fluid dynamics (CFD) applications or advanced visualization applications. The Blue Gene supercomputer presents special challenges for remote access because the compute nodes reside on private networks. This thesis presents an implemented solution and demonstrates it on representative applications. Thorough implementation details and application enablement steps are also presented in this thesis to encourage direct usage of this framework.
Advisors:
Keyes, David E. ( 0000-0002-4052-7224 )
Committee Member:
Ahmadia, Aron ( 0000-0002-2573-2481 ) ; Douglas, Craig C.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Program:
Computer Science
Issue Date:
Jul-2012
Type:
Thesis
Appears in Collections:
Theses; Computer Science Program; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.advisorKeyes, David E.en
dc.contributor.authorDanani, Bob K.en
dc.date.accessioned2013-01-27T06:18:41Z-
dc.date.available2013-01-27T06:18:41Z-
dc.date.issued2012-07en
dc.identifier.urihttp://hdl.handle.net/10754/267252en
dc.description.abstractComputational steering has revolutionized the traditional workflow in high performance computing (HPC) applications. The standard workflow that consists of preparation of an application’s input, running of a simulation, and visualization of simulation results in a post-processing step is now transformed into a real-time interactive workflow that significantly reduces development and testing time. Computational steering provides the capability to direct or re-direct the progress of a simulation application at run-time. It allows modification of application-defined control parameters at run-time using various user-steering applications. In this project, we propose a computational steering framework for HPC environments that provides an innovative solution and easy-to-use platform, which allows users to connect and interact with running application(s) in real-time. This framework uses RealityGrid as the underlying steering library and adds several enhancements to the library to enable steering support for Blue Gene systems. Included in the scope of this project is the development of a scalable and efficient steering relay server that supports many-to-many connectivity between multiple steered applications and multiple steering clients. Steered applications can range from intermediate simulation and physical modeling applications to complex computational fluid dynamics (CFD) applications or advanced visualization applications. The Blue Gene supercomputer presents special challenges for remote access because the compute nodes reside on private networks. This thesis presents an implemented solution and demonstrates it on representative applications. Thorough implementation details and application enablement steps are also presented in this thesis to encourage direct usage of this framework.en
dc.language.isoenen
dc.titleDevelopment of a Computational Steering Framework for High Performance Computing Environments on Blue Gene/P Systemsen
dc.typeThesisen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
thesis.degree.grantorKing Abdullah University of Science and Technologyen_GB
dc.contributor.committeememberAhmadia, Aronen
dc.contributor.committeememberDouglas, Craig C.en
thesis.degree.disciplineComputer Scienceen
thesis.degree.nameMaster of Scienceen
dc.person.id113239en
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.