Grant: Engage Grants Program
University of Toronto
$25,000.00
Optimization of astrophysical simulation performance on accelerator platforms
FY 2016-17
Summary
High Performance Computing (HPC) methods are widely implemented in modern science to solve highly nonlinear problems,which can be solved only numerically. Considerations of electricity, time, and funds lead to an increase in usage of heterogeneousarchitectures co-processors and graphics processing units (GPUs) to conduct massive floating point computations much moreefficiently. However, the variations of difficult, and often proprietary, languages (Cell, CUDA, OpenCL, HCC etc.) still limit the GPUadoption in the scientific community, even for common algorithms (fluid dynamics, N-body simulations). Scientists are weary ofhaving to reimplement codes for migration to new platforms. The AMD Firepro S9300x2 GPUs plans to support the common andstandard OpenMP 4 model, which opens the path to broader acceptance.This project will implement and optimize High Performance Computing (HPC) Computational Fluid Dynamics (CFD) codes (Panget al. 2010, arXiv.org/1004.1680) on the AMD Firepro S9300x2, and compare performance to a range of other platforms in whichthe code has been bench marked, including Nvidia, Intel PHI, and IBM Cell. The project will provide open source implementationof an astrophysical production code on the AMD platform, which benefits adaption of the product, and position AMD morecompetitively in the Compute Canada upgrade in 2017.
Program
Engage Grants Program
Recipient
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
506154-2016
Name nom
Pen, UeLi
Department département
Cdn Inst. for Theoretical Astrophysics
Organization id
31
Institution établissement
University of Toronto
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2016
Competition year année de concours
2016
Award amount
25000
Program id
EGP
Program name en
Engage Grants Program
Program name fr
Programme de subventions d'engagement partenariat
Group en
Research partnerships programs
Group fr
Programmes de partenariat
Committee code
1553
Committee name en
Ontario Internal Decision Committee
Committee name fr
Comité de décision interne pour l'Ontario
Area of application code
801
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Computer software
Area of application fr
Logiciel
Research subject code
3107
Research subject group en
Physics
Research subject group fr
Physique
Research subject en
Plasma physics
Research subject fr
Physique des plasmas
Partie
1
Nb partie
1
Application title
Optimization of astrophysical simulation performance on accelerator platforms
Keyword
NSERC
Application summary
High Performance Computing (HPC) methods are widely implemented in modern science to solve highly nonlinear problems,which can be solved only numerically. Considerations of electricity, time, and funds lead to an increase in usage of heterogeneousarchitectures co-processors and graphics processing units (GPUs) to conduct massive floating point computations much moreefficiently. However, the variations of difficult, and often proprietary, languages (Cell, CUDA, OpenCL, HCC etc.) still limit the GPUadoption in the scientific community, even for common algorithms (fluid dynamics, N-body simulations). Scientists are weary ofhaving to reimplement codes for migration to new platforms. The AMD Firepro S9300x2 GPUs plans to support the common andstandard OpenMP 4 model, which opens the path to broader acceptance.This project will implement and optimize High Performance Computing (HPC) Computational Fluid Dynamics (CFD) codes (Panget al. 2010, arXiv.org/1004.1680) on the AMD Firepro S9300x2, and compare performance to a range of other platforms in whichthe code has been bench marked, including Nvidia, Intel PHI, and IBM Cell. The project will provide open source implementationof an astrophysical production code on the AMD platform, which benefits adaption of the product, and position AMD morecompetitively in the Compute Canada upgrade in 2017.