TU Darmstadt / ULB / TUbiblio

Iterative SLE Solvers over a CPU-GPU Platform

Binotto, Alécio P. D. and Daniel, Christian G. and Weber, Daniel and Kuijper, Arjan and Stork, André and Pereira, Carlos Eduardo and Fellner, Dieter W. (2010):
Iterative SLE Solvers over a CPU-GPU Platform.
IEEE Computer Society Conference Publishing Services (CPS), Los Alamitos, Calif., In: Proceedings 2010 12th IEEE International Conference on High Performance Computing and Communications, pp. 305-313, DOI: 10.1109/HPCC.2010.40,
[Conference or Workshop Item]

Abstract

GPUs (Graphics Processing Units) have become one of the main co-processors that contributed to desktops towards high performance computing. Together with multi-core CPUs, a powerful heterogeneous execution platform is built for massive calculations. To improve application performance and explore this heterogeneity, a distribution of workload in a balanced way over the PUs (Processing Units) plays an important role for the system. However, this problem faces challenges since the cost of a task at a PU is non-deterministic and can be influenced by several parameters not known a priori, like the problem size domain. We present a comparison of iterative SLE (Systems of Linear Equations) solvers, used in many scientific and engineering applications, over a heterogeneous CPU-GPUs platform and characterize scenarios where the solvers obtain better performances. A new technique to improve memory access on matrix vector multiplication used by SLEs on GPUs is described and compared to standard implementations for CPU and GPUs. Such timing profiling is analyzed and break-even points based on the problem sizes are identified for this implementation, pointing whether our technique is faster to use GPU instead of CPU. Preliminary results show the importance of this study applied to a real-time CFD (Computational Fluid Dynamics) application with geometry modification.

Item Type: Conference or Workshop Item
Erschienen: 2010
Creators: Binotto, Alécio P. D. and Daniel, Christian G. and Weber, Daniel and Kuijper, Arjan and Stork, André and Pereira, Carlos Eduardo and Fellner, Dieter W.
Title: Iterative SLE Solvers over a CPU-GPU Platform
Language: English
Abstract:

GPUs (Graphics Processing Units) have become one of the main co-processors that contributed to desktops towards high performance computing. Together with multi-core CPUs, a powerful heterogeneous execution platform is built for massive calculations. To improve application performance and explore this heterogeneity, a distribution of workload in a balanced way over the PUs (Processing Units) plays an important role for the system. However, this problem faces challenges since the cost of a task at a PU is non-deterministic and can be influenced by several parameters not known a priori, like the problem size domain. We present a comparison of iterative SLE (Systems of Linear Equations) solvers, used in many scientific and engineering applications, over a heterogeneous CPU-GPUs platform and characterize scenarios where the solvers obtain better performances. A new technique to improve memory access on matrix vector multiplication used by SLEs on GPUs is described and compared to standard implementations for CPU and GPUs. Such timing profiling is analyzed and break-even points based on the problem sizes are identified for this implementation, pointing whether our technique is faster to use GPU instead of CPU. Preliminary results show the importance of this study applied to a real-time CFD (Computational Fluid Dynamics) application with geometry modification.

Publisher: IEEE Computer Society Conference Publishing Services (CPS), Los Alamitos, Calif.
Uncontrolled Keywords: Graphics processors, Parallel processing, Computational fluid dynamics (CFD)
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Interactive Graphics Systems
Event Title: Proceedings 2010 12th IEEE International Conference on High Performance Computing and Communications
Date Deposited: 12 Nov 2018 11:16
DOI: 10.1109/HPCC.2010.40
Export:
Suche nach Titel in: TUfind oder in Google
Send an inquiry Send an inquiry

Options (only for editors)

View Item View Item