Optimized FFT computations on heterogeneous platforms with application to the Poisson equation

We develop optimized multi-dimensional FFT implementations on CPU–GPU heterogeneous platforms for the case when the input is too large to fit on the GPU global memory, and use the resulting techniques to develop a fast Poisson solver. The solver involves memory bound computations for which the large...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of parallel and distributed computing Jg. 74; H. 8; S. 2745 - 2756
Hauptverfasser: Wu, Jing, JaJa, Joseph
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Amsterdam Elsevier Inc 01.08.2014
Elsevier
Schlagworte:
Applied sciences
Boundary conditions
Buses (vehicles)
Computation
Computer science; control theory; systems
Computer systems and distributed systems. User interface
CUDA GPU
Data transfer (computers)
Exact sciences and technology
Fast Fourier transforms
Operation, maintenance, reliability
Parallel and vector implementations
Platforms
Poisson equation
Poisson equations
Software
Solvers
Systems, networks and services of telecommunications
Telecommunications
Telecommunications and information theory
Three dimensional
Poisson equations
Fast Fourier transforms
Parallel and vector implementations
CUDA GPU
Performance evaluation
Lower bound
Neumann problem
Parallel algorithm
Fast Fourier transformation
Data transmission
Poisson equation
Bus(channel)
Distributed system
Boundary condition
Synchronization
Heterogeneity
Computation time
Graphic processing unit
Bandwidth
Central unit
Accelerator
Execution time
ISSN:0743-7315, 1096-0848
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:We develop optimized multi-dimensional FFT implementations on CPU–GPU heterogeneous platforms for the case when the input is too large to fit on the GPU global memory, and use the resulting techniques to develop a fast Poisson solver. The solver involves memory bound computations for which the large 3D data may have to be transferred over the PCIe bus several times during the computation. We develop a new strategy to decompose and allocate the computation between the GPU and the CPU such that the 3D data is transferred only once to the device memory, and the executions of the GPU kernels are almost completely overlapped with the PCI data transfer. We were able to achieve significantly better performance than what has been reported in previous related work, including over 145 GFLOPS for the three periodic boundary conditions (single precision version), and over 105 GFLOPS for the two periodic, one Neumann boundary conditions (single precision version). The effective bidirectional PCIe bus bandwidth achieved is 9–10 GB/s, which is close to the best possible on our platform. For all the cases tested, the single 3D data PCIe transfer time, which constitutes a lower bound on what is possible on our platform, takes almost 70% of the total execution time of the Poisson solver. •New strategy to decompose large multi-dimensional FFTs on CPU–GPU platforms.•Executions of GPU kernels are almost completely overlapped with PCI bus transfer.•Multi-dimensional data is transferred only once between the GPU and CPU.•Scheme is equally effective for the single and double precision computations.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0743-7315
1096-0848
DOI:10.1016/j.jpdc.2014.03.009