Scheduling of computations between GPGPU and CPU for cellular automata models

Authors:
Lukasz Rauch;Artur Rodzaj
Affiliations:
University of Science and Technology, Krakow, Poland;University of Science and Technology, Krakow, Poland
Venue:
Proceedings of the 2012 SpringSim Poster & Work-In-Progress Track
Year:
2012

Citing 1
Cited 0

A comparison of CPUs, GPUs, FPGAs, and massively parallel processor arrays for random number generation

Proceedings of the ACM/SIGDA international symposium on Field programmable gate arrays

Quantified Score

Hi-index	0.00

Visualization

Abstract

Heterogeneous hardware architectures play very important role in high performance computing. This trend is also visible in multiscale simulations, where algorithms characterized by different computational behavior are responsible for calculations in different scales. This paper presents analysis of cellular automata model, used for modeling of material phenomena in micro scale, dependently on a problem size and a number of iterations. The OpenCL kernels generated by the framework are performed on GPGPU and CPU separately as well as simultaneously. The obtained results are used to detect the optimal load balancing between CPU and GPGPU and minimize a time of calculations.