Comparing Hardware Accelerators in Scientific Applications: A Case Study
University of Tennessee, Knoxville
IEEE Transactions on Parallel and Distributed Systems, Vol. 99, No. 1. (5555)
@article{weber2010comparing,
title={Comparing hardware accelerators in scientific applications: A case study},
author={Weber, R. and Gothandaraman, A. and Hinde, R.J. and Peterson, G.D.},
journal={IEEE Transactions on Parallel and Distributed Systems},
issn={1045-9219},
year={2010},
publisher={Published by the IEEE Computer Society}
}
Multi-core processors and a variety of accelerators have allowed scientific applications to scale to larger problem sizes. We present a performance, design methodology, platform, and architectural comparison of several application accelerators executing a Quantum Monte Carlo application. We compare the application’s performance and programmability on a variety of platforms including CUDA with Nvidia GPUs, Brook+ with ATI graphics accelerators, OpenCL running on both multi-core and graphics processors, C++ running on multi-core processors, and a VHDL implementation running on a Xilinx FPGA. We show that OpenCL provides application portability between multi-core processors and GPUs, but may incur a performance cost. Furthermore we illustrate that graphics accelerators can make simulations involving large numbers of particles feasible.
January 10, 2011 by hgpu
