Bone structure analysis on multiple GPGPUs
ETH Zurich, Computer Science Department, Universitatsstrasse 6, 8092 Zurich, Switzerland
Eidgenossische Technische Hochschule Zurich, 2014
@article{arbenz2014bone,
title={Bone structure analysis on multiple GPGPUs},
author={Arbenz, Peter and Flaig, Cyril and Kellenberger, Daniel},
year={2014},
publisher={Eidgen{"o}ssische Technische Hochschule Z{"u}rich}
}
Osteoporosis is a disease that affects a growing number of people by increasing the fragility of their bones. To improve the understanding of the bone quality, large scale computer simulations are applied. A fast, scalable and memory efficient solver for such problems is ParOSol. It uses the preconditioned conjugate gradient algorithm with a multigrid preconditioner. A modification of ParOSol is presented that profits from the exorbitant compute capabilities of recent general-purpose graphics processing units (GPGPUs). Adaptations of data structures for the GPGPU are discussed. The fastest implementation on a GPGPU achieves a speedup of more than five compared with the CPU implementation and scales from 1 to at least 256 GPGPUs.
February 5, 2014 by hgpu
