Toward a Multi-level Parallel Framework on GPU Cluster with PetSC-CUDA for PDE-based Optical Flow Computation
University of Naples Federico II, Department of Mathematics and Applications, Naples, Italy
Procedia Computer Science, Volume 51, Pages 170-179, 2015
@article{cuomo2015toward,
title={Toward a Multi-level Parallel Framework on GPU Cluster with PetSC-CUDA for PDE-based Optical Flow Computation},
author={Cuomo, S and Galletti, A and Giunta, G and Marcellino, L},
journal={Procedia Computer Science},
volume={51},
pages={170–179},
year={2015},
publisher={Elsevier}
}
In this work we present a multi-level parallel framework for the Optical Flow computation on a GPUs cluster, equipped with a scientific computing middleware (the PetSc library). Starting from a flow-driven isotropic method, which models the optical flow problem through a parabolic partial differential equation (PDE), we have designed a parallel algorithm and its software implementation that is suitable for heterogeneous computing environments (multiprocessor, single GPU and cluster of GPUs). The proposed software has been tested on real SAR images sequences. Numerical experiments highlight the performance of the proposed software framework, which can reach a gain of about 95% with respect to the sequential implementation.
June 24, 2015 by hgpu
