Optimization Techniques for CUDA Application
Department of Computer and Information Science and Engineering, University of Florida, Gainesville, FL, US
Project Report of CIS6930 GPU: Parallel Architecture and Programming, 2012
@article{athawale2012optimization,
title={Optimization Techniques for CUDA Application},
author={Athawale, T. and Xu, X.},
year={2012}
}
In this paper, we summarize our experiment results of applying various optimization techniques for CUDA application running on NVIDIA Fermi GPUs. Our experiments on matrix multiplication and breadth first search algorithms show that optimization techniques such as coalesced global memory access, conflict-free shared memory access and data pre-fetching improve the performance of applications running on GPUs.
August 28, 2012 by hgpu
