A Comparison of Modern GPU and CPU Architectures: And the Common Convergence of Both
Oregon State University
Oregon State University, 2011
@article{palacios2011comparison,
title={A Comparison of Modern GPU and CPU Architectures: And the Common Convergence of Both},
author={Palacios, J. and Triska, J.},
year={2011}
}
In the past few decades, processor technology specifically designed for the processing and output of graphical data has become a major market. With the rise of parallelism as an important method of improving processor throughput, Graphics Processing Units (GPUs) have come to drive architecture demands in many ways. In this work, we plan to explore the differences in modern GPU and CPU architectures. For the GPU, we will focus on the latest NVIDIA and AMD technologies (named Fermi and Cayman, respectively), and for the CPU, we will focus on the hardware studied in class. Our main goal is to highlight the major differences in data flow and management, hardware layout, and optimization strategies.
December 17, 2011 by hgpu
