Modeling of Heterogeneous Architecture with GPU to Exascale System

Daniel A. G. de Oliveira, Pedro Velho, Edson L. Padoin, Philippe O. A. Navaux
Institute of Informatics, Federal University of Rio Grande do Sul (UFRGS) – Porto Alegre, RS – Brazil
Workshop de Processamento Paralelo e Distribuido (WSPPD 2012), 2012


   title={Modeling of Heterogeneous Architecture with GPU to Exascale System},

   author={de Oliveira, Daniel AG and Velho, Pedro and Padoin, Edson L and Navaux, Philippe OA},



Download Download (PDF)   View View   Source Source   



The High-Performance Computing (HPC) community aimed for many years at increasing performance regardless of energy consumption. However, energy is limiting the scalability of the next generation of supercomputers. Current HPC systems already consume huge amounts of power, in the order of a few MegaWatts (MW). The future HPC systems intend to achieve 10 to 100 times more performance, but the accepted energy to power those machines must remain below 20 MW. Therefore, the scientific community is investigating ways to improve energy efficiency. We argue that Graphics Processing Units (GPUs) bring an alternative base to build HPC clusters that respect the limits of electric energy. This paper presents a way to model the GPU. Our objective is to help to understand the behavior of GPU’s applications in exascale systems. We show a simple model for GPUs that can be used for large scale simulators.
No votes yet.
Please wait...

* * *

* * *

HGPU group © 2010-2021 hgpu.org

All rights belong to the respective authors

Contact us: