FATSEA-An Architectural Simulator for General Purpose Computing on GPUs
Computer Engineering and Technology Dept., University of Murcia, Spain
2nd Workshop on Rapid Simulation and Performance Evaluation: Methods and Tools (co-located with 5th Int’l. Conf. on High-Performance and Embedded Architectures and Compilers, HiPEAC’10), 2010
@article{ostby2010fatsea,
title={FATSEA–An Architectural Simulator for General Purpose Computing on GPUs},
author={O}stby, KE and Arag{‘o}n, JL and Garc{i}a, JM and Ujald{‘o}n, M.},
publisher={Citeseer},
year={2010}
}
We present FATSEA, a functional and performance evaluation simulator written in C++ to handle kernels written in the CUDA programming language aimed for GPGPU computing. FATSEA takes a Parallel Thread eXecution (PTX ) code as input, which is a device independent code format generated by the Nvidia CUDA compiler, to validate results and estimate performance on Nvidia platforms. This paper shows results on a G80-based architecture for a set of well-known kernels to illustrate the usefulness of our framework while performing a preliminary validation for the tool.
September 30, 2011 by hgpu
