A comparative benchmarking of the FFT on Fermi and Evergreen GPUs

Mohamed F. Ahmed; Omar Haridy

doi:10.1109/ISPASS.2011.5762726

(IEEE ISPASS) IEEE International Symposium on Performance Analysis of Systems and Software

A comparative benchmarking of the FFT on Fermi and Evergreen GPUs

Year: 2011, Pages: 127-128

DOI Bookmark: 10.1109/ISPASS.2011.5762726

Authors

Mohamed F. Ahmed, Computer Science and Engineering, The American University in Cairo, Cairo, Egypt
Omar Haridy, Media Engineering and Technology, German University in Cairo, Cairo, Egypt

Abstract

NVIDIA and AMD GPUs are are gaining traction in HPC for their performance and architectural aspects. It is very important to measure and analyze the relative power of each architecture. In this paper, we analyze the architecture of NVIDIA's Fermi and AMD's Evergreen processors and demonstrate the best practices and techniques to best utilize the capabilities of each architecture. We implemented the FFT on both cards utilizing our findings to reach new performance ceilings on both GPUs.

Like what you’re reading?

Already a member?

Get this article FREE with a new membership!

Static WCET Analysis of GPUs with Predictable Warp Scheduling
2017 IEEE 20th International Symposium on Real-Time Distributed Computing (ISORC)
Auto-tuning 3-D FFT library for CUDA GPUs
2009 SC Conference on High Performance Computing Networking, Storage and Analysis
Defend GPUs against DoS attacks
2013 IEEE 32nd International Performance Computing and Communications Conference (IPCCC)
A software-based self test of CUDA Fermi GPUs
2013 18th IEEE European Test Symposium (ETS 2013)
Automatic FFT Performance Tuning on OpenCL GPUs
Parallel and Distributed Systems, International Conference on
Bandwidth intensive 3-D FFT kernel for GPUs using CUDA
SC Conference
Benchmarking GPUs to tune dense linear algebra
SC Conference
Directive-based Programming for GPUs: A Comparative Study
High Performance Computing and Communication & IEEE International Conference on Embedded Software and Systems, IEEE International Conference on
Low Communication FMM-Accelerated FFT on GPUs
SC17: International Conference for High Performance Computing, Networking, Storage and Analysis
DO-GPU: Domain Optimizable Soft GPUs
2021 31st International Conference on Field-Programmable Logic and Applications (FPL)

A comparative benchmarking of the FFT on Fermi and Evergreen GPUs

Authors

Abstract

Related Articles