FFTW and Complex Ambiguity Function performance on the Maestro processor

Author

Singh, Karandeep ; Walters, John Paul ; Hestness, Joel ; Suh, Jinwoo ; Rogers, Craig M. ; Crago, Stephen P.

Author_Institution

Inf. Sci. Inst., Univ. of Southern California, Arlington, VA, USA

fYear

2011

fDate

5-12 March 2011

Firstpage

1

Lastpage

8

Abstract

Maestro is a 49-core general-purpose, radiation-hardened processor for space. Fourier Transform is an important computation in many space applications. FFTW is a high performance package that computes one and multidimensional Fourier Transforms. We ported FFTW library to the Maestro processor and used it extensively in implementing the Complex Ambiguity Function (CAF), which is a reference space application. We were able to obtain 145 MFLOPS (0.36 FLOPs/Cycle) on a single Maestro core with double-precision computations and 202 MFLOPS (0.50 FLOPs/Cycle) on a single Maestro core with single-precision computations running at 260 MHz.

Keywords

astronomy computing; fast Fourier transforms; multiprocessing systems; FFTW; MFLOPS; Maestro processor; complex ambiguity function performance; multicore general purpose processor; multidimensional Fourier Transforms; radiation-hardened processor; Computer architecture; Discrete Fourier transforms; Hardware; Libraries; System-on-a-chip; Tiles;

fLanguage

English

Publisher

ieee

Conference_Titel

Aerospace Conference, 2011 IEEE

Conference_Location

Big Sky, MT

ISSN

1095-323X

Print_ISBN

978-1-4244-7350-2

Type

conf

DOI

10.1109/AERO.2011.5747455

Filename

5747455