Domain-specific library generation for parallel software and hardware platforms

Author

Franchetti, Franz ; Voronenko, Yevgen ; Milder, Peter A. ; Chellappa, Srinivas ; Telgarsky, Marek R. ; Shen, Hao ; Alberto, Paolo D. ; De Mesmay, Frédéric ; Hoe, James C. ; Moura, José M F ; Püschel, Markus

Author_Institution

Carnegie Mellon Univ., Pittsburg, PA

fYear

2008

Firstpage

1

Lastpage

5

Abstract

We overview a library generation framework called Spiral. For the domain of linear transforms, Spiral automatically generates implementations for parallel platforms including SIMD vector extensions, multicore processors, field-programmable gate arrays (FPGAs) and FPGA accelerated processors. The performance of the generated code is competitive with the best available hand-written libraries.

Keywords

field programmable gate arrays; parallel programming; software libraries; FPGA accelerated processors; SIMD vector extensions; Spiral; domain specific library generation; field programmable gate arrays; multicore processors; parallel hardware platforms; parallel software platforms; Discrete Fourier transforms; Discrete transforms; Discrete wavelet transforms; Field programmable gate arrays; Fourier transforms; Hardware; Parallel processing; Signal processing algorithms; Software libraries; Spirals;

fLanguage

English

Publisher

ieee

Conference_Titel

Parallel and Distributed Processing, 2008. IPDPS 2008. IEEE International Symposium on

Conference_Location

Miami, FL

ISSN

1530-2075

Print_ISBN

978-1-4244-1693-6

Electronic_ISBN

1530-2075

Type

conf

DOI

10.1109/IPDPS.2008.4536398

Filename

4536398