Developing Fast Fourier Transforms using the EXTENT Parallel Programming Environmentz

نویسندگان

  • D. L. Dai
  • S. K. S. Gupta
  • S. D. Kaushik
  • T.-H. Lin
  • J. H. Lu
  • R. V. Singh
  • D. Zhou
  • C.-H. Huang
  • P. Sadayappan
چکیده

EXTENT stands for an EXpert system for TENsor product formula Translation. In this paper, we describe the development of programs for fast Fourier transforms (FFT) for vector multiprocessors and distributed-memory machines using the EXTENT system for formula translation. Several forms of the FFT have been presented in the literature. Each form has diierent characteristics and is best suited for good performance on a particular architecture. The tensor product has been shown to be a uniform representation for all of these forms. We augment the tensor product representation of the FFT to include a mixed radix and two-level representation. The code generation methodology , including the generation of the complicated indexing code, characteristic of FFT programs, is described. Extensions of the code generation for the FFT forms on vector multiprocessor and distributed-memory machines are described. Performance results for the automatically generated FFT codes on the Cray Y-MP and Intel iPSC/860 are presented.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Pathologies cardiac discrimination using the Fast Fourir Transform (FFT) The short time Fourier transforms (STFT) and the Wigner distribution (WD)

This paper is concerned with a synthesis study of the fast Fourier transform (FFT), the short time Fourier transform (STFT and the Wigner distribution (WD) in analysing the phonocardiogram signal (PCG) or heart cardiac sounds.     The FFT (Fast Fourier Transform) can provide a basic understanding of the frequency contents of the heart sounds. The STFT is obtained by calculating the Fourier tran...

متن کامل

Parallel Architecture for Fast Transforms with Trigonometric Kernel

We present an unified parallel architecture for four of the most important fast orthogonal transforms with trigonometric kernel: Complex Valued Fourier (CFFT), Real Valued Fourier (RFFT), Hartley (FHT), and Cosine (FCT). Out of these, only the CFFT has a data flow coinciding with the one generated by the successive doubling method, which can be transformed on a constant geometry flow using perf...

متن کامل

Parallel Three-Dimensional Nonequispaced Fast Fourier Transforms and Their Application to Particle Simulation

In this paper we describe a parallel algorithm for calculating nonequispaced fast Fourier transforms on massively parallel distributed memory architectures. These algorithms are implemented in an open source software library called PNFFT. Furthermore, we derive a parallel fast algorithm for the computation of the Coulomb potentials and forces in a charged particle system, which is based on the ...

متن کامل

A Self-sorting In-place Fast Fourier Transform Algorithm Suitable for Vector and Parallel Processing

We propose a new algorithm for fast Fourier transforms. This algorithm features uniformly long vector lengths and stride one data access. Thus it is well adapted to modern vector computers like the Fujitsu VP2200 having several oating point pipelines per CPU and very fast stride one data access. It also has favorable properties for distributed memory computers as all communication is gathered t...

متن کامل

Implementing 2-d and 3-d Discrete Hartley Transforms on a Massively Parallel Simd Mesh Computer

Discrete Hartley transform (DHT) is known to outperform fast Fourier transform (FFT) on sequential machines. Here we investigate parallel algorithms and implementations of twoand three-dimensional DHT in order to determine if the advantage of Hartley transforms over Fourier transforms carries over to parallel environment as well. Our extensive empirical study of the performances of DHT and FFT ...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2007