The dataflow time and space complexity of FFTs

Journal Article · Thu Jul 01 00:00:00 EDT 1993 · Journal of Parallel and Distributed Computing; (United States)

DOI:https://doi.org/10.1006/jpdc.1993.1066· OSTI ID:6303064

Boehm, A P ^[1]; Hiromoto, R E ^[2]

---

1. Colorado State Univ., Fort Collins, CO (United States). Computer Science Dept.
2. Los Alamos National Lab., Los Alamos, NM (United States). Computer Research and Applications

In this paper the authors analyze the performance of a recursive and an iterative fast Fourier transform algorithm, written in Id and run on MINT, a simulator for the Monsoon dataflow machine. Their complexity measures are: the number of instructions executed, the critical path length of the dataflow graph, and the storage occupancy. Using a set of simple functions, they calibrate loop and divide-and-conquer behavior of Monsoon and compare it with the tagged token dataflow architecture. They discuss the issues in explicit resource management in a functional language, introduce the language features required for this, and show results of their resource managed FFT programs.

• DOE Contract Number: W-7405-ENG-36
• OSTI ID: 6303064
• Journal Information: Journal of Parallel and Distributed Computing; (United States), Journal Name: Journal of Parallel and Distributed Computing; (United States) Vol. 18:3; ISSN JPDCER; ISSN 0743-7315
• Country of Publication: United States
• Language: English

Related Subjects

99 GENERAL AND MISCELLANEOUS
990200* -- Mathematics & Computers
ALGORITHMS
ANALOG SYSTEMS
CALCULATION METHODS
COMPARATIVE EVALUATIONS
COMPUTER ARCHITECTURE
DATA-FLOW PROCESSING
EVALUATION
FOURIER TRANSFORMATION
FUNCTIONAL MODELS
INTEGRAL TRANSFORMATIONS
ITERATIVE METHODS
MATHEMATICAL LOGIC
MEMORY DEVICES
PERFORMANCE
PROGRAMMING
PROGRAMMING LANGUAGES
SIMULATORS
TRANSFORMATIONS