Adaptive Computation of Self Sorting In-place FFTs on Hierarchical Memory Architectures
2007 (English)In: HIGH PERFORMANCE COMPUTING AND COMMUNICATIONS, PROCEEDINGS, 2007, 372-383 p.Conference paper (Refereed)
Computing ”in-place and in-order”FFT poses a very difficult problem on hierarchical memory architectures where data movement can seriously degrade the performance. In this paper we present recursive formulation of a self sorting in-place FFT algorithm that adapts to the target architecture. For transform sizes where an in-place, in-order execution is not possible, we show how schedules can be constructed that use minimum work-space to perform the computation efficiently. In order to express and construct FFT schedules, we present a context free grammar that generates the FFT Schedule Specification Language. We conclude by comparing the performance of our in-place in-order FFT implementation with that of other well known FFT libraries. We also present a performance comparison between the out-of-place and in-place execution of various FFT sizes.
Place, publisher, year, edition, pages
2007. 372-383 p.
, Lecture Notes in Computer Science, ISSN 0302-9743 ; 4782
Adaptive computation, Data movement, Schedule Specification Language
Computer and Information Science
IdentifiersURN: urn:nbn:se:kth:diva-63146ISI: 000250940200032ISBN: 978-354075443-5OAI: oai:DiVA.org:kth-63146DiVA: diva2:481697
3rd International Conference on High Performance Computing and Communications, HPCC 2007, Houston, TX, 26 September 2007 through 28 September 2007
QC 201202012012-01-222012-01-222012-02-04Bibliographically approved