Communication efficient adaptive matrix transpose algorithm for FFT on symmetric multiprocessors

Computing fast Fourier transform (FFT) on parallel computers has the same communication requirement to transpose matrices one or more times. In this paper, we propose an efficient algorithm (the adaptive matrix-transpose algorithm) for transposing matrices, which is based on all-to-all communication...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Proceedings of the Thirty-Seventh Southeastern Symposium on System Theory, 2005. SSST '05 s. 312 - 315
Hlavní autoři: Al Na'mneh, R., Pan, W.D., Adhami, R.
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 2005
Témata:
Bandwidth
Computational modeling
Concurrent computing
Fast Fourier transforms
Hypercubes
Image analysis
Image processing
Message passing
Plasma simulation
Symmetric matrices
ISBN:0780388089, 9780780388086
ISSN:0094-2898
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Computing fast Fourier transform (FFT) on parallel computers has the same communication requirement to transpose matrices one or more times. In this paper, we propose an efficient algorithm (the adaptive matrix-transpose algorithm) for transposing matrices, which is based on all-to-all communication. The adaptive matrix-transpose algorithm is efficient since it minimizes the overhead associated with transposing matrices by adaptively choosing the suitable radix based on data size, number of processors, start-up time, and the effective bandwidth. We study the effect of the adaptive matrix-transpose algorithm on the 6-step 1-D FFT using symmetric multiprocessors (SMP).
ISBN:0780388089
9780780388086
ISSN:0094-2898
DOI:10.1109/SSST.2005.1460928