Multipartite Modular Multiplication

Pascal Giorgi 1 Laurent Imbert 1 Thomas Izard 1
1 ARITH - Arithmétique informatique
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
Abstract : Current processors typically embeds many cores running at high speed. We may then ask ourselves whether software parallelism is practical for low level arithmetic operations. In this paper we generalize the bipartite modular multiplication method of Kaihara and Takagi. We present a multipartite multiplication. We carefully analyze its asymptotic complexity and measure its practical efficiency and scalability for nowadays multi-core architectures. We present several experimental results which illustrate the efficiency of our method and which bring a positive answer to the above question for a wide range of operand's sizes.
Complete list of metadatas

Cited literature [15 references]  Display  Hide  Download

https://hal-lirmm.ccsd.cnrs.fr/lirmm-00618437
Contributor : Pascal Giorgi <>
Submitted on : Thursday, September 1, 2011 - 4:57:53 PM
Last modification on : Wednesday, February 13, 2019 - 5:58:02 PM
Long-term archiving on : Sunday, December 4, 2016 - 12:42:24 PM

File

kpartite-report.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : lirmm-00618437, version 1

Collections

Citation

Pascal Giorgi, Laurent Imbert, Thomas Izard. Multipartite Modular Multiplication. RR-11024, 2011, pp.25. ⟨lirmm-00618437⟩

Share

Metrics

Record views

463

Files downloads

515