We experiment with the sequential and parallel Karatsuba algorithm under the paclib system on a Sequent Symmetry shared-memory architecture, obtaining efficiency 88% on 9 processors for the controlled-depth version, and 71% on 18 processors for the scalable version. Moreover, we use the Karatsuba algorithm within long integer division, by a recent divide-and-conquer technique. The parallel version exhibits modest speed-ups (2.3 on 3 processors and 3.4 on 9 processors), however the combined speed-up over the classical sequential method is more than 10 at 500 words.
CITATION STYLE
Jebelean, T. (1997). Using the parallel karatsuba algorithm for long integer multiplication and division. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1300 LNCS, pp. 1169–1172). Springer Verlag. https://doi.org/10.1007/bfb0002869
Mendeley helps you to discover research relevant for your work.