Evaluation of Parallel Computations of Gr¨obner and Involutive Bases on the Massive SMP Computer

Cover Page

Abstract


In previous papers author presented realizations of two different approaches to parallelization of computation of Gr¨obner and involutive bases of polynomial systems with benchmarking on the 8-cores SMP computer: reduction-level parallelism with coefficients of polynomials in Z-ring and basis-level parallelism using modular basis computation and lifting. In this work further development of this algorithms described, benchmarking results and maximal speedup achieved on the massive 32-cores computer presented, scalability differences of algorithms investigated.

D A Yanovich

Joint Institute for Nuclear Research

Email: yan@jinr.ru
Laboratory of Information Technologies

Views

Abstract - 63

PDF (Russian) - 29


Copyright (c) 2014 Янович Д.А.

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.