=ADD=
=reftype=
14
=number=
01-15
=url=
ftp://ftp.risc.uni-linz.ac.at/pub/techreports/2001/01-15.ps.gz
=year=
2001
=month=
07
=author=
Schreiner; Wolfgang
=title=
Parallelizing the Big Prime Berlekamp Algorithm with Distributed Maple
=abstract=
We report on a parallel implementation of the Big Prime Berlekamp algorithm
with Distributed Maple. This implementation is based on three parallel
subalgorithms with different parallelization strategies and dynamic
behaviors. The experimental results on a cluster and a multiprocessor system
yield superlinear speedups due to a performance problem of the sequential
Maple implementation and demonstrate varying efficiencies for the parallel
subalgorithms.
=note=
=howpublished=
=sponsor=
FWF projects P11160-TEC (HySaX) und SFB F013/F1304
=keywords=
computer algebra, parallel computation, polynomial factorization, cluster, multiprocessor.