Efficient Parallelization Strategy for
a High Accuracy Industrial Navier-Stokes Solver

Sergey Peigin
Israel Aircraft Industries,
Department 4473, Engineering Division,
Ben-Gurion Intern. Airport, 70100.

Abstarct:

A highly scalable parallelization strategy for a
high-accuracy 3-dimensional multiblock Navier-Stokes
algorithm is presented.
The parallelization strategy is based on the geometrical
decomposition principle, and on the overlapped communication
and computation concept. The important advantage of the
strategy is that suggested type of message-passing ensures a
very high scalability of the algorithm from network point
of view, because, on average, the communication work per
processor is not increased if the number of processors is
increased. The proposed approch allowed the reduction of
the execution time for large-scale computations employing
10 million of grid points from an estimated 46 days on
the SGI ORIGIN 2000 computer (in the serial mode) to
5-6 hours on 106 processors cluster.