Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Parallelization of a spherical S sub n transport theory algorithm

Conference · · Transactions of the American Nuclear Society; (USA)
OSTI ID:5228706
The work described in this paper derives a parallel algorithm for an R-dependent spherical S{sub N} transport theory algorithm and studies its performance by testing different sample problems. The S{sub N} transport method is one of the most accurate techniques used to solve the linear Boltzmann equation. Several studies have been done on the vectorization of the S{sub N} algorithms; however, very few studies have been performed on the parallelization of this algorithm. Weinke and Hommoto have looked at the parallel processing of the different energy groups, and Azmy recently studied the parallel processing of the inner iterations of an X-Y S{sub N} nodal transport theory method. Both studies have reported very encouraging results, which have prompted us to look at the parallel processing of an R-dependent S{sub N} spherical geometry algorithm. This geometry was chosen because, in spite of its simplicity, it contains the complications of the curvilinear geometries (i.e., redistribution of neutrons over the discretized angular bins).
OSTI ID:
5228706
Report Number(s):
CONF-890604--
Conference Information:
Journal Name: Transactions of the American Nuclear Society; (USA) Journal Volume: 59
Country of Publication:
United States
Language:
English