PARTISN - 022

PARTISN (Parallel, Time-Dependent S_N) is designed for UNIX-like systems. The specific computers supported fall into two categories - long word and short word. The program has been implemented on the long word Cray J90 and T90 computers. It has also been implemented on SGI, IBM RS6000, HP9000, and DEC Alpha short word workstations. The workstation versions use double precision arithmetic. The program has been run in parallel on clusters of SGI workstations (ASCI Blue Mountain) and the IBM SP₂ (ASCI Blue Pacific). The ASCI Red machine at SNL is also supported. An older version of the program has also been run on the Cray T3D. PARTISN solves the linear Boltzmann transport equation for neutral particles using the deterministic (S_N) method. Both the static (fixed source or eigenvalue) and time-dependent forms of the transport equation are solved in forward or adjoint mode. Vacuum, reflective, periodic, white, or inhomogeneous boundary conditions are solved. General anisotropic scattering and inhomogeneous sources are permitted. PARTISN solves the transport equation on orthogonal (single level or block-structured AMR) grids in 1-D (slab, two-angle slab, cylindrical, or spherical), 2-D (X-Y, R-Z, or R-T) and 3-D (X-Y-Z or R-Z-T) geometries. PARTISN numerically solves the multigroup form of the neutral-particle Boltzmann transport equation. The discrete-ordinates form of approximation is used for treating the angular variation of the particle distribution. For curvilinear geometries, diamond differencing is used for angular discretization. The spatial discretizations may be either low-order (diamond difference or Adaptive Weighted Diamond Difference (AWDD)) or higher-order (linear discontinuous or exponential discontinuous). Negative fluxes are eliminated by a local set-to-zero and-correct algorithm for the diamond case (DD/STZ). Time differencing is Crank-Nicholson (diamond), also with a set-to-zero fix up scheme. Both inner and outer iterations are accelerated using the diffusion synthetic acceleration method. The diffusion solver uses the conjugate gradient or multigrid method and Chebyshev acceleration of the fission source. First-collision source treatment options are provided for the elimination of primary ray effects in fixed-source calculations. The angular source terms may be treated either via standard PN expansions or Galerkin scattering. An option is provided for strictly positive scattering sources. Parallelization is performed via a 2D spatial decomposition, which retains the ability to invert the source iteration equation in a single sweep. The code is thoroughly variably dimensioned, with memory requirements determined from the input parameters. At present, no out-of-core (i.e., disk) storage capability is provided. Running time on a single processor is directly related to problem size and to central processor and data transfer speeds. On a SGI R10000, a four-group eigenvalue calculation of an X-Y-Z model of the Fast Test Reactor (FTR) took 9 seconds. The calculation used transport corrected P0 cross sections, an S8 angular quadrature, DD/STZ spatial differencing, and a 14x14x30 spatial mesh. Running time on parallel platforms is sensitive to the latency and topology of the interconnect, as well as the single processor performance. PARTISN is modularly structured in a form that separates the input and output (edit) functions from the main calculational (solver) section of the code. The code makes use of binary, sequential data files, called interface files, to transfer data between modules. Standard interface files whose specifications have been defined by the Reactor Physics Committee on Computer Code Coordination are accepted, used, and created by the code. A free-field card-image input capability is provided for the user. The code provides the user with considerable flexibility in using both card-image or sequential file input and in controlling the execution of modules. PARTISN is the evolutionary successor to DANTSYS. User input is very similar to that of the DANTSYS code