wiki:AthenaDocsUGDomainBlck

Version 3 (modified by jmstone, 4 years ago) (diff)

--

The <Domain> Block(s)

Wiki/Documentation/UserGuide/Domain Block(s)

Parameters in these blocks control the properties of the Domains, and with MPI how each Domain is decomposed into Grids. Without SMR, there should be only one Domain block, and it must have level=0. With SMR, there should be one block for each Domain at each refinement level, and there must be one level=0 Domain (see the User Guide to SMR? for details of Domains and Grids within the Mesh).

With MPI, each Domain can be decomposed in any coordinate direction, and into an arbitrary number of Grids. This allows slab, pencil, and block decompositions. Different Domains can have different decompositions, and the same processors can be used to update Grids on different Domains. Athena can also automatically figure out the optimal domain decomposition that minimizes the amount of data communicated; this option can be chosen with AutoWithNProc parameter in each Domain block. Again, see the User Guide to SMR? for details of parallelization of Domains with SMR.

Example:

<domain1>
level           = 0         # refinement level this Domain (root=0)
Nx1             = 8         # Number of zones in X-direction
x1min           = -0.5      # minimum value of X
x1max           = 0.5       # maximum value of X
bc_ix1          = 4         # boundary condition flag for inner-I (X1)
bc_ox1          = 4         # boundary condition flag for outer-I (X1)
NGrid_x1        = 1         # with MPI, number of Grids in X1 coordinate
AutoWithNProc   = 0         # set to Nproc for auto domain decomposition

Nx2             = 64        # Number of zones in Y-direction
x2min           = -4.0      # minimum value of Y
x2max           = 4.0       # maximum value of Y
bc_ix2          = 4         # boundary condition flag for inner-J (X2)
bc_ox2          = 4         # boundary condition flag for outer-J (X2)
NGrid_x2        = 1         # with MPI, number of Grids in X2 coordinate
AutoWithNProc   = 0         # set to Nproc for auto domain decomposition

Nx3             = 96        # Number of zones in X3-direction
x3min           = -6.0      # minimum value of X3
x3max           = 6.0       # maximum value of X3
bc_ix3          = 4         # boundary condition flag for inner-K (X3)
bc_ox3          = 4         # boundary condition flag for outer-K (X3)
NGrid_x3        = 1         # with MPI, number of Grids in X3 coordinate
AutoWithNProc   = 0         # set to Nproc for auto domain decomposition

<domain2>
level           = 1         # refinement level this Domain (root=0)
Nx1             = 16        # Number of zones in X1-direction
Nx2             = 128       # Number of zones in X2-direction
Nx3             = 96        # Number of zones in X3-direction
iDisp           = 0         # i-displacement measured in cells of this level
jDisp           = 0         # j-displacement measured in cells of this level
kDisp           = 48        # k-displacement measured in cells of this level
AutoWithNProc   = 0         # set to Nproc for auto domain decomposition
NGrid_x1        = 1         # with MPI, number of Grids in X1 coordinate
NGrid_x2        = 1         # with MPI, number of Grids in X2 coordinate
NGrid_x3        = 1         # with MPI, number of Grids in X3 coordinate

level: Refinement level of this Domain. There must be one and only one root (level=0) Domain block. It does not need to be given first.

Nx1, Nx2, Nx3: number of grid cells in the x1-, x2-, and x3-directions in each Domain.

x1min, x2min, x3min: x1-, x2-, x3-coordinate of left-edge of first cell for the root (level=0) Domain. Only specified for root level.

x1max, x2max, x3max: x1-, x2-, x3-coordinate of right-edge of last cell for the root (level=0) Domain. The root level domain in the x1-direction spans x1max-x1min, the grid spacing is dx1=(x1max-x1min)/Nx1, and the center of the first cell is located at x1=x1min + dx1/2. Also, x1max > x1min is required. Similarly for the x2- and x3-directions. Only specified for root level.

bc_ix1, bc_ox1: integer flags for boundary conditions applied at inner (left) and outer (right) edges of grid. Currently three values are implemented: 1 = reflecting, 2 = outflow (projection), and 4 = periodic. See Boundary Conditions for more information. Similarly for the x2- and x3-directions.

iDisp, jDisp, kDisp: Offset of origin of Domain from origin of root level. Given in units of cells at this level. Only specified for levels>0.

NGrid_x1, NGrid_x2, NGrid_x3: Number of MPI blocks in the x1-, x2-, x3-directions in this Domain.

AutoWithNProc: Set to number of processers desired for this Domain for automatic domain decomposition that optimizes efficiency. In this case, the NGrid_x1, etc. values are ignored.