| 1463 | | For parallel runs, the total number of processors to be used is given by the '''mrun'''-option -X. By default, depending on the type of the parallel computer, PALM generates a 1d processor net (domain decomposition along x, [#npey npey] = ''1'') or a 2d-net (this is favored on machines with fast communication network and/or large number of processors (>256)). In case of a 2d-net, it is tried to make it more or less square-shaped. If, for example, 16 processors are assigned (-X 16), a 4 * 4 processor net is generated ('''npex''' = 4, '''npey''' = 4). This choice is optimal for square total domains ([#nx nx] = [#ny ny]), since then the number of ghost points at the lateral boundarys of the subdomains reaches a minimum. If nx and ny differ extremely, the processor net should be manually adjusted using adequate values for '''npex''' and '''npey'''.\\\\ |
| | 1463 | For parallel runs, the total number of processors to be used is given by the '''mrun'''-option -X. By default, depending on the type of the parallel computer, PALM generates a 1d processor net (domain decomposition along x, [#npey npey] = ''1'') or a 2d-net (this is favored on machines with fast communication network and/or large number of processors (>256)). In case of a 2d-net, it is tried to make it more or less square-shaped. If, for example, 16 processors are assigned (-X 16), a 4 * 4 processor net is generated ('''npex''' = 4, '''npey''' = 4). This choice is optimal for square total domains ([../inipar#nx nx] = [../inipar#ny ny]), since then the number of ghost points at the lateral boundarys of the subdomains reaches a minimum. If nx and ny differ extremely, the processor net should be manually adjusted using adequate values for '''npex''' and '''npey'''.\\\\ |
