Use -pc_type lu -pc_factor_mat_solver_package superlu_dist to us this direct solver
Works with AIJ matrices
| -mat_superlu_dist_r <n> | - number of rows in processor partition | |
| -mat_superlu_dist_c <n> | - number of columns in processor partition | |
| -mat_superlu_dist_matinput <0,1> | - matrix input mode; 0=global, 1=distributed | |
| -mat_superlu_dist_equil | - equilibrate the matrix | |
| -mat_superlu_dist_rowperm <LargeDiag,NATURAL> | - row permutation | |
| -mat_superlu_dist_colperm <MMD_AT_PLUS_A,MMD_ATA,NATURAL> | - column permutation | |
| -mat_superlu_dist_replacetinypivot | - replace tiny pivots | |
| -mat_superlu_dist_fact <SamePattern> | - (choose one of) SamePattern SamePattern_SameRowPerm | |
| -mat_superlu_dist_iterrefine | - use iterative refinement | |
| -mat_superlu_dist_statprint | - print factorization information |
Level:beginner
Location:src/mat/impls/aij/mpi/superlu_dist/superlu_dist.c
Index of all Mat routines
Table of Contents for all manual pages
Index of all manual pages