1DM Object: Distributed Mesh 5 MPI processes 2 type: plex 3Distributed Mesh in 2 dimensions: 4 0-cells: 29 29 30 15 15 5 1-cells: 60 60 60 30 30 6 2-cells: 32 32 32 16 16 7Labels: 8 celltype: 3 strata with value/size (1 (60), 3 (32), 0 (29)) 9 depth: 3 strata with value/size (0 (29), 1 (60), 2 (32)) 10 marker: 2 strata with value/size (1 (8), 4 (17)) 11 Face Sets: 2 strata with value/size (1 (7), 4 (14)) 12 markerBottom: 1 strata with value/size (1 (8)) 13 markerRight: 0 strata with value/size () 14 markerTop: 0 strata with value/size () 15 markerLeft: 1 strata with value/size (1 (17)) 16Integral of pressure: 0. 17SNES Object: 5 MPI processes 18 type: newtonls 19 tolerances: relative=1e-08, absolute=1e-50, solution=1e-08 20 norm schedule ALWAYS 21 SNESLineSearch Object: 5 MPI processes 22 type: bt 23 interpolation: cubic 24 alpha=1.000000e-04 25 maxstep=1.000000e+08, minlambda=1.000000e-12 26 tolerances: relative=1.000000e-08, absolute=1.000000e-15, lambda=1.000000e-08 27 maximum iterations=40 28 KSP Object: 5 MPI processes 29 type: fetidp 30 fully redundant: 0 31 saddle point: 1 32 Inner KSP solver details 33 KSP Object: (fetidp_) 5 MPI processes 34 type: cg 35 maximum iterations=10000, initial guess is zero 36 tolerances: relative=1e-12, absolute=1e-10, divergence=10000. 37 left preconditioning 38 using NATURAL norm type for convergence test 39 PC Object: (fetidp_) 5 MPI processes 40 type: fieldsplit 41 FieldSplit with Schur preconditioner, factorization DIAG 42 Preconditioner for the Schur complement formed from user provided matrix 43 Split info: 44 Split number 0 Defined by IS 45 Split number 1 Defined by IS 46 KSP solver for A00 block 47 KSP Object: (fetidp_fieldsplit_lag_) 5 MPI processes 48 type: preonly 49 maximum iterations=10000, initial guess is zero 50 tolerances: relative=1e-05, absolute=1e-50, divergence=10000. 51 left preconditioning 52 using NONE norm type for convergence test 53 PC Object: (fetidp_fieldsplit_lag_) 5 MPI processes 54 type: shell 55 FETI-DP multipliers 56 Dirichlet preconditioner (just from rank 0) 57 Mat Object: 1 MPI processes 58 type: schurcomplement 59 rows=59, cols=59 60 Schur complement A11 - A10 inv(A00) A01 61 A11 62 Mat Object: 1 MPI processes 63 type: seqaij 64 rows=59, cols=59 65 total: nonzeros=651, allocated nonzeros=651 66 total number of mallocs used during MatSetValues calls=0 67 using I-node routines: found 25 nodes, limit used is 5 68 A10 69 Mat Object: 1 MPI processes 70 type: seqaij 71 rows=59, cols=105 72 total: nonzeros=476, allocated nonzeros=476 73 total number of mallocs used during MatSetValues calls=0 74 using I-node routines: found 25 nodes, limit used is 5 75 KSP of A00 76 KSP Object: (fetidp_harmonic_) 1 MPI processes 77 type: preonly 78 maximum iterations=10000, initial guess is zero 79 tolerances: relative=1e-05, absolute=1e-50, divergence=10000. 80 left preconditioning 81 using NONE norm type for convergence test 82 PC Object: (fetidp_harmonic_) 1 MPI processes 83 type: cholesky 84 out-of-place factorization 85 tolerance for zero pivot 2.22045e-14 86 matrix ordering: nd 87 factor fill ratio given 5., needed 1.34037 88 Factored matrix follows: 89 Mat Object: 1 MPI processes 90 type: seqsbaij 91 rows=105, cols=105 92 package used to perform factorization: petsc 93 total: nonzeros=1079, allocated nonzeros=1079 94 total number of mallocs used during MatSetValues calls=0 95 block size is 1 96 linear system matrix = precond matrix: 97 Mat Object: 1 MPI processes 98 type: seqaij 99 rows=105, cols=105 100 total: nonzeros=1505, allocated nonzeros=1505 101 total number of mallocs used during MatSetValues calls=0 102 using I-node routines: found 63 nodes, limit used is 5 103 A01 104 Mat Object: 1 MPI processes 105 type: seqaij 106 rows=105, cols=59 107 total: nonzeros=476, allocated nonzeros=476 108 total number of mallocs used during MatSetValues calls=0 109 using I-node routines: found 52 nodes, limit used is 5 110 linear system matrix = precond matrix: 111 Mat Object: (fetidp_fieldsplit_lag_) 5 MPI processes 112 type: submatrix 113 rows=112, cols=112 114 KSP solver for S = A11 - A10 inv(A00) A01 115 KSP Object: (fetidp_fieldsplit_p_) 5 MPI processes 116 type: preonly 117 maximum iterations=10000, initial guess is zero 118 tolerances: relative=1e-05, absolute=1e-50, divergence=10000. 119 left preconditioning 120 using NONE norm type for convergence test 121 PC Object: (fetidp_fieldsplit_p_) 5 MPI processes 122 type: jacobi 123 linear system matrix followed by preconditioner matrix: 124 Mat Object: (fetidp_fieldsplit_p_) 5 MPI processes 125 type: schurcomplement 126 rows=31, cols=31 127 Schur complement A11 - A10 inv(A00) A01 128 A11 129 Mat Object: (fetidp_fieldsplit_p_) 5 MPI processes 130 type: submatrix 131 rows=31, cols=31 132 A10 133 Mat Object: 5 MPI processes 134 type: submatrix 135 rows=31, cols=112 136 KSP of A00 137 KSP Object: (fetidp_fieldsplit_lag_) 5 MPI processes 138 type: preonly 139 maximum iterations=10000, initial guess is zero 140 tolerances: relative=1e-05, absolute=1e-50, divergence=10000. 141 left preconditioning 142 using NONE norm type for convergence test 143 PC Object: (fetidp_fieldsplit_lag_) 5 MPI processes 144 type: shell 145 FETI-DP multipliers 146 Dirichlet preconditioner (just from rank 0) 147 Mat Object: 1 MPI processes 148 type: schurcomplement 149 rows=59, cols=59 150 Schur complement A11 - A10 inv(A00) A01 151 A11 152 Mat Object: 1 MPI processes 153 type: seqaij 154 rows=59, cols=59 155 total: nonzeros=651, allocated nonzeros=651 156 total number of mallocs used during MatSetValues calls=0 157 using I-node routines: found 25 nodes, limit used is 5 158 A10 159 Mat Object: 1 MPI processes 160 type: seqaij 161 rows=59, cols=105 162 total: nonzeros=476, allocated nonzeros=476 163 total number of mallocs used during MatSetValues calls=0 164 using I-node routines: found 25 nodes, limit used is 5 165 KSP of A00 166 KSP Object: (fetidp_harmonic_) 1 MPI processes 167 type: preonly 168 maximum iterations=10000, initial guess is zero 169 tolerances: relative=1e-05, absolute=1e-50, divergence=10000. 170 left preconditioning 171 using NONE norm type for convergence test 172 PC Object: (fetidp_harmonic_) 1 MPI processes 173 type: cholesky 174 out-of-place factorization 175 tolerance for zero pivot 2.22045e-14 176 matrix ordering: nd 177 factor fill ratio given 5., needed 1.34037 178 Factored matrix follows: 179 Mat Object: 1 MPI processes 180 type: seqsbaij 181 rows=105, cols=105 182 package used to perform factorization: petsc 183 total: nonzeros=1079, allocated nonzeros=1079 184 total number of mallocs used during MatSetValues calls=0 185 block size is 1 186 linear system matrix = precond matrix: 187 Mat Object: 1 MPI processes 188 type: seqaij 189 rows=105, cols=105 190 total: nonzeros=1505, allocated nonzeros=1505 191 total number of mallocs used during MatSetValues calls=0 192 using I-node routines: found 63 nodes, limit used is 5 193 A01 194 Mat Object: 1 MPI processes 195 type: seqaij 196 rows=105, cols=59 197 total: nonzeros=476, allocated nonzeros=476 198 total number of mallocs used during MatSetValues calls=0 199 using I-node routines: found 52 nodes, limit used is 5 200 linear system matrix = precond matrix: 201 Mat Object: (fetidp_fieldsplit_lag_) 5 MPI processes 202 type: submatrix 203 rows=112, cols=112 204 A01 205 Mat Object: 5 MPI processes 206 type: submatrix 207 rows=112, cols=31 208 Mat Object: 5 MPI processes 209 type: is 210 rows=31, cols=31 211 total: nonzeros=216, allocated nonzeros=216 212 total number of mallocs used during MatSetValues calls=0 213 linear system matrix = precond matrix: 214 Mat Object: G (fetidp_) 5 MPI processes 215 type: shell 216 rows=143, cols=143 217 Inner BDDC solver details 218 PC Object: (fetidp_bddc_) 5 MPI processes 219 type: bddc 220 Use verbose output: 0 221 Use user-defined CSR: 0 222 Use local mat graph: 1 223 Connectivity graph topological dimension: 2 224 Graph max count: 2 225 Use vertices: 1 (vertex size 3) 226 Use edges: 1 227 Use faces: 0 228 Use true near null space: 0 229 Use QR for single constraints on cc: 0 230 Use change of basis on local edge nodes: 0 231 Use change of basis on local face nodes: 0 232 User defined change of basis matrix: 0 233 Has change of basis matrix: 0 234 Eliminate dirichlet boundary dofs: 1 235 Switch on static condensation ops around the interface preconditioner: 1 236 Use exact dirichlet trick: 0 237 Interface extension: DIRICHLET 238 Multilevel max levels: 0 239 Multilevel coarsening ratio: 8 240 Use estimated eigs for coarse problem: 0 241 Use deluxe scaling: 0 242 Use deluxe zerorows: 0 243 Use deluxe singlemat: 0 244 Rebuild interface graph for Schur principal minors: 0 245 Number of dofs' layers for the computation of principal minors: -1 246 Use user CSR graph to compute successive layers: 0 247 Adaptive constraint selection threshold (active 0, userdefined 0): 0. 248 Min constraints / connected component: 0 249 Max constraints / connected component: 0 250 Invert exact Schur complement for adaptive selection: 0 251 Symmetric computation of primal basis functions: 1 252 Num. Procs. to map coarse adjacency list: 0 253 Coarse eqs per proc (significant at the coarsest level): 1 254 Detect disconnected: 1 (filter 0) 255 Benign subspace trick: 0 (change explicit 0) 256 Benign subspace trick is active: 0 257 Algebraic computation of no-net-flux: 0 258 ********************************** STATISTICS AT LEVEL 0 ********************************** 259 Global dofs sizes: all 591 interface 149 coarse 6 260 Coarsening ratios: all/coarse 98 interface/coarse 24 261 Active processes : 5 262 Total subdomains : 8 263 Dofs type : MIN MAX MEAN 264 Interior dofs : 45 121 88 265 Interface dofs : 40 95 62 266 Primal dofs : 3 5 4 267 Local dofs : 95 195 151 268 Local subs : 1 2 269 --- Interior solver (rank 0) 270 KSP Object: (fetidp_bddc_pc_bddc_dirichlet_) 1 MPI processes 271 type: preonly 272 maximum iterations=10000, initial guess is zero 273 tolerances: relative=1e-05, absolute=1e-50, divergence=10000. 274 left preconditioning 275 using NONE norm type for convergence test 276 PC Object: (fetidp_bddc_pc_bddc_dirichlet_) 1 MPI processes 277 type: none 278 linear system matrix = precond matrix: 279 Mat Object: (fetidp_bddc_pc_bddc_dirichlet_) 1 MPI processes 280 type: seqaij 281 rows=121, cols=121 282 total: nonzeros=2225, allocated nonzeros=2225 283 total number of mallocs used during MatSetValues calls=0 284 using I-node routines: found 64 nodes, limit used is 5 285 --- Correction solver (rank 0) 286 KSP Object: (fetidp_bddc_pc_bddc_neumann_) 1 MPI processes 287 type: preonly 288 maximum iterations=1000, initial guess is zero 289 tolerances: relative=1e-10, absolute=1e-10, divergence=10000. 290 left preconditioning 291 using NONE norm type for convergence test 292 PC Object: (fetidp_bddc_pc_bddc_neumann_) 1 MPI processes 293 type: svd 294 All singular values smaller than 1e-12 treated as zero 295 Provided essential rank of the matrix 0 (all other eigenvalues are zeroed) 296 linear system matrix = precond matrix: 297 Mat Object: (fetidp_bddc_pc_bddc_neumann_) 1 MPI processes 298 type: seqaij 299 rows=176, cols=176 300 total: nonzeros=3856, allocated nonzeros=3856 301 total number of mallocs used during MatSetValues calls=0 302 using I-node routines: found 85 nodes, limit used is 5 303 --- Coarse solver 304 KSP Object: (fetidp_bddc_pc_bddc_coarse_) 5 MPI processes 305 type: preonly 306 maximum iterations=1000, initial guess is zero 307 tolerances: relative=1e-10, absolute=1e-10, divergence=10000. 308 left preconditioning 309 using NONE norm type for convergence test 310 PC Object: (fetidp_bddc_pc_bddc_coarse_) 5 MPI processes 311 type: redundant 312 First (color=0) of 5 PCs follows 313 linear system matrix = precond matrix: 314 Mat Object: (fetidp_bddc_pc_bddc_coarse_) 5 MPI processes 315 type: mpiaij 316 rows=6, cols=6 317 total: nonzeros=34, allocated nonzeros=36 318 total number of mallocs used during MatSetValues calls=0 319 using I-node (on process 0) routines: found 1 nodes, limit used is 5 320 KSP Object: (fetidp_bddc_pc_bddc_coarse_redundant_) 1 MPI processes 321 type: preonly 322 maximum iterations=10000, initial guess is zero 323 tolerances: relative=1e-05, absolute=1e-50, divergence=10000. 324 left preconditioning 325 using NONE norm type for convergence test 326 PC Object: (fetidp_bddc_pc_bddc_coarse_redundant_) 1 MPI processes 327 type: cholesky 328 out-of-place factorization 329 tolerance for zero pivot 2.22045e-14 330 matrix ordering: nd 331 factor fill ratio given 5., needed 1. 332 Factored matrix follows: 333 Mat Object: 1 MPI processes 334 type: seqsbaij 335 rows=6, cols=6 336 package used to perform factorization: petsc 337 total: nonzeros=20, allocated nonzeros=20 338 total number of mallocs used during MatSetValues calls=0 339 block size is 1 340 linear system matrix = precond matrix: 341 Mat Object: 1 MPI processes 342 type: seqaij 343 rows=6, cols=6 344 total: nonzeros=34, allocated nonzeros=34 345 total number of mallocs used during MatSetValues calls=0 346 using I-node routines: found 3 nodes, limit used is 5 347 linear system matrix = precond matrix: 348 Mat Object: 5 MPI processes 349 type: is 350 rows=591, cols=591 351 total: nonzeros=17047, allocated nonzeros=17047 352 total number of mallocs used during MatSetValues calls=0 353 has attached near null space 354 maximum iterations=10000, initial guess is zero 355 tolerances: relative=1e-12, absolute=1e-10, divergence=10000. 356 left preconditioning 357 using NONE norm type for convergence test 358 PC Object: 5 MPI processes 359 type: none 360 linear system matrix followed by preconditioner matrix: 361 Mat Object: 5 MPI processes 362 type: is 363 rows=591, cols=591 364 total: nonzeros=17047, allocated nonzeros=20789 365 total number of mallocs used during MatSetValues calls=0 366 has attached null space 367 Mat Object: 5 MPI processes 368 type: is 369 rows=591, cols=591 370 total: nonzeros=17047, allocated nonzeros=20789 371 total number of mallocs used during MatSetValues calls=0 372L_2 Error: 0.00151 [9.32e-06, 0.00151] 373Integral of pressure: 0. 374