xref: /petsc/src/snes/tutorials/output/ex12_tri_bddc.out (revision c4762a1b19cd2af06abeed90e8f9d34fb975dd94)
1*c4762a1bSJed Brown  0 SNES Function norm 5.83179
2*c4762a1bSJed Brown    0 KSP Residual norm 5.11122
3*c4762a1bSJed Brown    1 KSP Residual norm 0.0792098
4*c4762a1bSJed Brown    2 KSP Residual norm 0.0036192
5*c4762a1bSJed Brown    3 KSP Residual norm 0.000168263
6*c4762a1bSJed Brown    4 KSP Residual norm 1.59554e-06
7*c4762a1bSJed Brown    5 KSP Residual norm 2.40291e-08
8*c4762a1bSJed Brown    6 KSP Residual norm 1.023e-10
9*c4762a1bSJed Brown  Linear solve converged due to CONVERGED_RTOL iterations 6
10*c4762a1bSJed Brown  1 SNES Function norm 4.026e-10
11*c4762a1bSJed BrownL_2 Error: 0.00527027
12*c4762a1bSJed BrownNonlinear solve converged due to CONVERGED_FNORM_RELATIVE iterations 1
13*c4762a1bSJed BrownSNES Object: 5 MPI processes
14*c4762a1bSJed Brown  type: newtonls
15*c4762a1bSJed Brown  maximum iterations=50, maximum function evaluations=10000
16*c4762a1bSJed Brown  tolerances: relative=1e-08, absolute=1e-50, solution=1e-08
17*c4762a1bSJed Brown  total number of linear solver iterations=6
18*c4762a1bSJed Brown  total number of function evaluations=2
19*c4762a1bSJed Brown  norm schedule ALWAYS
20*c4762a1bSJed Brown  SNESLineSearch Object: 5 MPI processes
21*c4762a1bSJed Brown    type: bt
22*c4762a1bSJed Brown      interpolation: cubic
23*c4762a1bSJed Brown      alpha=1.000000e-04
24*c4762a1bSJed Brown    maxstep=1.000000e+08, minlambda=1.000000e-12
25*c4762a1bSJed Brown    tolerances: relative=1.000000e-08, absolute=1.000000e-15, lambda=1.000000e-08
26*c4762a1bSJed Brown    maximum iterations=40
27*c4762a1bSJed Brown  KSP Object: 5 MPI processes
28*c4762a1bSJed Brown    type: gmres
29*c4762a1bSJed Brown      restart=100, using Classical (unmodified) Gram-Schmidt Orthogonalization with no iterative refinement
30*c4762a1bSJed Brown      happy breakdown tolerance 1e-30
31*c4762a1bSJed Brown    maximum iterations=10000, initial guess is zero
32*c4762a1bSJed Brown    tolerances:  relative=1e-09, absolute=1e-50, divergence=10000.
33*c4762a1bSJed Brown    left preconditioning
34*c4762a1bSJed Brown    using PRECONDITIONED norm type for convergence test
35*c4762a1bSJed Brown  PC Object: 5 MPI processes
36*c4762a1bSJed Brown    type: bddc
37*c4762a1bSJed Brown      Use verbose output: 0
38*c4762a1bSJed Brown      Use user-defined CSR: 0
39*c4762a1bSJed Brown      Use local mat graph: 1
40*c4762a1bSJed Brown      Connectivity graph topological dimension: 2
41*c4762a1bSJed Brown      Use vertices: 1 (vertex size 1)
42*c4762a1bSJed Brown      Use edges: 1
43*c4762a1bSJed Brown      Use faces: 0
44*c4762a1bSJed Brown      Use true near null space: 0
45*c4762a1bSJed Brown      Use QR for single constraints on cc: 0
46*c4762a1bSJed Brown      Use change of basis on local edge nodes: 0
47*c4762a1bSJed Brown      Use change of basis on local face nodes: 0
48*c4762a1bSJed Brown      User defined change of basis matrix: 0
49*c4762a1bSJed Brown      Has change of basis matrix: 0
50*c4762a1bSJed Brown      Eliminate dirichlet boundary dofs: 0
51*c4762a1bSJed Brown      Switch on static condensation ops around the interface preconditioner: 0
52*c4762a1bSJed Brown      Use exact dirichlet trick: 0
53*c4762a1bSJed Brown      Interface extension: DIRICHLET
54*c4762a1bSJed Brown      Multilevel max levels: 0
55*c4762a1bSJed Brown      Multilevel coarsening ratio: 8
56*c4762a1bSJed Brown      Use estimated eigs for coarse problem: 0
57*c4762a1bSJed Brown      Use deluxe scaling: 0
58*c4762a1bSJed Brown      Use deluxe zerorows: 0
59*c4762a1bSJed Brown      Use deluxe singlemat: 0
60*c4762a1bSJed Brown      Rebuild interface graph for Schur principal minors: 0
61*c4762a1bSJed Brown      Number of dofs' layers for the computation of principal minors: -1
62*c4762a1bSJed Brown      Use user CSR graph to compute successive layers: 0
63*c4762a1bSJed Brown      Adaptive constraint selection threshold (active 0, userdefined 0): 0.
64*c4762a1bSJed Brown      Min constraints / connected component: 0
65*c4762a1bSJed Brown      Max constraints / connected component: 0
66*c4762a1bSJed Brown      Invert exact Schur complement for adaptive selection: 0
67*c4762a1bSJed Brown      Symmetric computation of primal basis functions: 1
68*c4762a1bSJed Brown      Num. Procs. to map coarse adjacency list: 0
69*c4762a1bSJed Brown      Coarse eqs per proc (significant at the coarsest level): 1
70*c4762a1bSJed Brown      Detect disconnected: 0 (filter 0)
71*c4762a1bSJed Brown      Benign subspace trick: 0 (change explicit 0)
72*c4762a1bSJed Brown      Benign subspace trick is active: 0
73*c4762a1bSJed Brown      Algebraic computation of no-net-flux: 0
74*c4762a1bSJed Brown    ********************************** STATISTICS AT LEVEL 0 **********************************
75*c4762a1bSJed Brown      Global dofs sizes: all 49 interface 25 coarse 9
76*c4762a1bSJed Brown      Coarsening ratios: all/coarse 5 interface/coarse 2
77*c4762a1bSJed Brown      Active processes : 5
78*c4762a1bSJed Brown      Total subdomains : 5
79*c4762a1bSJed Brown      Dofs type        :	MIN	MAX	MEAN
80*c4762a1bSJed Brown      Interior  dofs   :	3	6	4
81*c4762a1bSJed Brown      Interface dofs   :	7	16	10
82*c4762a1bSJed Brown      Primal    dofs   :	3	6	4
83*c4762a1bSJed Brown      Local     dofs   :	10	22	15
84*c4762a1bSJed Brown      Local     subs   :	1	1
85*c4762a1bSJed Brown    --- Interior solver (rank 0)
86*c4762a1bSJed Brown      KSP Object: (pc_bddc_dirichlet_) 1 MPI processes
87*c4762a1bSJed Brown        type: preonly
88*c4762a1bSJed Brown        maximum iterations=10000, initial guess is zero
89*c4762a1bSJed Brown        tolerances:  relative=1e-05, absolute=1e-50, divergence=10000.
90*c4762a1bSJed Brown        left preconditioning
91*c4762a1bSJed Brown        using NONE norm type for convergence test
92*c4762a1bSJed Brown      PC Object: (pc_bddc_dirichlet_) 1 MPI processes
93*c4762a1bSJed Brown        type: lu
94*c4762a1bSJed Brown          out-of-place factorization
95*c4762a1bSJed Brown          tolerance for zero pivot 2.22045e-14
96*c4762a1bSJed Brown          matrix ordering: nd
97*c4762a1bSJed Brown          factor fill ratio given 5., needed 1.
98*c4762a1bSJed Brown            Factored matrix follows:
99*c4762a1bSJed Brown              Mat Object: 1 MPI processes
100*c4762a1bSJed Brown                type: seqaij
101*c4762a1bSJed Brown                rows=6, cols=6
102*c4762a1bSJed Brown                package used to perform factorization: petsc
103*c4762a1bSJed Brown                total: nonzeros=18, allocated nonzeros=18
104*c4762a1bSJed Brown                total number of mallocs used during MatSetValues calls=0
105*c4762a1bSJed Brown                  using I-node routines: found 2 nodes, limit used is 5
106*c4762a1bSJed Brown        linear system matrix = precond matrix:
107*c4762a1bSJed Brown        Mat Object: (pc_bddc_dirichlet_) 1 MPI processes
108*c4762a1bSJed Brown          type: seqaij
109*c4762a1bSJed Brown          rows=6, cols=6
110*c4762a1bSJed Brown          total: nonzeros=18, allocated nonzeros=18
111*c4762a1bSJed Brown          total number of mallocs used during MatSetValues calls=0
112*c4762a1bSJed Brown            using I-node routines: found 2 nodes, limit used is 5
113*c4762a1bSJed Brown    --- Correction solver (rank 0)
114*c4762a1bSJed Brown      KSP Object: (pc_bddc_neumann_) 1 MPI processes
115*c4762a1bSJed Brown        type: preonly
116*c4762a1bSJed Brown        maximum iterations=10000, initial guess is zero
117*c4762a1bSJed Brown        tolerances:  relative=1e-05, absolute=1e-50, divergence=10000.
118*c4762a1bSJed Brown        left preconditioning
119*c4762a1bSJed Brown        using NONE norm type for convergence test
120*c4762a1bSJed Brown      PC Object: (pc_bddc_neumann_) 1 MPI processes
121*c4762a1bSJed Brown        type: lu
122*c4762a1bSJed Brown          out-of-place factorization
123*c4762a1bSJed Brown          tolerance for zero pivot 2.22045e-14
124*c4762a1bSJed Brown          matrix ordering: nd
125*c4762a1bSJed Brown          factor fill ratio given 5., needed 1.18462
126*c4762a1bSJed Brown            Factored matrix follows:
127*c4762a1bSJed Brown              Mat Object: 1 MPI processes
128*c4762a1bSJed Brown                type: seqaij
129*c4762a1bSJed Brown                rows=15, cols=15
130*c4762a1bSJed Brown                package used to perform factorization: petsc
131*c4762a1bSJed Brown                total: nonzeros=77, allocated nonzeros=77
132*c4762a1bSJed Brown                total number of mallocs used during MatSetValues calls=0
133*c4762a1bSJed Brown                  not using I-node routines
134*c4762a1bSJed Brown        linear system matrix = precond matrix:
135*c4762a1bSJed Brown        Mat Object: (pc_bddc_neumann_) 1 MPI processes
136*c4762a1bSJed Brown          type: seqaij
137*c4762a1bSJed Brown          rows=15, cols=15
138*c4762a1bSJed Brown          total: nonzeros=65, allocated nonzeros=65
139*c4762a1bSJed Brown          total number of mallocs used during MatSetValues calls=0
140*c4762a1bSJed Brown            not using I-node routines
141*c4762a1bSJed Brown    --- Coarse solver
142*c4762a1bSJed Brown      KSP Object: (pc_bddc_coarse_) 5 MPI processes
143*c4762a1bSJed Brown        type: preonly
144*c4762a1bSJed Brown        maximum iterations=1, initial guess is zero
145*c4762a1bSJed Brown        tolerances:  relative=1e-05, absolute=1e-50, divergence=10000.
146*c4762a1bSJed Brown        left preconditioning
147*c4762a1bSJed Brown        using NONE norm type for convergence test
148*c4762a1bSJed Brown      PC Object: (pc_bddc_coarse_) 5 MPI processes
149*c4762a1bSJed Brown        type: redundant
150*c4762a1bSJed Brown          First (color=0) of 5 PCs follows
151*c4762a1bSJed Brown        linear system matrix = precond matrix:
152*c4762a1bSJed Brown        Mat Object: (pc_bddc_coarse_) 5 MPI processes
153*c4762a1bSJed Brown          type: mpiaij
154*c4762a1bSJed Brown          rows=9, cols=9
155*c4762a1bSJed Brown          total: nonzeros=61, allocated nonzeros=68
156*c4762a1bSJed Brown          total number of mallocs used during MatSetValues calls=0
157*c4762a1bSJed Brown            using I-node (on process 0) routines: found 1 nodes, limit used is 5
158*c4762a1bSJed Brown              KSP Object:     (pc_bddc_coarse_redundant_)     1 MPI processes
159*c4762a1bSJed Brown                type: preonly
160*c4762a1bSJed Brown                maximum iterations=10000, initial guess is zero
161*c4762a1bSJed Brown                tolerances:  relative=1e-05, absolute=1e-50, divergence=10000.
162*c4762a1bSJed Brown                left preconditioning
163*c4762a1bSJed Brown                using NONE norm type for convergence test
164*c4762a1bSJed Brown              PC Object:     (pc_bddc_coarse_redundant_)     1 MPI processes
165*c4762a1bSJed Brown                type: lu
166*c4762a1bSJed Brown                  out-of-place factorization
167*c4762a1bSJed Brown                  tolerance for zero pivot 2.22045e-14
168*c4762a1bSJed Brown                  matrix ordering: nd
169*c4762a1bSJed Brown                  factor fill ratio given 5., needed 1.13115
170*c4762a1bSJed Brown                    Factored matrix follows:
171*c4762a1bSJed Brown                      Mat Object:     1 MPI processes
172*c4762a1bSJed Brown                        type: seqaij
173*c4762a1bSJed Brown                        rows=9, cols=9
174*c4762a1bSJed Brown                        package used to perform factorization: petsc
175*c4762a1bSJed Brown                        total: nonzeros=69, allocated nonzeros=69
176*c4762a1bSJed Brown                        total number of mallocs used during MatSetValues calls=0
177*c4762a1bSJed Brown                          using I-node routines: found 5 nodes, limit used is 5
178*c4762a1bSJed Brown                linear system matrix = precond matrix:
179*c4762a1bSJed Brown                Mat Object:     1 MPI processes
180*c4762a1bSJed Brown                  type: seqaij
181*c4762a1bSJed Brown                  rows=9, cols=9
182*c4762a1bSJed Brown                  total: nonzeros=61, allocated nonzeros=61
183*c4762a1bSJed Brown                  total number of mallocs used during MatSetValues calls=0
184*c4762a1bSJed Brown                    using I-node routines: found 7 nodes, limit used is 5
185*c4762a1bSJed Brown    linear system matrix = precond matrix:
186*c4762a1bSJed Brown    Mat Object: 5 MPI processes
187*c4762a1bSJed Brown      type: is
188*c4762a1bSJed Brown      rows=49, cols=49
189*c4762a1bSJed Brown      total: nonzeros=361, allocated nonzeros=463
190*c4762a1bSJed Brown      total number of mallocs used during MatSetValues calls=0
191