1c4762a1bSJed Brownlid velocity = 0.0204082, prandtl # = 1., grashof # = 1. 2c4762a1bSJed Brown 0 SNES Function norm 0.146194 3c4762a1bSJed Brown 0 KSP Residual norm 0.146194 4c4762a1bSJed Brown 1 KSP Residual norm 0.0479011 5c4762a1bSJed Brown 2 KSP Residual norm 0.010778 6c4762a1bSJed Brown 3 KSP Residual norm 0.00272661 7c4762a1bSJed Brown 4 KSP Residual norm 0.000690594 8c4762a1bSJed Brown 5 KSP Residual norm 0.000210687 9c4762a1bSJed Brown 6 KSP Residual norm 0.000105322 10c4762a1bSJed Brown 7 KSP Residual norm 6.5975e-05 11c4762a1bSJed Brown 8 KSP Residual norm 2.80726e-05 12c4762a1bSJed Brown 9 KSP Residual norm 1.87879e-05 13c4762a1bSJed Brown 10 KSP Residual norm 8.24594e-06 14c4762a1bSJed Brown 11 KSP Residual norm 1.72955e-06 15c4762a1bSJed Brown 12 KSP Residual norm 5.17053e-07 16c4762a1bSJed Brown 1 SNES Function norm 2.5023e-05 17c4762a1bSJed Brown 0 KSP Residual norm 2.5023e-05 18c4762a1bSJed Brown 1 KSP Residual norm 5.03001e-06 19c4762a1bSJed Brown 2 KSP Residual norm 1.40061e-06 20c4762a1bSJed Brown 3 KSP Residual norm 5.68423e-07 21c4762a1bSJed Brown 4 KSP Residual norm 1.98909e-07 22c4762a1bSJed Brown 5 KSP Residual norm 9.6167e-08 23c4762a1bSJed Brown 6 KSP Residual norm 4.64061e-08 24c4762a1bSJed Brown 7 KSP Residual norm 9.70042e-09 25c4762a1bSJed Brown 8 KSP Residual norm 2.76341e-09 26c4762a1bSJed Brown 9 KSP Residual norm 9.531e-10 27c4762a1bSJed Brown 10 KSP Residual norm 4.491e-10 28c4762a1bSJed Brown 11 KSP Residual norm 2.978e-10 29c4762a1bSJed Brown 12 KSP Residual norm 1.484e-10 30c4762a1bSJed Brown 2 SNES Function norm 1.483e-10 31c4762a1bSJed BrownSNES Object: 3 MPI processes 32c4762a1bSJed Brown type: newtonls 33c4762a1bSJed Brown maximum iterations=50, maximum function evaluations=10000 34c4762a1bSJed Brown tolerances: relative=1e-08, absolute=1e-50, solution=1e-08 35c4762a1bSJed Brown total number of linear solver iterations=24 36c4762a1bSJed Brown total number of function evaluations=3 37c4762a1bSJed Brown norm schedule ALWAYS 38c4762a1bSJed Brown Jacobian is built using colored finite differences on a DMDA 39c4762a1bSJed Brown SNESLineSearch Object: 3 MPI processes 40c4762a1bSJed Brown type: bt 41c4762a1bSJed Brown interpolation: cubic 42c4762a1bSJed Brown alpha=1.000000e-04 43c4762a1bSJed Brown maxstep=1.000000e+08, minlambda=1.000000e-12 44c4762a1bSJed Brown tolerances: relative=1.000000e-08, absolute=1.000000e-15, lambda=1.000000e-08 45c4762a1bSJed Brown maximum iterations=40 46c4762a1bSJed Brown KSP Object: 3 MPI processes 47c4762a1bSJed Brown type: fgmres 48c4762a1bSJed Brown restart=30, using Classical (unmodified) Gram-Schmidt Orthogonalization with no iterative refinement 49c4762a1bSJed Brown happy breakdown tolerance 1e-30 50c4762a1bSJed Brown maximum iterations=10000, initial guess is zero 51c4762a1bSJed Brown tolerances: relative=1e-05, absolute=1e-50, divergence=10000. 52c4762a1bSJed Brown right preconditioning 53c4762a1bSJed Brown using UNPRECONDITIONED norm type for convergence test 54c4762a1bSJed Brown PC Object: 3 MPI processes 55c4762a1bSJed Brown type: fieldsplit 56c4762a1bSJed Brown FieldSplit with SYMMETRIC_MULTIPLICATIVE composition: total splits = 4 57c4762a1bSJed Brown Solver info for each split is in the following KSP objects: 58c4762a1bSJed Brown Split number 0 Defined by IS 59c4762a1bSJed Brown KSP Object: (fieldsplit_x_velocity_) 3 MPI processes 60c4762a1bSJed Brown type: preonly 61c4762a1bSJed Brown maximum iterations=10000, initial guess is zero 62c4762a1bSJed Brown tolerances: relative=1e-05, absolute=1e-50, divergence=10000. 63c4762a1bSJed Brown left preconditioning 64c4762a1bSJed Brown using NONE norm type for convergence test 65c4762a1bSJed Brown PC Object: (fieldsplit_x_velocity_) 3 MPI processes 66c4762a1bSJed Brown type: bjacobi 67c4762a1bSJed Brown number of blocks = 3 688d76b567SPatrick Sanan Local solver is the same for all blocks, as in the following KSP and PC objects on rank 0: 69c4762a1bSJed Brown KSP Object: (fieldsplit_x_velocity_sub_) 1 MPI processes 70c4762a1bSJed Brown type: preonly 71c4762a1bSJed Brown maximum iterations=10000, initial guess is zero 72c4762a1bSJed Brown tolerances: relative=1e-05, absolute=1e-50, divergence=10000. 73c4762a1bSJed Brown left preconditioning 74c4762a1bSJed Brown using NONE norm type for convergence test 75c4762a1bSJed Brown PC Object: (fieldsplit_x_velocity_sub_) 1 MPI processes 76c4762a1bSJed Brown type: ilu 77c4762a1bSJed Brown out-of-place factorization 78c4762a1bSJed Brown 0 levels of fill 79c4762a1bSJed Brown tolerance for zero pivot 2.22045e-14 80c4762a1bSJed Brown matrix ordering: natural 81c4762a1bSJed Brown factor fill ratio given 1., needed 1. 82c4762a1bSJed Brown Factored matrix follows: 83c4762a1bSJed Brown Mat Object: 1 MPI processes 84c4762a1bSJed Brown type: seqaij 85c4762a1bSJed Brown rows=21, cols=21 86c4762a1bSJed Brown package used to perform factorization: petsc 87c4762a1bSJed Brown total: nonzeros=85, allocated nonzeros=85 88c4762a1bSJed Brown not using I-node routines 89c4762a1bSJed Brown linear system matrix = precond matrix: 90*793a3527SPierre Jolivet Mat Object: (fieldsplit_x_velocity_sub_) 1 MPI processes 91c4762a1bSJed Brown type: seqaij 92c4762a1bSJed Brown rows=21, cols=21 93c4762a1bSJed Brown total: nonzeros=85, allocated nonzeros=85 94c4762a1bSJed Brown total number of mallocs used during MatSetValues calls=0 95c4762a1bSJed Brown not using I-node routines 96c4762a1bSJed Brown linear system matrix = precond matrix: 97c4762a1bSJed Brown Mat Object: (fieldsplit_x_velocity_) 3 MPI processes 98c4762a1bSJed Brown type: mpiaij 99c4762a1bSJed Brown rows=49, cols=49 100c4762a1bSJed Brown total: nonzeros=217, allocated nonzeros=217 101c4762a1bSJed Brown total number of mallocs used during MatSetValues calls=0 102c4762a1bSJed Brown not using I-node (on process 0) routines 103c4762a1bSJed Brown Split number 1 Defined by IS 104c4762a1bSJed Brown KSP Object: (fieldsplit_y_velocity_) 3 MPI processes 105c4762a1bSJed Brown type: preonly 106c4762a1bSJed Brown maximum iterations=10000, initial guess is zero 107c4762a1bSJed Brown tolerances: relative=1e-05, absolute=1e-50, divergence=10000. 108c4762a1bSJed Brown left preconditioning 109c4762a1bSJed Brown using NONE norm type for convergence test 110c4762a1bSJed Brown PC Object: (fieldsplit_y_velocity_) 3 MPI processes 111c4762a1bSJed Brown type: bjacobi 112c4762a1bSJed Brown number of blocks = 3 1138d76b567SPatrick Sanan Local solver is the same for all blocks, as in the following KSP and PC objects on rank 0: 114c4762a1bSJed Brown KSP Object: (fieldsplit_y_velocity_sub_) 1 MPI processes 115c4762a1bSJed Brown type: preonly 116c4762a1bSJed Brown maximum iterations=10000, initial guess is zero 117c4762a1bSJed Brown tolerances: relative=1e-05, absolute=1e-50, divergence=10000. 118c4762a1bSJed Brown left preconditioning 119c4762a1bSJed Brown using NONE norm type for convergence test 120c4762a1bSJed Brown PC Object: (fieldsplit_y_velocity_sub_) 1 MPI processes 121c4762a1bSJed Brown type: ilu 122c4762a1bSJed Brown out-of-place factorization 123c4762a1bSJed Brown 0 levels of fill 124c4762a1bSJed Brown tolerance for zero pivot 2.22045e-14 125c4762a1bSJed Brown matrix ordering: natural 126c4762a1bSJed Brown factor fill ratio given 1., needed 1. 127c4762a1bSJed Brown Factored matrix follows: 128c4762a1bSJed Brown Mat Object: 1 MPI processes 129c4762a1bSJed Brown type: seqaij 130c4762a1bSJed Brown rows=21, cols=21 131c4762a1bSJed Brown package used to perform factorization: petsc 132c4762a1bSJed Brown total: nonzeros=85, allocated nonzeros=85 133c4762a1bSJed Brown not using I-node routines 134c4762a1bSJed Brown linear system matrix = precond matrix: 135*793a3527SPierre Jolivet Mat Object: (fieldsplit_y_velocity_sub_) 1 MPI processes 136c4762a1bSJed Brown type: seqaij 137c4762a1bSJed Brown rows=21, cols=21 138c4762a1bSJed Brown total: nonzeros=85, allocated nonzeros=85 139c4762a1bSJed Brown total number of mallocs used during MatSetValues calls=0 140c4762a1bSJed Brown not using I-node routines 141c4762a1bSJed Brown linear system matrix = precond matrix: 142c4762a1bSJed Brown Mat Object: (fieldsplit_y_velocity_) 3 MPI processes 143c4762a1bSJed Brown type: mpiaij 144c4762a1bSJed Brown rows=49, cols=49 145c4762a1bSJed Brown total: nonzeros=217, allocated nonzeros=217 146c4762a1bSJed Brown total number of mallocs used during MatSetValues calls=0 147c4762a1bSJed Brown not using I-node (on process 0) routines 148c4762a1bSJed Brown Split number 2 Defined by IS 149c4762a1bSJed Brown KSP Object: (fieldsplit_Omega_) 3 MPI processes 150c4762a1bSJed Brown type: preonly 151c4762a1bSJed Brown maximum iterations=10000, initial guess is zero 152c4762a1bSJed Brown tolerances: relative=1e-05, absolute=1e-50, divergence=10000. 153c4762a1bSJed Brown left preconditioning 154c4762a1bSJed Brown using NONE norm type for convergence test 155c4762a1bSJed Brown PC Object: (fieldsplit_Omega_) 3 MPI processes 156c4762a1bSJed Brown type: bjacobi 157c4762a1bSJed Brown number of blocks = 3 1588d76b567SPatrick Sanan Local solver is the same for all blocks, as in the following KSP and PC objects on rank 0: 159c4762a1bSJed Brown KSP Object: (fieldsplit_Omega_sub_) 1 MPI processes 160c4762a1bSJed Brown type: preonly 161c4762a1bSJed Brown maximum iterations=10000, initial guess is zero 162c4762a1bSJed Brown tolerances: relative=1e-05, absolute=1e-50, divergence=10000. 163c4762a1bSJed Brown left preconditioning 164c4762a1bSJed Brown using NONE norm type for convergence test 165c4762a1bSJed Brown PC Object: (fieldsplit_Omega_sub_) 1 MPI processes 166c4762a1bSJed Brown type: ilu 167c4762a1bSJed Brown out-of-place factorization 168c4762a1bSJed Brown 0 levels of fill 169c4762a1bSJed Brown tolerance for zero pivot 2.22045e-14 170c4762a1bSJed Brown matrix ordering: natural 171c4762a1bSJed Brown factor fill ratio given 1., needed 1. 172c4762a1bSJed Brown Factored matrix follows: 173c4762a1bSJed Brown Mat Object: 1 MPI processes 174c4762a1bSJed Brown type: seqaij 175c4762a1bSJed Brown rows=21, cols=21 176c4762a1bSJed Brown package used to perform factorization: petsc 177c4762a1bSJed Brown total: nonzeros=85, allocated nonzeros=85 178c4762a1bSJed Brown not using I-node routines 179c4762a1bSJed Brown linear system matrix = precond matrix: 180*793a3527SPierre Jolivet Mat Object: (fieldsplit_Omega_sub_) 1 MPI processes 181c4762a1bSJed Brown type: seqaij 182c4762a1bSJed Brown rows=21, cols=21 183c4762a1bSJed Brown total: nonzeros=85, allocated nonzeros=85 184c4762a1bSJed Brown total number of mallocs used during MatSetValues calls=0 185c4762a1bSJed Brown not using I-node routines 186c4762a1bSJed Brown linear system matrix = precond matrix: 187c4762a1bSJed Brown Mat Object: (fieldsplit_Omega_) 3 MPI processes 188c4762a1bSJed Brown type: mpiaij 189c4762a1bSJed Brown rows=49, cols=49 190c4762a1bSJed Brown total: nonzeros=217, allocated nonzeros=217 191c4762a1bSJed Brown total number of mallocs used during MatSetValues calls=0 192c4762a1bSJed Brown not using I-node (on process 0) routines 193c4762a1bSJed Brown Split number 3 Defined by IS 194c4762a1bSJed Brown KSP Object: (fieldsplit_temperature_) 3 MPI processes 195c4762a1bSJed Brown type: preonly 196c4762a1bSJed Brown maximum iterations=10000, initial guess is zero 197c4762a1bSJed Brown tolerances: relative=1e-05, absolute=1e-50, divergence=10000. 198c4762a1bSJed Brown left preconditioning 199c4762a1bSJed Brown using NONE norm type for convergence test 200c4762a1bSJed Brown PC Object: (fieldsplit_temperature_) 3 MPI processes 201c4762a1bSJed Brown type: bjacobi 202c4762a1bSJed Brown number of blocks = 3 2038d76b567SPatrick Sanan Local solver is the same for all blocks, as in the following KSP and PC objects on rank 0: 204c4762a1bSJed Brown KSP Object: (fieldsplit_temperature_sub_) 1 MPI processes 205c4762a1bSJed Brown type: preonly 206c4762a1bSJed Brown maximum iterations=10000, initial guess is zero 207c4762a1bSJed Brown tolerances: relative=1e-05, absolute=1e-50, divergence=10000. 208c4762a1bSJed Brown left preconditioning 209c4762a1bSJed Brown using NONE norm type for convergence test 210c4762a1bSJed Brown PC Object: (fieldsplit_temperature_sub_) 1 MPI processes 211c4762a1bSJed Brown type: ilu 212c4762a1bSJed Brown out-of-place factorization 213c4762a1bSJed Brown 0 levels of fill 214c4762a1bSJed Brown tolerance for zero pivot 2.22045e-14 215c4762a1bSJed Brown matrix ordering: natural 216c4762a1bSJed Brown factor fill ratio given 1., needed 1. 217c4762a1bSJed Brown Factored matrix follows: 218c4762a1bSJed Brown Mat Object: 1 MPI processes 219c4762a1bSJed Brown type: seqaij 220c4762a1bSJed Brown rows=21, cols=21 221c4762a1bSJed Brown package used to perform factorization: petsc 222c4762a1bSJed Brown total: nonzeros=85, allocated nonzeros=85 223c4762a1bSJed Brown not using I-node routines 224c4762a1bSJed Brown linear system matrix = precond matrix: 225*793a3527SPierre Jolivet Mat Object: (fieldsplit_temperature_sub_) 1 MPI processes 226c4762a1bSJed Brown type: seqaij 227c4762a1bSJed Brown rows=21, cols=21 228c4762a1bSJed Brown total: nonzeros=85, allocated nonzeros=85 229c4762a1bSJed Brown total number of mallocs used during MatSetValues calls=0 230c4762a1bSJed Brown not using I-node routines 231c4762a1bSJed Brown linear system matrix = precond matrix: 232c4762a1bSJed Brown Mat Object: (fieldsplit_temperature_) 3 MPI processes 233c4762a1bSJed Brown type: mpiaij 234c4762a1bSJed Brown rows=49, cols=49 235c4762a1bSJed Brown total: nonzeros=217, allocated nonzeros=217 236c4762a1bSJed Brown total number of mallocs used during MatSetValues calls=0 237c4762a1bSJed Brown not using I-node (on process 0) routines 238c4762a1bSJed Brown linear system matrix = precond matrix: 239c4762a1bSJed Brown Mat Object: 3 MPI processes 240c4762a1bSJed Brown type: mpiaij 241c4762a1bSJed Brown rows=196, cols=196, bs=4 242c4762a1bSJed Brown total: nonzeros=3472, allocated nonzeros=3472 243c4762a1bSJed Brown total number of mallocs used during MatSetValues calls=0 244c4762a1bSJed BrownNumber of SNES iterations = 2 245