PCCHOLESKY#

Uses a direct solver, based on Cholesky factorization, as a preconditioner

Options Database Keys#

  • -pc_factor_reuse_ordering - Activate PCFactorSetReuseOrdering()

  • -pc_factor_mat_solver_type - Actives PCFactorSetMatSolverType() to choose the direct solver, like superlu

  • -pc_factor_reuse_fill - Activates PCFactorSetReuseFill()

  • -pc_factor_fill - Sets the explected fill amount

  • -pc_factor_in_place - Activates in-place factorization

  • -pc_factor_mat_ordering_type <nd,rcm,…> - Sets ordering routine used to determine the order the rows are used in the factorization to reduce fill and thus be more effective

Notes#

The Cholesky factorization direct solver, PCCHOLESKY is only for symmetric positive-definite (SPD) matrices. For such SPD matrices it is more efficient than using the LU factorization direct solver, PCLU.

Not all options work for all matrix formats

Usually this will compute an “exact” solution in one iteration and does not need a Krylov method (i.e. you can use -ksp_type preonly, or KSPSetType(ksp,KSPPREONLY) for the Krylov method

See Also#

KSP: Linear System Solvers, PCCreate(), PCSetType(), PCType, PC, PCILU, PCLU, PCICC, PCFactorSetReuseOrdering(), PCFactorSetReuseFill(), PCFactorGetMatrix(), PCFactorSetFill(), PCFactorSetShiftType(), PCFactorSetShiftAmount() PCFactorSetUseInPlace(), PCFactorGetUseInPlace(), PCFactorSetMatOrderingType(), PCFactorSetReuseOrdering()

Level#

beginner

Location#

src/ksp/pc/impls/factor/cholesky/cholesky.c

Examples#

src/ksp/ksp/tutorials/ex52.c
src/tao/constrained/tutorials/ex1.c
src/ksp/pc/tutorials/ex4.c


Index of all PC routines
Table of Contents for all manual pages
Index of all manual pages