:orphan: # MatMPIAdjToSeqRankZero Converts an parallel `MATMPIADJ` matrix to complete `MATMPIADJ` on rank zero (needed by sequential partitioner) ## Synopsis ``` #include "petscmat.h" PetscErrorCode MatMPIAdjToSeqRankZero(Mat A, Mat *B) ``` Logically Collective ## Input Parameter - ***A -*** the matrix ## Output Parameter - ***B -*** the same matrix on rank zero, not set on other ranks ## Note This routine has the advantage on systems with multiple ranks per node since only one copy of the matrix is stored on the first node, instead of the number of ranks copies. This can allow partitioning much larger parallel graph sequentially. ## See Also [](ch_matrices), `Mat`, `MATMPIADJ`, `MatCreate()`, `MatCreateMPIAdj()`, `MatSetValues()`, `MatMPIAdjToSeq()` ## Level intermediate ## Location src/mat/impls/adj/mpi/mpiadj.c ## Implementations MatMPIAdjToSeqRankZero_MPIAdj in src/mat/impls/adj/mpi/mpiadj.c
--- [Edit on GitLab](https://gitlab.com/petsc/petsc/-/edit/release/src/mat/impls/adj/mpi/mpiadj.c) [Index of all Mat routines](index.md) [Table of Contents for all manual pages](/manualpages/index.md) [Index of all manual pages](/manualpages/singleindex.md)