MATMPIAIJCUSPARSE#
A matrix type to be used for sparse matrices on NVIDIA GPUs.
Options Database Keys#
-mat_type mpiaijcusparse - sets the matrix type to
MATMPIAIJCUSPARSE-mat_cusparse_storage_format csr - sets the storage format of diagonal and off-diagonal matrices. Other options include ell (ellpack) or hyb (hybrid).
-mat_cusparse_mult_diag_storage_format csr - sets the storage format of diagonal matrix. Other options include ell (ellpack) or hyb (hybrid).
-mat_cusparse_mult_offdiag_storage_format csr - sets the storage format of off-diagonal matrix. Other options include ell (ellpack) or hyb (hybrid).
Notes#
These matrices can be in either CSR, ELL, or HYB format. The ELL and HYB formats require CUDA 4.2 or later.
All matrix calculations are performed on NVIDIA GPUs using the cuSPARSE library.
Uses 32-bit integers internally. If PETSc is configured --with-64-bit-indices, the integer row and column indices are stored on the GPU with int. It is unclear what happens
if some integer values passed in do not fit in int.
See Also#
Matrices, Mat, MatCreateAIJCUSPARSE(), MATSEQAIJCUSPARSE, MATMPIAIJCUSPARSE, MatCreateSeqAIJCUSPARSE(), MatCUSPARSESetFormat(), MatCUSPARSEStorageFormat, MatCUSPARSEFormatOperation
Level#
beginner
Location#
Examples#
Index of all Mat routines
Table of Contents for all manual pages
Index of all manual pages