[petsc-users] MatPtAP for involving MPIDENSE and MPIAIJ matrices

Jed Brown jed at jedbrown.org
Thu Oct 15 14:57:41 CDT 2015


r<#secure method=pgpmime mode=sign>
Hong <hzhang at mcs.anl.gov> writes:

> I plan to implement MatTransposeMatMult_MPIDense_MPIDense via
>
> 1. add MatTransposeMatMult_elemental_elemental()
> 2. C_dense = P_dense^T * B_dense
>     via MatConvert_dense_elemental() and MatConvert_elemental_dense()

The above involves a ton of data movement and MPIDense is a logical
distribution for matrices with a modest number of columns.  I think I
would just do the local GEMM and then MPI_Reduce_scatter it.


More information about the petsc-users mailing list