Re: [petsc-users] Understanding preallocation for MPI

> On Jul 16, 2017, at 9:53 PM, Florian Lindner wrote: > > Hi Barry, > > Am 17.07.2017 um 10:42 schrieb Barry Smith: >>The intention with the AO is you map everything to the PETSc ordering and >> then you just work in the PETSc ordering completely, this way there is no

Re: [petsc-users] Understanding preallocation for MPI

Hi Barry, Am 17.07.2017 um 10:42 schrieb Barry Smith: > The intention with the AO is you map everything to the PETSc ordering and > then you just work in the PETSc ordering completely, this way there is no > "sending around with MPI preallocation information". That is your matrix >

Re: [petsc-users] Understanding preallocation for MPI

Florian, The intention with the AO is you map everything to the PETSc ordering and then you just work in the PETSc ordering completely, this way there is no "sending around with MPI preallocation information". That is your matrix assembly and preallocation (as well as vectors) always

Re: [petsc-users] Understanding preallocation for MPI

Hello, Am 11.07.2017 um 02:45 schrieb Barry Smith: > > You might consider using > http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/Mat/MatPreallocateInitialize.html > and >