Re: [petsc-users] superlu_dist+parmetis

Thanks for the information. METIS works very well, so it is no big thing that parmetis doesn'.   best, marius   Gesendet: Samstag, 16. Februar 2019 um 00:01 Uhr Von: "Xiaoye S. Li" An: "Smith, Barry F." Cc: "Marius Buerkle" , "PETSc users list" Betreff: Re: [petsc-users] superlu_dist+parmet

Re: [petsc-users] KSP ex2.c and Parallelization

Unfortunately converting a sequential code to MPI parallelism requires a great deal of restructuring of the code (often little of the original code can be retained) to get good performance. This is because the data has to be partitioned well in parallel to get good performance. src/ksp/ksp

Re: [petsc-users] KSP ex2.c and Parallelization

Maybe we could provide more guidance on how to proceed if we knew a bit more about the 1) problem you are trying to solve, 2) the discretization you are using and 3) the mesh you are using? From your code fragment with a m*n I am guessing you are solving something on a structured grid

Re: [petsc-users] KSP ex2.c and Parallelization

Maahi : Hello All, I have some questions regarding ex2.c in /petsc/src/ksp/ksp/examples/tutorials/ex2.c. 1. The local sizes of the matrix was decided by PETSc using ierr = MatSetSizes(A,PETSC_DECIDE,PETSC_DECIDE,m*n,m*n);CHKERRQ(ierr); But without knowing the local sizes, how was the followi

Re: [petsc-users] superlu_dist+parmetis

I am pretty sure this is a bug in parmetis. A few years ago, I nailed down this "divide by zero" bug, and reported to them. They said they would take a look, but never did. This usually happens when the graph is relatively dense. You can try sequential Metis. Sherry On Thu, Feb 14, 2019 at 10:

Re: [petsc-users] Missing Fortran interface for PetscFree?

On Fri, Feb 15, 2019 at 6:26 AM Marco Tiberga via petsc-users < petsc-users@mcs.anl.gov> wrote: > Dear PETSc developers, > > > > Unfortunately, I need to continue the previous conversation (which I > attached), because I am getting another compiler error. > > I need to call PetscFree() after PCFie

[petsc-users] Missing Fortran interface for PetscFree?

Dear PETSc developers, Unfortunately, I need to continue the previous conversation (which I attached), because I am getting another compiler error. I need to call PetscFree() after PCFieldSplitGetSubKSP (I am trying to implement what is described in lines 117—120 of this example