On Tue, Mar 25, 2014 at 12:01 PM, Fischer, Greg A. < [email protected]> wrote:
> > > *From:* Matthew Knepley [mailto:[email protected]] > *Sent:* Tuesday, March 25, 2014 12:50 PM > *To:* Fischer, Greg A. > *Cc:* [email protected] > *Subject:* Re: [petsc-users] function that returns the coordinates in the > DMDA group? > > > > On Tue, Mar 25, 2014 at 11:41 AM, Fischer, Greg A. < > [email protected]> wrote: > > Hello, > > > > The FAQ indicates: > > > > The MPI_Cart_create() first divides the mesh along the z > direction, then the y, then the x. DMDA divides along the x, then y, then z. > > > > Is there a PETSc function call that returns the coordinates of the calling > process in the DMDA group? > > > > I am not sure I understand exactly what you want. Can you do a small 2D > example, with 4 or 6 cells? > > > > > > In a 2D DMDA, my understanding is that the process ranks would be arranged > as: > > > > 2 3 > > 0 1 > > > > I would like to be able to call some function and have it return > coordinates: > > > > [Rank 0] (x,y) = (0, 0) > > [Rank 1] (x,y) = (0, 1) > > [Rank 2] (x,y) = (1, 0) > > [Rank 3] (x,y) = (1, 1) > We do not have an API method for that. We give you the rank, (m,n,p) for the number of processors in each direction, and that it is numbered lexicographically. Thanks, Matt > Greg > > > > Thanks, > > > > Matt > > > > Thanks, > > Greg > > > > > > -- > What most experimenters take for granted before they begin their > experiments is infinitely more interesting than any results to which their > experiments lead. > -- Norbert Wiener > -- What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead. -- Norbert Wiener
