Re: [petsc-users] Question regarding DMPlex reordering

2021-10-27 Thread Matthew Knepley
On Wed, Oct 27, 2021 at 9:54 AM Pierre Seize wrote: > Hi, thanks for the fix. It seems to work fine. > > Out of curiosity, I noticed the MatOrderingType of DMPlexGetOrdering is > not used. Is this intentional ? To match MatGetOrdering ? > True. It only does RCM. I noticed that when fixing it. I h

Re: [petsc-users] Question regarding DMPlex reordering

2021-10-27 Thread Pierre Seize
Hi, thanks for the fix. It seems to work fine. Out of curiosity, I noticed the MatOrderingType of DMPlexGetOrdering is not used. Is this intentional ? To match MatGetOrdering ? Pierre On 27/10/21 15:03, Matthew Knepley wrote: On Wed, Oct 27, 2021 at 3:15 AM Pierre Seize

Re: [petsc-users] Question regarding DMPlex reordering

2021-10-27 Thread Matthew Knepley
On Wed, Oct 27, 2021 at 3:15 AM Pierre Seize wrote: > > > On 26/10/21 22:28, Matthew Knepley wrote: > > On Tue, Oct 26, 2021 at 10:17 AM Pierre Seize > wrote: > >> Hi, I had the idea to try and renumber my mesh cells, as I've heard it's >> better: "neighbouring cells are stored next to one anoth

Re: [petsc-users] Question regarding DMPlex reordering

2021-10-27 Thread Pierre Seize
On 26/10/21 22:28, Matthew Knepley wrote: On Tue, Oct 26, 2021 at 10:17 AM Pierre Seize > wrote: Hi, I had the idea to try and renumber my mesh cells, as I've heard it's better: "neighbouring cells are stored next to one another, and memory access are

Re: [petsc-users] Question regarding DMPlex reordering

2021-10-26 Thread Matthew Knepley
On Tue, Oct 26, 2021 at 10:17 AM Pierre Seize wrote: > Hi, I had the idea to try and renumber my mesh cells, as I've heard it's > better: "neighbouring cells are stored next to one another, and memory > access are faster". > Right now, I load the mesh then I distribute it over the processes. I >

[petsc-users] Question regarding DMPlex reordering

2021-10-26 Thread Pierre Seize
Hi, I had the idea to try and renumber my mesh cells, as I've heard it's better: "neighbouring cells are stored next to one another, and memory access are faster". Right now, I load the mesh then I distribute it over the processes. I thought I'd try to permute the numbering between those two s