Sorry, forgot to reply all: ---
I liked SNES ex48, I managed to do some scaling on 256 KNL nodes. Some results here, but didn't compare to Broadwell https://arxiv.org/abs/1705.03625 On Tue, Jul 25, 2017 at 5:59 PM, Barry Smith <[email protected]> wrote: > > I'm looking for a good PETSc example for a weak scaling study up to > 256*65 MPI processes. This means that > > 1) the example should not have non-scalable parts to it > > 2) it should be easy to increase the problem size properly with the number > of MPI processes. > > 3) the algorithms used should be scalable, or possible to run with a fixed > number of iterations with a simple preconditioner > > This is for studying a new machine, not studying PETSc, and comparing > KNL to Broadwell. > > Thanks > > Barry > >
