Re: [petsc-dev] PETSc future starting as a new design layer that runs on top of PETSc 3?

FWIW, vendors have poured a lot of effort into making C++ the industry standard for HPC, and it will remain that way for a very long time. Switching PETSc to a non-C/C++/Python/Fortran language today while still enabling GPU/accelerated computing could get ugly from a code implementation

Re: [petsc-dev] odd log behavior

I think N/A (not applicable) would be a better message than NaN. Prior to this mail, even I thought I broke something with these NaN's On Tue, Apr 26, 2022 at 2:49 PM Matthew Knepley wrote: > On Tue, Apr 26, 2022 at 12:03 PM Mark Adams wrote: > >> Well, Nans are a clear sign that something is

Re: [petsc-dev] PETSc amg solver with gpu seems run slowly

Hi Qi, Mark, My colleague Suyash Tandon has almost completed a PETSc HIP port (essentially a hipification of the CUDA port) and has been trying to test it on the same OpenFOAM 3D Lid-driven case. It would be interesting to see what the optimal HYPRE parameters are as we could experiment from the

Re: [petsc-dev] Kokkos/Crusher perforance

show line numbers in the code. On Wed, Jan 26, 2022 at 1:29 PM Mark Adams wrote: > > > On Wed, Jan 26, 2022 at 1:54 PM Justin Chang wrote: > >> Couple suggestions: >> >> 1. Set the environment variable "export AMD_LOG_LEVEL=3" <- this will >> tell

Re: [petsc-dev] Kokkos/Crusher perforance

Couple suggestions: 1. Set the environment variable "export AMD_LOG_LEVEL=3" <- this will tell you everything that's happening at the HIP level (memcpy's, mallocs, kernel execution time, etc) 2. Try rocgdb, AFAIK this is the closest "HIP variant of valgrind" that we officially support. There are

Re: [petsc-dev] Kokkos/Crusher perforance

a more "official" and reproducible PETSc GPU benchmark, which I've already expressed to some folks in this thread, and as others already commented on the difficulty of such a task. Hopefully I will have more time soon to illustrate what I am thinking. On Mon, Jan 24, 2022 at 1:57 PM Ju

Re: [petsc-dev] Kokkos/Crusher perforance

n-ECP > technical support contact. > >But if this is an AMD machine then maybe contact Matt's student Justin > Chang? > > > > > > On Mon, Jan 24, 2022 at 2:18 PM Barry Smith wrote: > >> >> I think you should contact the crusher ECP technical support team and >> tell t

Re: [petsc-dev] HIP / hypre

for the HIP part of HYPRE, but we’re unsure what direction you guys want to go with the GPU route. Thanks, Justin On Wed, Nov 10, 2021 at 12:19 PM Justin Chang wrote: > Cc’ing Paul since I misspelled his email address initially > > On Wed, Nov 10, 2021 at 12:17 PM Jacob Faibussowitsch > wro

Re: [petsc-dev] HIP / hypre

; > Best regards, > > Jacob Faibussowitsch > (Jacob Fai - booss - oh - vitch) > > On Nov 10, 2021, at 11:50, Justin Chang wrote: > > Paul Bauman was also involved with the HIP port of HYPRE. Several of us at > AMD are interested in getting HIP support for PETSc in

Re: [petsc-dev] HIP / hypre

Paul Bauman was also involved with the HIP port of HYPRE. Several of us at AMD are interested in getting HIP support for PETSc in general, and having HYPRE support would greatly help On Wed, Nov 10, 2021 at 11:47 AM Stefano Zampini wrote: > I did the work last summer. It's already available in

Re: [petsc-dev] good PETSc example for weak scaling study?

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 wrote: > >I'm looking for a good PETSc

Re: [petsc-dev] Hardware counter logging in PETSc (was Re: Where next with PETSc and KNL?)

The so called "perfect cache" model in my paper , which was counted by hand, only works for solvers and preconditioners which rely on sparse matrix-vector multiply. I reluctantly used ILU in the first part of the paper for that very

Re: [petsc-dev] WTF

That guy's results actually make sense to me. I also get poor strong-scaling for the FEM version of the poisson equation (via firedrake) using HYPRE's boomerAMG. The studies were done on Intel E5-2670 machines and had proper OpenMPI bindings.No HYPRE configure options were set via command line so

[petsc-dev] PETSc meeting this summer?

Is there going to be another PETSc meeting this summer? Thanks, Justin

[petsc-dev] Always compiling and installing Ctetgen even without updates

Hi developers, Is there any particular reason why PETSc ./configure always compiles and installs Ctetgen even if no updates are made to the source code? Not that it's a big deal or anything, but it just gives me that same uncomfortable itchy feeling of seeing my high school math teacher erase