search

lanl / spiner

Performance portable routines for generic, tabulated, multi-dimensional data
https://lanl.github.io/spiner
BSD 3-Clause "New" or "Revised" License
7 stars 3 forks source link

`benchmark.bin` fails silently for large grid size #38

Open lgarrison opened 2 years ago

lgarrison commented 2 years ago

Per the title, the benchmark seems to complete instantly if the fine grid gets too big:

lgarrison@workergpu142:~/spiner/bin/test$ ./benchmark.bin 64 6000
# ncoarse = 64
# nfine time/point (ns) L2 error
6000    2.03796296296296e-07    0.00000000000000e+00
lgarrison@workergpu142:~/spiner/bin/test$ echo $?
0

x-ref: https://github.com/openjournals/joss-reviews/issues/4367, but this is low-priority and doesn't need to be resolved as part of the review.

Yurlungur commented 2 years ago

Interesting. I'll take a look at this. In what mode did you run this? On GPU? My guess is that the GPU is running out of memory when the benchmark attempts to allocate the arrays. But the fact there's no runtime error is disturbing.

lgarrison commented 2 years ago

Yes, Kokkos CUDA.

On Wed, Jun 15, 2022 at 2:43 PM Jonah Miller @.***> wrote:

Interesting. I'll take a look at this. In what mode did you run this? On GPU? My guess is that the GPU is running out of memory when the benchmark attempts to allocate the arrays. But the fact there's no runtime error is disturbing.

— Reply to this email directly, view it on GitHub https://github.com/lanl/spiner/issues/38#issuecomment-1156806990, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABLA7S3VSBD7VGIVQLJY233VPIP3ZANCNFSM5Y4ILP7A . You are receiving this because you authored the thread.Message ID: @.***>

-- Lehman Garrison Flatiron Research Fellow, Cosmology X Data Science Group Center for Computational Astrophysics, Flatiron Institute lgarrison.github.io

Yurlungur commented 2 years ago

Cool, thanks for the pointer. I'll look at this.