You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I've been using ExaMPM (DamBreak) for performance and profiling on one Crusher node and continue to see what look like communications errors leading to spurious new velocities at processor boundaries which then lead to numerical blow-up and crashes (almost always in g2p->scatter->packBuffer). This only occurs for problems over about 50^3 cells and 4 or more MPI ranks (I test with srun -N1 -n8 -S16 --exclusive -t30:00 --cpus-per-task=1 --threads-per-core=1 --gpus-per-task=1 --gpu-bind=closest ./DamBreak 0.01 2 3 0.00004 10.0 2500 hip) and does not go away with wider halos, evenly-divisible ny, different Y boundary conditions (periodic, slip, noslip), or different versions of Cabana (0.5.0, head). It seems to be suppressed somewhat with fewer particles per cell and by using AMD_SANITIZE_KERNEL and AMD_SANITIZE_COPY vars, but never goes away. For a while, it seemed to always happen 2 or 3 time steps after a Silo write, but still occurs without any Silo writes. It can occur anywhere between steps 5000 and 100000.
The text was updated successfully, but these errors were encountered:
From Mark on slack:
I've been using ExaMPM (DamBreak) for performance and profiling on one Crusher node and continue to see what look like communications errors leading to spurious new velocities at processor boundaries which then lead to numerical blow-up and crashes (almost always in
g2p->scatter->packBuffer
). This only occurs for problems over about 50^3 cells and 4 or more MPI ranks (I test withsrun -N1 -n8 -S16 --exclusive -t30:00 --cpus-per-task=1 --threads-per-core=1 --gpus-per-task=1 --gpu-bind=closest ./DamBreak 0.01 2 3 0.00004 10.0 2500 hip
) and does not go away with wider halos, evenly-divisible ny, different Y boundary conditions (periodic, slip, noslip), or different versions of Cabana (0.5.0, head). It seems to be suppressed somewhat with fewer particles per cell and by usingAMD_SANITIZE_KERNEL
andAMD_SANITIZE_COPY
vars, but never goes away. For a while, it seemed to always happen 2 or 3 time steps after a Silo write, but still occurs without any Silo writes. It can occur anywhere between steps 5000 and 100000.The text was updated successfully, but these errors were encountered: