You should be using KOKKOS OpenMP version of reaxff instead of deprecated serial version even if you’re not using gpus
@stamoor has been pretty clear about that …
“IMO anyone and everyone should be using the KOKKOS version of ReaxFF. Not only is it more memory robust and will never have these hbondchk errors, it is also faster on CPUs, at least in most cases that I’ve benchmarked, or same speed at the very least.”
– Stan Moore (2024/10) on MatSci.org :
Lammps hbondchk failed .
@wangchao There are a number of problems with your input.
for ReaxFF you need to use a shorter timestep than the default, specifically when your system contains hydrogen atoms and when you are running at a very high temperature
please also check your log file or screen output. There should be a warning about the neighbor list settings being changed due to your settings for fix reaxff/species. averaging over 100 steps is not going to work for anything with a large-ish timestep that is not a so…
“I highly suggest using the KOKKOS package for ReaxFF, works in serial for CPUs too.”
– Stan Moore (2024/10) on MatSci.org :
Segmentation fault: address not mapped to object at address 0xc2cfb87c .
It is very unlikely that a segmentation fault will be cause by an out-of-memory scenario these days and the fact that you can run with a single process confirms that. This rather hints at a design choice of the original ReaxFF code (in C): it was making guesses how large certain arrays would become during a run at the most. However, this assumption is only valid for pre-equilibrated bulk systems. When running in parallel and starting from a non-equilibrated configuration the changes can be larg…
“You could also try the KOKKOS version which doesn’t use the safezone, mincap, and minhbonds factors which can bloat the memory if you set them too high.”
– Stan Moore (2025/01) on MatSci.org :
Possible memory problem with Reaxff when the total atom number increased .
There are limitations, but they require much, MUCH larger systems.
You can easily verify that more than 90000 atoms are possible, by inserting a replicate 2 2 2 command into the input right after the read_data command in the examples/reaxff/FC folder.
While the 30000 atom simulation does not crash, its thermo output is highly suspicious with its unusually large fluctuations. That indicates that some of your simulation settings are problematic or the potential parameters not suitable for your…
I’ve run reaxff/kk jobs with up to ~6000 cpu cores with no problems
2 Likes