[Non-DoD Source] AIREBO: parallel simulations (UNCLASSIFIED)


Have you verified that both clusters are running the same version of LAMMPS?

There have been some updates to AIREBO within the last year (http://lammps.sandia.gov/bug.html)



Hi Chris

Discrepancy in potential energy was indeed due to version differences. On stampede2, i was using default lammps that gets loaded with module command (I believe its March 2017).

I compiled the latest lammps release on stampede2 and i got the energies which are very close to the ones i got on university cluster (I have Aug17 version on university cluster).

Thanks for the quick reply and very useful suggestion.