Dear all,
When I run my lammps script on 2 cores (23 Aug 2016) on my laptop in parallel all things work fine, but run it on 16 cores ((14 May 2016)) on the cluster i get following error:
run 2000000
Neighbor list info …
1 neighbor list requests
update every 50 steps, delay 0 steps, check yes
max neighbors/atom: 2000, page size: 100000
master list distance cutoff = 5.1
ghost atom cutoff = 5.1
binsize = 2.55, bins = 21 21 166
Setting up Verlet run …
Unit style : metal
Current step: 0
Time step : 0.001
Memory usage per processor = 5.12159 Mbytes
Step
ERROR: Lost atoms: original 29973 current 10776 (…/thermo.cpp:398)
Can you please help me?
Thanks a lot
inR25.txt (3.9 KB)
Dear all,
When I run my lammps script on 2 cores (23 Aug 2016) on my laptop in
parallel all things work fine, but run it on 16 cores ((14 May 2016)) on the
cluster i get following error:
run 2000000
Neighbor list info ...
1 neighbor list requests
update every 50 steps, delay 0 steps, check yes
max neighbors/atom: 2000, page size: 100000
master list distance cutoff = 5.1
ghost atom cutoff = 5.1
binsize = 2.55, bins = 21 21 166
Setting up Verlet run ...
Unit style : metal
Current step: 0
Time step : 0.001
Memory usage per processor = 5.12159 Mbytes
Step
ERROR: Lost atoms: original 29973 current 10776 (../thermo.cpp:398)
Can you please help me?
you have set shrik-wrap boundary conditions. does your box dimensions
in the data file match the shrink-wrap conditions?
otherwise, LAMMPS will enforce it on the first MD step, and then you
will lose atoms when the change is too large for the subdomains (as it
would often happen when using more than just a couple of processors).
axel.