Dear Lammps Users, Developers,
I am using lammps to do some molecular statics. I displace the atoms and then I minimize the structure. I repeat such an operation for a large number of step.
At the end of the number of increment (I have XX time displace + minimize in the input file), the code just does not stop.
Also, because I only want the snapshot at the end of every minimization, I dump every 1000000 steps, which means never.
Do you have any idea of why?
Thanks,
Sebastien
Sebastien,
Your question seems unclear. Perhaps you could post an example input script.
When you say "the code just does not stop", I assume you mean that the minimization doesn't converge within the time you waited. You could try changing the minimization parameters to make it end sooner.
You can also change the dump frequency so that you get more snapshots.
Paul
Paul,
Actually, the minimization converged within the wall time. Lammps just does not stop after the last minimization and I have no idea of why. The last time it wrote something was at 7pm and it is not almost 11pm. lammps is still running and I have no idea of what is it doing.
here are the stat of the last minimization to show you the convergence.
Minimization stats:
Stopping criterion = energy tolerance
Energy initial, next-to-last, final =
-40388.9413657 -40388.984296 -40388.984296
Force two-norm initial, final = 0.184992 0.000106844
Force max component initial, final = 0.0129694 2.36187e-05
Final line search alpha, max atom move = 2.46447e-08 5.82077e-13
Iterations, force evaluations = 2363 17964
Pair time (\) = 236\.03 \(81\.2528\)
Neigh time \() = 0 (0)
Comm time (\) = 37\.0695 \(12\.7611\)
Outpt time \() = 1.16373 (0.400612)
Other time (%) = 16.2251 (5.58546)
And here, it the example of script I am using:
# Minimize after deformation to calculate a Peierls Stress
# we start with an unrelaxed configuration: first step, minimize with 0 displacement
units metal
boundary p s p
atom_style atomic
read_data data.meam
pair_style eam/fs
pair_coeff * * Mg_mm.eam.fs Mg
neighbor 2.0 bin
neigh_modify delay 0
#definition of the upper block
region upper block INF INF 323.491210937500000 INF INF INF units box
region lowerL block INF 6.4178000000000000009 INF 10.0000000000000000009 INF INF units box
region lowerM block 6.4178000000000000009 327.357793612670898 INF 10.0000000000000000009 INF INF units box
region lowerR block 327.357793612670898 INF INF 10.0000000000000000009 INF INF units box
region forstress1 block 6.4178000000000000009 327.357793612670898 INF 323.491210937500000 INF INF units box
group upper region upper
group lowerL region lowerL
group lowerM region lowerM
group lowerR region lowerR
group forstress1 region forstress1
displace_atoms lowerL move 0.0 0. 0. units box
displace_atoms lowerM move 0.0 0. 0. units box
displace_atoms lowerR move 0.0 0. 0. units box
fix 1 lowerL setforce NULL 0.0 NULL
fix 2 lowerR setforce NULL 0.0 NULL
fix 3 lowerM setforce NULL 0.0 NULL
fix 4 upper setforce NULL 0.0 NULL
thermo 1
thermo_style custom step f_3[1]
compute 2 all centro/atom
dump 3 all custom 50000000 dump.movie tag x y z c_2
dump 9 forstress1 custom 10000000 dump.syz.* fx
min_modify dmax 0.005
minimize 0.0 1.0e-14 100000 100000
unfix 1
unfix 2
unfix 3
unfix 4
fix 1 lowerL setforce 0.0 0.0 0.0
fix 2 lowerR setforce 0.0 0.0 0.0
fix 3 lowerM setforce 0.0 0.0 0.0
fix 4 upper setforce 0.0 0.0 0.0
displace_atoms lowerL move .03337775936126708979 0. 0. units box
displace_atoms lowerM move .03337775936126708979 0. 0. units box
displace_atoms lowerR move .03337775936126708979 0. 0. units box
min_modify dmax 0.005
minimize 0.0 1.0e-14 100000 100000
displace_atoms lowerL move .03337775936126708979 0. 0. units box
displace_atoms lowerM move .03337775936126708979 0. 0. units box
displace_atoms lowerR move .03337775936126708979 0. 0. units box
min_modify dmax 0.005
minimize 0.0 1.0e-14 100000 100000
displace_atoms lowerL move .03337775936126708979 0. 0. units box
displace_atoms lowerM move .03337775936126708979 0. 0. units box
displace_atoms lowerR move .03337775936126708979 0. 0. units box
min_modify dmax 0.005
minimize 0.0 1.0e-14 100000 100000
displace_atoms lowerL move .03337775936126708979 0. 0. units box
displace_atoms lowerM move .03337775936126708979 0. 0. units box
displace_atoms lowerR move .03337775936126708979 0. 0. units box
min_modify dmax 0.005
minimize 0.0 1.0e-14 100000 100000
displace_atoms lowerL move .03337775936126708979 0. 0. units box
displace_atoms lowerM move .03337775936126708979 0. 0. units box
displace_atoms lowerR move .03337775936126708979 0. 0. units box
min_modify dmax 0.005
minimize 0.0 1.0e-14 100000 100000
... (I repeated the last 5 lines 50 times).
I have the results of the last minimization adn everything is great but code still running.
Sebastien
Sebastien,
Are you using the current updated version of LAMMPS? If not, I'd strongly recommend that you update and try it again.
I had to slightly modify your input script to get it to parse properly, but it ran to completion here on my machine. The line I changed was as follows:
pair_coeff * * Mg_mm.eam.fs Mg Mg
I also assumed that you are using the Mg_mm.eam.fs and data.meam files as-is from the LAMMPS distribution. If not, please send the files that you're using.
As-is, the output from the run looked funny: zero everywhere for the f_3[1] quantity.
Here's the final output I'm seeing (pasted below). The run completed very quickly.
Paul
Setting up minimization ...
Memory usage per processor = 2.64327 Mbytes
Step 3[1]
2162 0
2163 0
Loop time of 0.016 on 1 procs for 1 steps with 128 atoms
Minimization stats:
Stopping criterion = failed linesearch
Energy initial, next-to-last, final =
-47.375139111 -47.375139111 -47.375139111
Force two-norm initial, final = 0 0
Force max component initial, final = 0 0
Final line search alpha, max atom move = 0 0
Iterations, force evaluations = 1 0
Pair time (\) = 0 \(0\)
Neigh time \() = 0 (0)
Comm time (\) = 0 \(0\)
Outpt time \() = 0 (0)
Other time (%) = 0.016 (100)
Nlocal: 128 ave 128 max 128 min
Histogram: 1 0 0 0 0 0 0 0 0 0
Nghost: 718 ave 718 max 718 min
Histogram: 1 0 0 0 0 0 0 0 0 0
Neighs: 12141 ave 12141 max 12141 min
Histogram: 1 0 0 0 0 0 0 0 0 0
FullNghs: 24282 ave 24282 max 24282 min
Histogram: 1 0 0 0 0 0 0 0 0 0
Total # of neighbors = 24282
Ave neighs/atom = 189.703
Neighbor list builds = 0
Dangerous builds = 0