GROMACS version: 2022
GROMACS modification: No
Hello. I’m trying to run a simulation across multiple nodes with HMR (4fs timestep) but I am plagued by small errors such as.
relative constraint deviation after LINCS:
rms 0.001098, max 0.010735 (between atoms 27733 and 27736)
bonds that rotated more than 30 degrees:
atom 1 atom 2 angle previous, current, constraint length
27733 27736 31.4 0.1111 0.1099 0.1111
I have found that these disappear on a single node when I use the option “constraints = all-bonds” but when I run this on multiple nodes it crashes with the error.
“Fatal error:
6 particles communicated to PME rank 38 are more than 2/3 times the cut-off
out of the domain decomposition cell of their charge group in dimension y.
This usually means that your system is not well equilibrated.”
It feels odd that I get different behaviour when I run over single or multiple nodes. Are the small LINCS errors just an artifact of the domain decomposition or do they speak to an instability in my system? The rotations aren’t much larger than the lincs warning threshold. Below is my mdp file. I will do an analysis of the energetics if you think it might be helpful.
integrator = md
dt = 0.004
nsteps = 500000000
nstxtcout = 50000
nstcalcenergy = 100
nstenergy = 1000
nstlog = 1000
;
cutoff-scheme = Verlet
nstlist = 20
rlist = 1.2
vdwtype = Cut-off
vdw-modifier = Force-switch
rvdw_switch = 1.0
rvdw = 1.2
coulombtype = PME
rcoulomb = 1.2
;
tcoupl = Nose-Hoover
tc_grps = SOLU MEMB SOLV
tau_t = 1.0 1.0 1.0
ref_t = 310 310 310
;
pcoupl = Parrinello-Rahman
pcoupltype = semiisotropic
tau_p = 5.0
compressibility = 4.5e-5 4.5e-5
ref_p = 1.0 1.0
;
constraints = h-bonds
constraint_algorithm = LINCS
lincs-order = 6
continuation = yes
;
nstcomm = 100
comm_mode = linear
comm_grps = SOLU_MEMB SOLV
~