|
|
老师您好，我之前在做蛋白的模拟的时候，正式MD过程使用了GPU加速，但是现在做蛋白-配体的模拟，发现使用GPU会报错，不使用GPU则没有问题Command line:
gmx_mpi mdrun -deffnm md01 -nb gpu
Reading file md01.tpr, VERSION 2016.4 (single precision)
Note: file tpx version 110, software tpx version 112
Multiple energy groups is not implemented for GPUs, falling back to the CPU. For better performance, run on the GPU without energy groups and then do gmx mdrun -rerun option on the trajectory with an energy group .tpr file.
-------------------------------------------------------
Program: gmx mdrun, version 2018.3
Source file: src/gromacs/taskassignment/decidegpuusage.cpp (line 292)
Function: bool gmx::decideWhetherToUseGpusForNonbonded(const gmx::TaskTarget, const std::vector<int> &, const EmulateGpuNonbonded, const bool, const bool, const bool)
Inconsistency in user input:
Nonbonded interactions on the GPU were required, but not supported for these
simulation settings. Change your settings, or do not require using GPUs.
这是我的mdp文件参数
integrator = md
dt = 0.002
nsteps = 250000
nstxout = 500
nstvout = 500
nstfout = 500
nstenergy = 500
nstlog = 500
energygrps = Protein_CAL NA_SOL
nstlist = 1000
ns-type = Grid
pbc = xyz
rlist = 300.0
cutoff-scheme = verlet
coulombtype = PME
pme_order = 4
fourierspacing = 3
rcoulomb = 300.0
vdw-type = Cut-off
rvdw = 300.0
Tcoupl = V-rescale
tc-grps = Protein_CAL NA_SOL
tau_t = 3 3
ref_t = 370 370
DispCorr = EnerPres
Pcoupl = no
gen_vel = no
constraints = all-bonds
continuation = yes
constraint_algorithm = lincs
lincs_iter = 1
lincs_order = 4
E-z = 1 0 0
请问老师这是说不能设置energygrps吗？
还有老师，如果使用nose-hoover热浴，会出现这个错误，是要扩大tau—t的值吗？
|
|
发表于 Post on 2019-7-13 15:38:11
|
只看该作者 Only view this author
|只看大图 View big picture
|倒序浏览 Reverse view
|阅读模式 Reading model
收藏 Add to favorites
楼主 Author
发表于 Post on 2019-7-14 01:15:07
