求助：模拟退火遇到错误

wh.shen · 发表于 Post on 2019-10-28 14:01:06

各位老师，我在模拟退火时遇到如下错误，以下是代码和错误信息，求指导
代码：

<b>constraints = h-bonds
constraint_algorithm = LINCS
lincs_order = 4
;shake_tol = 0.0001
comm_mode = Linear
comm_grps = Protein Non-protein
integrator = md
annealing = periodic periodic
annealing_npoints = 3 3
annealing_time = 0 5 10 0 5 10
annealing_temp = 300 400 500 300 400 500
dt = 0.002 ; ps !
nsteps = 5000000 ; total 10 ns.
nstcomm = 10
nstxout = 10000 ; collect data every 10 ps
nstenergy = 10000
nstvout = 10000
nstlog = 10000
nstxtcout = 5000
xtc_grps = non-water
nstfout = 0
nstlist = 10
ns_type = grid
rlist = 1.0
; Electrostatics
coulombtype = PME
rcoulomb = 1
; van der Waals
vdw-type = switch
rvdw-switch = 0.8
rvdw = 1.0
fourierspacing = 0.12
fourier_nx = 0
fourier_ny = 0
fourier_nz = 0
pme_order = 4
ewald_rtol = 1e-5
optimize_fft = yes
energygrps = system
; Berendsen temperature coupling is on in two groups
Tcoupl = v-rescale ;nose-hoover
tau_t = 0.5 0.5
tc-grps = Protein Non-protein
ref_t = 300 300
; Pressure coupling is on
Pcoupl = Berendsen
tau_p = 1.0
compressibility = 4.5e-5
ref_p = 1.0
refcoord_scaling = com
; Generate velocites is on at 300 K.
gen_vel = yes
gen_temp = 300.0
gen_seed = -1
</b>

复制代码

错误信息：

<b>Removing all charge groups because cutoff-scheme=Verlet
Simulated annealing for group Protein: Periodic, 3 timepoints
Time (ps) Temperature (K)
0.0 0.0
0.0 0.0
0.0 0.0
Simulated annealing for group non-Protein: Periodic, 3 timepoints
Time (ps) Temperature (K)
0.0 0.0
0.0 0.0
0.0 0.0
Number of degrees of freedom in T-Coupling group Protein is 185455.00
Number of degrees of freedom in T-Coupling group non-Protein is 77298.00
[node03:88987] *** Process received signal ***
[node03:88987] Signal: Segmentation fault (11)
[node03:88987] Signal code: Address not mapped (1)
[node03:88987] Failing at address: 0x41200008
[node03:88987] [ 0] /lib64/libpthread.so.0(+0xf370) [0x2b185cd40370]
[node03:88987] [ 1] /opt/software/openmpi-1.6.5/lib/libmpi.so.1(opal_memory_ptmalloc2_int_malloc+0x2cd) [0x2b185bd4afad]
[node03:88987] [ 2] /opt/software/openmpi-1.6.5/lib/libmpi.so.1(opal_memory_ptmalloc2_int_memalign+0x52) [0x2b185bd4d942]
[node03:88987] [ 3] /opt/software/openmpi-1.6.5/lib/libmpi.so.1(opal_memory_ptmalloc2_memalign+0xbf) [0x2b185bd4dd7f]
[node03:88987] [ 4] /lib64/libstdc++.so.6(_Znwm+0x1d) [0x2b18684350cd]
[node03:88987] [ 5] gmx_mpi_2019_s_plumed(_ZNSt6vectorI17VerletbufAtomtypeSaIS0_EE19_M_emplace_back_auxIIS0_EEEvDpOT_+0x42) [0x832fb2]
[node03:88987] [ 6] gmx_mpi_2019_s_plumed() [0x82e248]
[node03:88987] [ 7] gmx_mpi_2019_s_plumed(_Z23calc_verlet_buffer_sizePK10gmx_mtop_tfPK10t_inputreciifPK18VerletbufListSetupPiPf+0x49a) [0x82e6ea]
[node03:88987] [ 8] gmx_mpi_2019_s_plumed(_Z10gmx_gromppiPPc+0x705f) [0x719abf]
[node03:88987] [ 9] gmx_mpi_2019_s_plumed(_ZN3gmx24CommandLineModuleManager3runEiPPc+0x222) [0x4f39a2]
[node03:88987] [10] gmx_mpi_2019_s_plumed(main+0x7c) [0x4e6b1c]
[node03:88987] [11] /lib64/libc.so.6(__libc_start_main+0xf5) [0x2b1868b3cb35]
[node03:88987] [12] gmx_mpi_2019_s_plumed() [0x4eb61d]
[node03:88987] *** End of error message ***
/var/spool/torque/mom_priv/jobs/7094.node01.SC: 行 11: 88987 段错误 (吐核)gmx_mpi_2019_s_plumed grompp -f annealing.mdp -c shear_14ns.gro -p silk.top -o annealing.tpr -maxwarn 2
</b>

复制代码

sobereva · 发表于 Post on 2019-10-28 14:42:22

mdp没有显著问题
去掉gen_vel = yes、减小步长再试
还不行尝试非MPI版

wh.shen · 发表于 Post on 2019-10-28 18:46:44

本帖最后由 wh.shen 于 2019-10-28 19:25 编辑

sobereva 发表于 2019-10-28 14:42
mdp没有显著问题
去掉gen_vel = yes、减小步长再试
还不行尝试非MPI版

谢谢soberva老师，去掉速度可以了，但是gromacs忽略了我mdp的设置退火的时间和温度，生成的log显示温度一直控制在0k，我把温度输出同样是0k。

grpopts:
nrdf: 185455 77298
ref-t: 300 300
tau-t: 0.5 0.5
annealing: Periodic Periodic
annealing-npoints: 2 2
annealing-time [0]: 0.0 0.0
annealing-temp [0]: 0.0 0.0
annealing-time [1]: 0.0 0.0
annealing-temp [1]: 0.0 0.0
acc: 0 0 0
nfreeze: N N N
energygrp-flags[ 0]: 0<div class="blockcode"><blockquote>@ title "GROMACS Energies"
@ xaxis label "Time (ps)"
@ yaxis label "(K)"
@TYPE xy
@ view 0.15, 0.15, 0.75, 0.85
@ legend on
@ legend box on
@ legend loctype view
@ legend 0.78, 0.8
@ legend length 2
@ s0 legend "Temperature"
0.000000 309.465393
20.000000 0.033040
40.000000 0.046505
60.000000 0.085721
80.000000 0.157258
100.000000 0.273109
120.000000 0.473926
140.000000 0.736583
160.000000 1.055009

复制代码

sobereva · 发表于 Post on 2019-10-29 07:16:44

我不晓得你具体怎么弄的，gen_vel和退火设定的参考温度的变化在原理上没有任何直接关系。如果确认是程序bug，可以发到mailing list问问

wh.shen · 发表于 Post on 2019-10-30 20:31:40

sobereva 发表于 2019-10-29 07:16
我不晓得你具体怎么弄的，gen_vel和退火设定的参考温度的变化在原理上没有任何直接关系。如果确认是程序bug ...

用gromacs-2016.5就可以输出正确的淬火温度，而用gromacs-2019.5就得到上面的结果，而2019的手册关于淬火的命令也没有丝毫变化，很奇怪啊，bug不应该越来越少吗？

sobereva · 发表于 Post on 2019-10-31 05:52:29

wh.shen 发表于 2019-10-30 20:31
用gromacs-2016.5就可以输出正确的淬火温度，而用gromacs-2019.5就得到上面的结果，而2019的手册关于淬火 ...

如果bug越来越少，那就没bug了，显然是不可能的
加入新功能、做新改进的时候总免不了伴随bug的出现

		自动登录 Automatic login	找回密码 Forget password
密码 Password			注册 Register

[GROMACS] 求助：模拟退火遇到错误

浏览过的版块