(Solved) (Optimization task) calculation doesn't proceed at certain state

From NWChem

Viewed 2665 times, With a total of 5 Posts
Jump to: navigation, search

Clicked A Few Times
Threads 5
Posts 17
Hello everyone,




I'm trying to optimize a qmmm system.

Actually, what I really want to do is to calculate the Free energy of the system.
I saw the documentation page and found that I need .top / .rst / .xyzi / .esp files.
I think I can make .top and .rst file from prepare task, and .xyzi file by manual change of .xyz file.
I also found that .esp file can be obtained by optimization task or esp task.
At first I tried to do esp task but found a error. (I also want to ask about this error later...)
So now I'm trying to do optimize task.




When I run the NWCHEM, the calculation starts and the output file shows no error message.

The problem is that the calculation does not end.
When I open the output file (I'm using openmpi and able to open the .out file during calculation),
I see that the output file remains at certain script line and do not proceed anymore.

Now I'm waiting more than one hour.
I'm not sure whether NWCHEM need more time to proceed or not.
However the energy calculation in my task finished within 100sec. So I guess there is some error.


Here is the description about my system, the input file, and the last output script line




[System] I set solvation box and there is a amino-acid(tyrosine?) at the center.



[Input file]

start tyr

permanent_dir ./perm
scratch_dir ./scratch

md
system tyr_optimization
end

basis
* library "6-31G"
end

dft
xc b3lyp
iterations 500
convergence energy 1e-4
end

qmmm
region qm solvent
maxiter 100 1000
ncycles 5
density espfit
xyz xyz_tyr
end

task qmmm dft optimize




[Tail part of .out file]

                   NWChem DFT Gradient Module
--------------------------



charge = 0.00
wavefunction = closed shell



DFT ENERGY GRADIENTS

atom coordinates gradient
x y z x y z
1 C 1.020943 0.501344 -16.556675 0.027611 0.049265 0.023862
2 C 1.773129 -1.571252 -18.366403 0.025040 0.042327 -0.018517
3 C 0.631624 1.664066 -12.032204 -0.069200 0.003572 -0.033601
4 C -1.865233 1.828028 -11.172846 -0.008663 0.012754 0.010223
5 C -2.562252 3.750972 -9.495472 0.002641 -0.006136 -0.001670
6 C -0.764556 5.511519 -8.677121 -0.007786 0.014917 0.011669
7 C 1.732301 5.347557 -9.536479 0.002912 0.006850 0.009406
8 C 2.429895 3.423401 -11.215184 -0.000962 -0.008045 -0.005092
9 O -1.447872 7.392487 -7.036097 0.016169 -0.010029 -0.005915
10 N -1.693860 1.171657 -16.968381 0.012055 -0.009052 0.010000
11 C 0.000000 0.000000 22.490492 -0.005057 -0.003997 0.000969
12 O 3.471800 -1.216304 -19.884409 -0.068289 -0.054912 0.043898
13 O 0.639003 -3.686918 -18.306816 0.059139 0.027121 -0.019370
14 H -3.252258 0.469040 -11.803776 0.005123 0.006214 -0.003917
15 H -4.488996 3.877038 -8.832874 0.000656 -0.000740 -0.000128
16 H 3.119014 6.703927 -8.905982 -0.006194 -0.005727 -0.002011
17 H 4.356638 3.297336 -11.877782 -0.001450 -0.000934 -0.002433
18 H 0.213826 -2.067097 -13.511057 0.002625 -0.021090 0.047741
19 H 3.360676 -0.896246 -13.542224 0.022445 -0.034114 -0.045208
20 H 0.000000 8.447355 -6.654673 -0.003146 -0.001115 -0.006300
21 H 2.192494 2.159561 -16.886554 0.000203 -0.004876 -0.017041
22 H -1.940646 1.787807 -18.807798 0.006793 -0.005791 -0.001579
23 H -2.209890 2.593044 -15.727386 0.001415 0.004297 0.002539
24 H -2.808210 -0.405021 -16.653135 -0.011433 -0.002005 0.003300

----------------------------------------
| Time | 1-e(secs) | 2-e(secs) |
----------------------------------------
| CPU | 0.12 | 1.12 |
----------------------------------------
| WALL | 0.12 | 1.12 |
----------------------------------------
Bq nuclear interaction energy = 0.862503323818484





From the last line, calculation does not proceed anymore...

I see that even now, I mean when the calculation does not proceed anymore at above line, the NWCHEM is using memories and CPUs of computer.
(It is doing something but I can't see any progress, as like as my current state. )

Now I have no idea... Can you help me?

Thank you much in advance-
Edited On 10:58:28 PM PDT - Mon, Apr 2nd 2012 by Yjleedaniel

Clicked A Few Times
Threads 5
Posts 17
Still don't know what to do
I started the same calculation task yesterday and have waited more then 12 hours.

The reason that I have waited is to see whether this is just a problem of time; time needed in calculation.

I've run the job in parallel, using 34 number of CPUs. Now I guess this is not the problem about time.

Isn't there anyone who would like to help? Now I really don't know what to do...


Or, (actually, my purpose of optimization is to get the ".esp" file) is there any other method to get ".esp" file?

Clicked A Few Times
Threads 5
Posts 17
One clue
I found that the calculation proceeds without any trouble when I run the nwchem with single core.

It's strange because I could do other kind of tasks using multi core cpu.

I don't know the reason but now I'm trying to re-install NWCHEM with some different environment setup.

Clicked A Few Times
Threads 5
Posts 17
Solved
My advisor tried to re-install NWCHEM with a little bit different environment setup and now it works well.

These things are changed from previous setup.

1. Fortran and C compiler : from ifort&icc to mpif90. (didn't indicate C compiler)
2. BLASOPT libraries : changed and added
3. LIBMPI library : path was clarified
4. etc. : clarified path in OPENMPI setup file.

Clicked A Few Times
Threads 6
Posts 12
a question about QMMM NEB Calculation
Hello everyone,

Here is an example(from documentation) about QMMMM NEB Calculation:

memory total 1500 Mb

start cloh

permanent_dir ./perm
scratch_dir ./data
charge -1

basis "ao basis"
* library "6-31+G*"
end

dft
mult 2
XC b3lyp
iterations 5000
end

md
system cloh_neb
cutoff 1.0
end

qmmm
bqzone 10.0
region solvent
maxiter 1000
ncycles 1
density espfit
end

set qmmm:neb_path_limits cloh_rs.rst cloh_ps.rst
set neb:nbeads 10
set neb:stepsize 1.0
set neb:steps 5

task qmmm dft neb


My question is how to get cloh_rs.rst and cloh_ps.rst file, that is, if I want to get these .rst files , what calculation should I do.

Thanks for your attention

Clicked A Few Times
Threads 6
Posts 12
I'm really sorry, I made a mistake, I should not post here.


Forum >> NWChem's corner >> QMMM



Who's here now Members 0 Guests 0 Bots/Crawler 1


AWC's: 2.5.10 MediaWiki - Stand Alone Forum Extension
Forum theme style by: AWC