How to Control CP2K output file generation?

563 views
Skip to first unread message

Anthony Debellis

unread,
Mar 20, 2017, 4:33:07 PM3/20/17
to cp2k
I have been running a metadynamics calculation with the enclosed input sections.  As the run progresses, restart files are generated every 500 steps.  From the input I would guess that they should be generated every 10000 steps because of the input:

&RESTART

&EACH

MD 10000

&END

&END


The corresponding section for trajectory output is similar:

&TRAJECTORY

&EACH

MD 50

&END

&END


and the trajectory file contains coordinates dumped every 50 steps (as I would expect).

Furthermore, at the start of the run, the HILLS.metadynLog files are created first at 150 steps, and every 150 steps thereafter until 2150, at which point no additional files are created even though the simulation continues to run (presently at 10900 steps) as can be seen in the .ener file.  In addition, the HILLS files each contain exactly one line.  I would like a single HILLS file that contains each step.

I obviously do not understand how to control cp2k's output.  The documentation does not help.  Can someone please explain the correct use of the keywords to control output?  (For example, what does COMMON_ITERATION_LEVELS mean?)

Anthony
===========================================================================

&MOTION

&MD

ENSEMBLE NVT

STEPS 100000

TIMESTEP 1.0

TEMPERATURE 300.0

ANGVEL_ZERO

COMVEL_TOL 2.0E-7

&THERMOSTAT

&NOSE

TIMECON 100.0

&END NOSE

 

# TYPE CSVR

# &CSVR

# TIMECON 100

# &END

 

&END THERMOSTAT

&PRINT

&ENERGY

&EACH

MD 100

&END

&END

 

# &PROGRAM_RUN_INFO

# &EACH

# MD 100

# &END

# &END

&END PRINT

&END MD

 

&FREE_ENERGY

METHOD METADYN

&METADYN

DO_HILLS .TRUE.

NT_HILLS 150

MIN_NT_HILLS 50

MIN_DISP 0.03

WW 1.0e-3

&METAVAR

COLVAR 1

SCALE 0.5

&END METAVAR

 

&PRINT

&COLVAR

COMMON_ITERATION_LEVELS 1

&EACH

MD 50

&END

&END

&HILLS

# COMMON_ITERATION_LEVELS 1

&EACH

MD 50

&END

&END

&END PRINT

&END METADYN

&END FREE_ENERGY

 

&PRINT

&TRAJECTORY

&EACH

MD 50

&END

&END

&VELOCITIES OFF

&END

&RESTART

&EACH

MD 10000

&END

# ADD_LAST NUMERIC

&END

# &RESTART_HISTORY

# &EACH

# MD 2000

# &END

# &END

&END PRINT

 

&END MOTION

Marcella Iannuzzi

unread,
Mar 21, 2017, 5:15:00 AM3/21/17
to cp2k
 Hi Antony

To learn more on the common iteration level, see the answer of Teo to the post 'Printing of wannier localisation center' 
of January 2008
regards 
Marcella

Anthony Debellis

unread,
Mar 21, 2017, 4:18:37 PM3/21/17
to cp2k
I'm sorry Marcella, but I don't understand the reply by Teo to that post.  Can I ask if you would try to explain and to answer directly the questions I have in my post?  That would be greatly appreciated.
Reply all
Reply to author
Forward
0 new messages