Topic Options
#29809 - 04/12/12 01:36 AM GCMC parallel error
Manish Kesherwan Offline
Forum Member

Registered: 10/22/11
Posts: 50
Good Morning sir,
I am running absolute free energy calculation using GCMC/SSBP method by charmm gui provided script In step 5.1_site_fes.pbs giving these below error. I equilibrated system water molecule of active site sphere for 10ps trajactory 20 times .
This step5.1 script is running in serial mode but it's not runnig in parallel version using MPI Please give me some suggestion why this is showing error.
I installed mpich2-1.2.1p1 and charmm c35b6
using
./install.com em64t xxlarge M MPICH +CGENFF

Script error in step5.1_site_fes calculation
CHARMM> nbonds extend grad quad group switch cdie eps 1.0 -
CHARMM> vgroup vswitch -
CHARMM> ctonnb 14.5 ctofnb 14.5 cutnb 14.5 wmin 1.2 wrnmxd 3.5 ! wrnmxd 1.2
WARNING: NBOND CONFIGURATION HAS BEEN CHANGED SINCE LAST UPDATE


NONBOND OPTION FLAGS:
ELEC VDW GROUps CDIElec SWITch VGROup VSWItch
BYGRoup EXTEnd GRADient QUADripo NOEWald
CUTNB = 14.500 CTEXNB =999.000 CTONNB = 14.500 CTOFNB = 14.500
WMIN = 1.200 WRNMXD = 3.500 E14FAC = 1.000 EPS = 1.000
WARNING: NBOND CONFIGURATION HAS BEEN CHANGED SINCE LAST UPDATE

NBXMOD = 5
There are 0 atom pairs and 0 atom exclusions.
There are 0 group pairs and 0 group exclusions.

***** LEVEL 1 WARNING FROM *****
***** CUTNB and CTOFNB are too close for efficient heuristic update.
******************************************
BOMLEV ( 0) IS NOT REACHED. WRNLEV IS 5

GTNBCT> CUTNB,CTOFNB,CTONNB= 14.5 14.5 14.5

***** LEVEL 1 WARNING FROM *****
***** CUTNB,CTOFNB,CTONNB are not in correct order.
******************************************
BOMLEV ( 0) IS NOT REACHED. WRNLEV IS 5


***** LEVEL 1 WARNING FROM *****
***** CUTNB and CTOFNB are too close for efficient heuristic update.
******************************************
BOMLEV ( 0) IS NOT REACHED. WRNLEV IS 5


***** LEVEL 1 WARNING FROM *****
***** CUTNB,CTOFNB,CTONNB are not in correct order.
******************************************
BOMLEV ( 0) IS NOT REACHED. WRNLEV IS 5


***** LEVEL -4 WARNING FROM *****
***** A SHARED DATA STRUCTURE MAY NOT BE RESIZED
******************************************
BOMLEV ( 0) IS REACHED - TERMINATING. WRNLEV IS 5

Top
#29811 - 04/12/12 10:10 AM Re: GCMC parallel error [Re: Manish Kesherwan]
mgarraha Offline
Forum Member

Registered: 05/13/09
Posts: 125
I don't think any of the MC stuff is MPI safe.

Top
#29812 - 04/12/12 11:01 AM Re: GCMC parallel error [Re: Manish Kesherwan]
Manish Kesherwan Offline
Forum Member

Registered: 10/22/11
Posts: 50
Sorry sir, After nbond above command program is getting terminated and not allowing to proceed next command of script that' why i am worrying and i want to run full script.please give some suggestion

Top
#29813 - 04/12/12 11:16 AM Re: GCMC parallel error [Re: Manish Kesherwan]
Manish Kesherwan Offline
Forum Member

Registered: 10/22/11
Posts: 50
It's also showing


CHARMM> nbonds extend grad quad group switch cdie eps 1.0 -
CHARMM> vgroup vswitch -
CHARMM> ctonnb 14.5 ctofnb 14.5 cutnb 14.5 wmin 1.2 wrnmxd 3.5 ! wrnmxd 1.2
WARNING: NBOND CONFIGURATION HAS BEEN CHANGED SINCE LAST UPDATE
WARNING: NBOND CONFIGURATION HAS BEEN CHANGED SINCE LAST UPDATE

WARNING: NBOND CONFIGURATION HAS BEEN CHANGED SINCE LAST UPDATE


***** LEVEL 1 WARNING FROM *****
***** CUTNB and CTOFNB are too close for efficient heuristic update.
******************************************
BOMLEV ( 0) IS NOT REACHED. WRNLEV IS 5


***** LEVEL 1 WARNING FROM *****
***** CUTNB,CTOFNB,CTONNB are not in correct order.
******************************************
BOMLEV ( 0) IS NOT REACHED. WRNLEV IS 5

WARNING: NBOND CONFIGURATION HAS BEEN CHANGED SINCE LAST UPDATE
WARNING: NBOND CONFIGURATION HAS BEEN CHANGED SINCE LAST UPDATE
WARNING: NBOND CONFIGURATION HAS BEEN CHANGED SINCE LAST UPDATE





***** LEVEL -4 WARNING FROM *****
***** A SHARED DATA STRUCTURE MAY NOT BE RESIZED
******************************************
BOMLEV ( 0) IS REACHED - TERMINATING. WRNLEV IS 5



/---------\
/ \
/ \
/ \
! XXXX XXXX !
! XXXX XXXX !
! XXX XXX !
! X !
--\ XXX /--
! ! XXX ! !
! ! ! !
! I I I I I !
! I I I I !
\ /
-- --


\---/
XXX XXX
XXXX XXXX
XXXXX XXXXX
XXX XXX
XXX XXX
XXXXX
XXX XXX
XXX XXX
XXX XXX
XXXXX XXXXX
XXXX XXXX
XXX XXX


Execution terminated due to the detection of a fatal error.

ABNORMAL TERMINATION
MAXIMUM STACK SPACE USED IS 44810
STACK CURRENTLY IN USE IS 32994
MOST SEVERE WARNIN ***** LEVEL 1 WARNING FROM *****
***** LEVEL 1 WARNING FROM *****
***** LEVEL 1 WARNING FROM *****
***** CUTNB and CTOFNB are too close for efficient heuristic update.
******************************************
G WAS AT LEVEL -4

Top
#29814 - 04/12/12 11:16 AM Re: GCMC parallel error [Re: Manish Kesherwan]
Manish Kesherwan Offline
Forum Member

Registered: 10/22/11
Posts: 50
What's mean of A SHARED DATA STRUCTURE MAY NOT BE RESIZED

Top
#29815 - 04/12/12 11:24 AM Re: GCMC parallel error [Re: Manish Kesherwan]
Manish Kesherwan Offline
Forum Member

Registered: 10/22/11
Posts: 50
and this step energy is not getting able to calculate in parallel version but in single process it's calculating energy and proceding further steps of script.
parallel
mpiexec -np 8 charmm gcomp=grmsd win=1 ccycle=1 step5_output.out
application called MPI_Abort(MPI_COMM_WORLD, 12) - process 2
application called MPI_Abort(MPI_COMM_WORLD, 12) - process 3
application called MPI_Abort(MPI_COMM_WORLD, 12) - process 4
one processor
charmm gcomp=grmsd win=1 ccycle=1 step5_output.out
this command is working and calculating energy and further commands of script

NONBOND OPTION FLAGS:
ELEC VDW GROUps CDIElec SWITch VGROup VSWItch
BYGRoup EXTEnd GRADient QUADripo NOEWald
CUTNB = 14.500 CTEXNB =999.000 CTONNB = 14.500 CTOFNB = 14.500
WMIN = 1.200 WRNMXD = 3.500 E14FAC = 1.000 EPS = 1.000
NBXMOD = 5
There are 0 atom pairs and 14946 atom exclusions.
There are 160270 group pairs and 2123 group exclusions.

***** LEVEL 1 WARNING FROM *****
***** CUTNB and CTOFNB are too close for efficient heuristic update.
******************************************
BOMLEV ( 0) IS NOT REACHED. WRNLEV IS 5

GTNBCT> CUTNB,CTOFNB,CTONNB= 14.5 14.5 14.5

***** LEVEL 1 WARNING FROM *****
***** CUTNB,CTOFNB,CTONNB are not in correct order.
******************************************
BOMLEV ( 0) IS NOT REACHED. WRNLEV IS 5

Generating nonbond list with Exclusion mode = 5
== PRIMARY == SPACE FOR 1255618 ATOM PAIRS AND 221748 GROUP PAIRS

General group nonbond list generation found:
160270 GROUP PAIRS WERE FOUND FOR GROUP LIST
585670 PAIRS USED EXTENDED ELECTROSTATICS
0 GROUP PAIRS WERE BEYOND CUTOFFS
91 GROUP PAIRS WERE NOT SELECTED
0 QM group pairs excluded
ENER ENR: Eval# ENERgy Delta-E GRMS
ENER INTERN: BONDs ANGLes UREY-b DIHEdrals IMPRopers
ENER CROSS: CMAPs
ENER EXTERN: VDWaals ELEC HBONds ASP USER
ENER IMAGES: IMNBvdw IMELec IMHBnd RXNField EXTElec
ENER PBEQ: PBnp PBelec GBEnr GSBP
---------- --------- --------- --------- --------- ---------
ENER> 0 -8569.44988 0.00000 15.71324
ENER INTERN> 349.53469 818.64369 103.02668 622.87375 46.02293
ENER CROSS> -160.70674
ENER EXTERN> -171.79804 -9082.08639 0.00000 0.00000 0.00000
ENER IMAGES> 0.00000 0.00000 0.00000 0.00000 -26.79831
ENER PBEQ> 0.00000 0.00000 0.00000 -1068.16215
---------- --------- --------- --------- --------- ---------

CHARMM>

CHARMM> !
CHARMM> !flat bottom restraint to restrain the symmetrical rotatable parts to a single conformation
CHARMM> !
CHARMM>

Top
#29816 - 04/12/12 12:11 PM Re: GCMC parallel error [Re: Manish Kesherwan]
rmv Online   content

Forum Member

Registered: 09/17/03
Posts: 8314
Loc: 39 03 48 N, 77 06 54 W
Your symptoms are entirely consistent with the reply-- that the MC code has not been suitably modified to run in parallel. The short term solution is to use only a single processor for these calculations.
_________________________
Rick Venable
computational chemist


Top

Moderator:  BRBrooks, bucknerj, chmgr, John Legato