Dear Amber users,
We ran into a problem with MD simulation using AMBER6 on the Beowulf
cluster (RedHat, Myranet, PG compiler). To run an MPI job on 8 processors
we used "mpirun.ch_gm" script. Calculation stops abnormally. Could you
help us to find out what is going on? If anyone has some idea? I am
attaching the output file and error message(see below).
********************************************************************
# message - sander_06135:
Atom division among processors for gb:
| 0 1402 2804 4206 5608 7010 8412 9814
| 11220
| Running AMBER/MPI version on 8 nodes
Sum of charges from parm topology file = 0.00000000
Forcing neutrality...
---------------------------------------------------
APPROXIMATING switch and d/dx switch using CUBIC SPLINE INTERPOLATION
using 5000.0 points per unit in tabled values
TESTING RELATIVE ERROR over r ranging from 0.0 to cutoff
---------------------------------------------------
APPROXIMATING switch and d/dx switch using CUBIC SPLINE INTERPOLATION
using 5000.0 points per unit in tabled values
TESTING RELATIVE ERROR over r ranging from 0.0 to cutoff
---------------------------------------------------
APPROXIMATING switch and d/dx switch using CUBIC SPLINE INTERPOLATION
using 5000.0 points per unit in tabled values
TESTING RELATIVE ERROR over r ranging from 0.0 to cutoff
| Atom division among processors:
| 0 1404 2805 4206 5610 7011 8412 9816
| 11220
| Atom division among processors for gb:
| 0 1402 2804 4206 5608 7010 8412 9814
| 11220
| Running AMBER/MPI version on 8 nodes
Sum of charges from parm topology file = 0.00000000
Forcing neutrality...
---------------------------------------------------
APPROXIMATING switch and d/dx switch using CUBIC SPLINE INTERPOLATION
using 5000.0 points per unit in tabled values
TESTING RELATIVE ERROR over r ranging from 0.0 to cutoff
---------------------------------------------------
APPROXIMATING switch and d/dx switch using CUBIC SPLINE INTERPOLATION
using 5000.0 points per unit in tabled values
TESTING RELATIVE ERROR over r ranging from 0.0 to cutoff
---------------------------------------------------
APPROXIMATING switch and d/dx switch using CUBIC SPLINE INTERPOLATION
using 5000.0 points per unit in tabled values
TESTING RELATIVE ERROR over r ranging from 0.0 to cutoff
| Atom division among processors:
| 0 1404 2805 4206 5610 7011 8412 9816
| 11220
| Atom division among processors for gb:
| 0 1402 2804 4206 5608 7010 8412 9814
| 11220
| Running AMBER/MPI version on 8 nodes
Sum of charges from parm topology file = 0.00000000
Forcing neutrality...
---------------------------------------------------
APPROXIMATING switch and d/dx switch using CUBIC SPLINE INTERPOLATION
using 5000.0 points per unit in tabled values
TESTING RELATIVE ERROR over r ranging from 0.0 to cutoff
| CHECK switch(x): max rel err = 0.3242E-14 at 2.436720
| CHECK d/dx switch(x): max rel err = 0.8064E-11 at 2.761360
---------------------------------------------------
Total number of mask terms = 12578
Total number of mask terms = 25156
| CHECK switch(x): max rel err = 0.3242E-14 at 2.436720
| CHECK d/dx switch(x): max rel err = 0.8064E-11 at 2.761360
---------------------------------------------------
???????????????????
---------------------------------------------------
Total number of mask terms = 12578
Total number of mask terms = 25156
| CHECK switch(x): max rel err = 0.3242E-14 at 2.436720
| CHECK d/dx switch(x): max rel err = 0.8064E-11 at 2.761360
---------------------------------------------------
Total number of mask terms = 12578
Total number of mask terms = 25156
| Total Ewald setup time = 0.14000000
------------------------------------------------------------------------------
| Total Ewald setup time = 0.15000000| Total Ewald setup time =
0.19000000|
Total Ewald setup time = 0.19000000
Total Ewald setup time = 0.20000000| Total Ewald setup time =
0.22000000|
Total Ewald setup time = 0.19000000
Unit 7 Error on OPEN:
[7] MPI Abort by user Aborting program !
[7] Aborting program!
[4] MPI Abort by user Aborting program !
[4] Aborting program!
done
*****************************************************************************
# output file:
mdrest_5.out
-------------------------------------------------------
Amber 6 SANDER Scripps/UCSF 1999
-------------------------------------------------------
| Fri Oct 18 13:38:37 2002
File Assignments:
|MDIN : mdrst_5.in
|MDOUT: mdrst_5.out
|INPCR: mdrst_4_ahg21_W.xyz
|PARM : ahg21_W.top
|RESTR: mdrst_5_ahg21_W.xyz
|REFC : mdrst_4_ahg21_W.xyz
|MDVEL: mdvel
|MDEN : mden
|MDCRD: mdrst_5_ahg21_W.traj
|MDINF: test.info
Here is the input file:
MD run(mdrst_5.in) for P=const (ntb=2)with force constant 10.0 kcal/mol
&cntrl
imin=0, irest=1, ntx=7,
ntt=1, tempi=283.0, temp0=283.0, tautp=2.0,
ntb=2, ntp=1,
ntc=2, tol=0.000001,
scee=1.2, cut=9.0,
ntwx=100, ntpr=100,
nstlim=10000,
ntr=1,
&end
Group input for restrained atoms (Harmonic rest. on solute coord.)
10.0
RES 1 21
END
END
1. RESOURCE USE:
getting box info from bottom of parm
getting new box info from bottom of inpcrd
| peek_ewald_inpcrd: Box info found
EWALD SPECIFIC INPUT:
| Using the T3D specific (FFT3D0) Fast Fourier Transform
-------------------------------------------------
NO EWALD INPUT FOUND: USING DEFAULTS
-------------------------------------------------
Largest sphere to fit in unit cell has radius = 18.431
Calculating ew_coeff from dsum_tol,cutoff
Box X = 78.705 Box Y = 36.863 Box Z = 38.447
Alpha = 90.000 Beta = 90.000 Gamma = 90.000
NFFT1 = 80 NFFT2 = 36 NFFT3 = 40
Cutoff= 9.000 Tol =0.100E-04
Ewald Coefficient = 0.30768
Interpolation order = 4
NATOM = 11220 NTYPES = 16 NBONH = 11068 MBONA = 158
NTHETH = 334 MTHETA = 222 NPHIH = 687 MPHIA = 473
NHPARM = 0 NPARM = 0 NNB = 16230 NRES = 3662
NBONA = 158 NTHETA = 222 NPHIA = 473 NUMBND = 34
NUMANG = 67 NPTRA = 34 NATYP = 27 NPHB = 4
IFBOX = 1 NMXRS = 41 IFCAP = 0
EWALD MEMORY USE:
| Total heap storage needed = 1161
| Adjacent nonbond minimum mask = 16230
| Max number of pointers = 25
| List build maxmask = 32460
| Maximage = 16418
EWALD LOCMEM POINTER OFFSETS
| Real memory needed by PME = 1161
| Size of EEDTABLE = 20768
| Real memory needed by EEDTABLE = 83072
| Integer memory needed by ADJ = 32460
| Integer memory used by local nonb= 271519
| Real memory used by local nonb = 183894
| MAX NONBOND PAIRS = 5000000
| Memory Use Allocated Used
| Real 2500000 784668
| Hollerith 600000 70984
| Integer 2000000 561701
| Max Nonbonded Pairs: 5000000
BOX TYPE: RECTILINEAR
2. CONTROL DATA FOR THE RUN
TIMLIM= 999999. IREST = 1 IBELLY= 0
IMIN = 0
IPOL = 0
NTX = 7 NTXO = 1
IG = 71277 TEMPI = 283.00 HEAT = 0.000
NTB = 2 BOXX = 78.705
BOXY = 36.863 BOXZ = 38.447
NTT = 1 TEMP0 = 283.000
DTEMP = 0.000 TAUTP = 2.000
VLIMIT= 0.000
NTP = 1 PRES0 = 1.000 COMP = 44.600
TAUP = 0.200 NPSCAL= 0
NTCM = 0 NSCM = 9999999
NSTLIM=10000 NTU = 1
T = 0.000 DT = 0.00100
NTC = 2 TOL = 0.00000 JFASTW = 0
NTF = 1 NSNB = 25
CUT = 9.000 SCNB = 2.000
SCEE = 1.200 DIELC = 1.000
NTPR = 100 NTWR = 50 NTWX = 100
NTWV = 0 NTWE = 0 IOUTFM= 0
NTWPRT= 0 NTWPR0= 0 NTAVE= 0
NTR = 1 NTRX = 1
TAUR = 0.00000 NMROPT= 0 PENCUT= 0.10000
IVCAP = 0 MATCAP= 0 FCAP = 1.500
OTHER DATA:
IFCAP = 0 NATCAP= 0 CUTCAP= 0.000
XCAP = 0.000 YCAP = 0.000 ZCAP = 0.000
VRAND= 0
NATOM = 11220 NRES = 3662
Water definition for fast triangulated model:
Resname = WAT ; Oxygen_name = O ; Hyd1_name = H1 ; Hyd2_name = H2
| PLEVEL = 1: runmd parallelization, no EKCMR
LOADING THE CONSTRAINED ATOMS AS GROUPS
5. REFERENCE ATOM COORDINATES
----- READING GROUP 1; TITLE:
Group input for restrained atoms (Harmonic rest. on solute coord.)
GROUP 1 HAS HARMONIC CONSTRAINTS 10.00000
GRP 1 RES 1 TO 21
Number of atoms in this group = 297
----- END OF GROUP READ -----
3. ATOMIC COORDINATES AND VELOCITIES
Largest sphere to fit in unit cell has radius = 18.431
NEW EWALD BOX PARAMETERS from inpcrd file:
A = 78.70522 B = 36.86298 C = 38.44706
ALPHA = 90.00000 BETA = 90.00000 GAMMA = 90.00000
begin time read from input coords = 50.000 ps
Number of triangulated 3-point waters found: 3641
| Atom division among processors:
| 0 1404 2805 4206 5610 7011 8412 9816
| 11220
| Atom division among processors for gb:
| 0 1402 2804 4206 5608 7010 8412 9814
| 11220
| Running AMBER/MPI version on 8 nodes
Sum of charges from parm topology file = 0.00000000
Forcing neutrality...
---------------------------------------------------
APPROXIMATING switch and d/dx switch using CUBIC SPLINE INTERPOLATION
using 5000.0 points per unit in tabled values
TESTING RELATIVE ERROR over r ranging from 0.0 to cutoff
| CHECK switch(x): max rel err = 0.3242E-14 at 2.4367
***************************************************************
I very sorry for such big file.
Thanks a lot,
Alexander Rubinshtein, Ph.D.
UNMC Eppley Cancer Center
Molecular Modeling Core Facility
_________________________________
University of Nebraska Medical Center
986805 Nebraska Medical Center
Omaha, Nebraska 68198-6805
USA
Office: (402) 559-7809
Fax: (402) 559-4651
E-mail: arubin_at_unmc.edu
WWW: http://www.unmc.edu/Eppley
Received on Thu Nov 21 2002 - 15:10:39 PST
