[AMBER] "nstlim" is not reflected correctly in REMD with pmemd.MPI

From: kurisaki <kurisaki.ncube.human.nagoya-u.ac.jp>
Date: Tue, 21 Jul 2015 20:16:30 +0900

Dear Amber developers and users,

I'm trying REMD with Amber14's pmemd.MPI.
Although REMD simulation proceeds,
I encountered anomalous situation.

"nstlim" is not reflected correctly,
Instead, "anomalously huge number" was given for "Total steps" in mdinfo.

Here is one of input files:

=====================================================
REMD
 &cntrl
   irest=0, ntx=1,
   nstlim=500000, dt=0.002,
   irest=0, ntt=3, gamma_ln=1.0,
   temp0=360.0, ig=21016,
   ntc=2, ntf=2, nscm=1000,
   ntb=0, igb=5,
   cut=999.0, rgbmax=999.0, extdiel = 78.5,
   ntpr=500, ntwx=500, ntwr=100000,
   numexchg=50000,
/
======================================================

I used following command to submit job:
mpirun -np 24 pmemd.MPI -ng 12 -groupfile remd.groupfile < /dev/null


Shown below is mdinfo in the middle of simulation.
NSTLIM is over 500000 already.

======================================================

NSTEP = 746000 TIME(PS) = 1492.000 TEMP(K) = 551.07 PRESS = 0.0
 Etot = -395.2254 EKtot = 125.3866 EPtot = -520.6120
 BOND = 18.2740 ANGLE = 77.7230 DIHED = 32.1195
 1-4 NB = 17.8187 1-4 EEL = -1168.7411 VDWAALS = -7.6935
 EELEC = 590.2709 EGB = -80.3836 RESTRAINT = 0.0000
 TEMP0 = 540.0000 REPNUM = 8 EXCHANGE# = 2
 ------------------------------------------------------------------------------
| Current Timing Info
| -------------------
| Total steps : ********* | Completed : 746000 | Remaining : *********
|
| Average timings for last 96000 steps:
| Elapsed(s) = 59.93 Per Step(ms) = 0.62
| ns/day = 276.79 seconds/ns = 312.15
|
| Average timings for all steps:
| Elapsed(s) = 479.61 Per Step(ms) = 0.64
| ns/day = 268.78 seconds/ns = 321.46
|
|
| Estimated time remaining: -495397.8 seconds.
 ------------------------------------------------------------------------------

=========================================================

Maybe, this simulation takes 6 days in my platform.

Why does this kind of error appear?
Is this due to input file or Amber14 itself?

I'm most grateful if you give some advices to resolve this problem.

Best regards,

                                                    Ikuo KURISAKI



_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jul 21 2015 - 04:30:02 PDT
Custom Search