Hi Yao,
> is there any way I can continue a crashed run and a flag for sander to
> estimate what time the job will be done?
To continue a crashed run in &cntrl set
irest=1, ntx=5
Then give the restart file as the input coordinates with -c.
$AMBERHOME/bin/sander -o -i mdin -o mdout -p prmtop -c inpcrd -x mdcrd -inf
mdinfo -r restart
<crash>
$AMBERHOME/bin/sander -O -i mdin.1 -o mdout.1 -c restart -x mdcrd.1 -inf
mdinfo.1 -r restart.1
To see how long a job is estimated to take let it run for about 3 or 4
minutes and then cat the file specified by the -inf command line option.
(default name is mdinfo).
E.g.
NSTEP = 49924500 TIME(PS) = 100899.010 TEMP(K) = 293.55 PRESS =
0.0
Etot = -296.0964 EKtot = 222.2512 EPtot =
-518.3476
BOND = 46.4742 ANGLE = 156.2150 DIHED =
148.8680
1-4 NB = 67.3502 1-4 EEL = 901.4083 VDWAALS =
-101.4540
EELEC = -1388.9326 EGB = -348.2766 RESTRAINT =
0.0000
----------------------------------------------------------------------------
--
| Current Timing Info
| -------------------
| Total steps : 50000000 | Completed : 49924500 | Remaining : 75500
|
| Average timings for last 79250 steps:
| Elapsed(s) = 59.84 Per Step(ms) = 0.76
| ns/day = 228.86 seconds/ns = 377.53
|
| Average timings for all steps:
| Elapsed(s) = 37739.59 Per Step(ms) = 0.76
| ns/day = 228.59 seconds/ns = 377.97
|
|
| Estimated time remaining: 57.1 seconds.
----------------------------------------------------------------------------
--
All the best
Ross
/\
\/
|\oss Walker
---------------------------------------------------------
| Assistant Research Professor |
| San Diego Supercomputer Center |
| Adjunct Assistant Professor |
| Dept. of Chemistry and Biochemistry |
| University of California San Diego |
| NVIDIA Fellow |
| http://www.rosswalker.co.uk | http://www.wmd-lab.org/ |
| Tel: +1 858 822 0854 | EMail:- ross.rosswalker.co.uk |
---------------------------------------------------------
Note: Electronic Mail is not secure, has no guarantee of delivery, may not
be read every day, and should not be used for urgent or sensitive issues.
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Nov 04 2011 - 23:00:07 PDT