[gmx-users] error in a parallel job

[Lanyuan Lu]

A micelle in CO2. The only particular feature I can think about is that
I used dummy atoms to produce the linear rigid model of CO2. Can
this be a reason? I guess when the gromacs devided the box into 2,
it broke some bonds. maybe that's the reason I got the error for
SHAKE. The input file is like:
[ moleculetype ]
; Name   nrexcl
CO2      3
[ atoms ]
;   nr    type   resnr  residu    atom    cgnr        charge          mass
     1      CO       1     CO2        CO     1          0.6512
     2      OO       1     CO2        OO     1        - 0.3256
     3      OO       1     CO2        OO     1        - 0.3256
     4      RM       1     CO2        RM     1          0.0
     5      RM       1     CO2        RM     1          0.0
[dummies2]
1    4    5    1    0.5
2    4    5    1    -0.0862
3    4    5    1    1.0862
[ constraints ]
4    5    1    0.196
[exclusions]
1 2 3 4 5
2 1 3 4 5
3 1 2 4 5
I calculated the same micelle with single point CO2 model without problem.
But for this 3-site CO2, It breaks often.
Thanks.
Lanyuan Lu

----- Original Message ----- 
From: "David" <spoel at xray.bmc.uu.se>
To: <gmx-users at gromacs.org>
Sent: Friday, November 14, 2003 2:30 PM
Subject: Re: [gmx-users] error in a parallel job


> On Fri, 2003-11-14 at 20:24, lulanyuan wrote:
> > In short. For a certain system A I can calculate it using 1 processor but can not using 2. For other systems I can use any
> > number of processors (1,2,4 etc) as I like. So what's the problem about system A? 
> > The error message I got was shake didn't converge. I din't change any condition other than number of processors.
> > Thanks.
> > Lanyuan
> 
> either your starting conformation or something peculiar about system A.
> so what is system A?
> 
> > >Hello, everyone. I got a strange problem when I simulated
> > >a system using two processors. The job is ok with one processor.
> > >But when I try to calculate it using two, I got it broken in
> > >the middle. In the log file I found a lot of error message saying shake was not converged in 1000 steps.
> > >For other simulations, I got no problem for parallel jobs.
> > >Can anyone give me a clue?
> > >not without more information.
> > 
> > >Thank you very much.
> > >Lanyuan Lu
> > ______________________________________
> > ===================================================================
> > °®¹úÕßÊýÂëÏà»ú°éÂÂÍõ--´øÉÏ555¸ö½º¾íÈ¥ÂÃÓÎ (http://ad4.sina.com.cn/shc/zhuiyu_hprefresh.html)
> > _______________________________________________
> > gmx-users mailing list
> > gmx-users at gromacs.org
> > http://www.gromacs.org/mailman/listinfo/gmx-users
> > Please don't post (un)subscribe requests to the list. Use the 
> > www interface or send it to gmx-users-request at gromacs.org.
> -- 
> David.
> ________________________________________________________________________
> David van der Spoel, PhD, Assist. Prof., Molecular Biophysics group,
> Dept. of Cell and Molecular Biology, Uppsala University.
> Husargatan 3, Box 596,  75124 Uppsala, Sweden
> phone: 46 18 471 4205 fax: 46 18 511 755
> spoel at xray.bmc.uu.se spoel at gromacs.org   http://xray.bmc.uu.se/~spoel
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> 
> _______________________________________________
> gmx-users mailing list
> gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the list. Use the 
> www interface or send it to gmx-users-request at gromacs.org.
> 
> 
>