[gmx-users] lincs warning

[XAvier Periole]

On Jun 3, 2009, at 10:48 AM, Erik Marklund wrote:

> XAvier Periole skrev:
>>
>> On Jun 3, 2009, at 10:01 AM, Ran Friedman wrote:
>>
>>> Hi XAvier,
>>> Do you use virtual sites? I've seen this when I used virtual sites,
>> No virtual sites.
>>> large time steps and a system that probably wasn't equilibrated  
>>> enough.
>> time step 2 fs and a system totally equilibrated, nothing there.
>>
>> I would have considered this if the warning were reproduced when
>> restarting the simulation.
>>
>> It seems more that it is a loss of accuracy due to communication
>> between CPUs ?
>>
> If you use dynamic load balancing then you are sacrificing binary  
> reproducibility. Because of this, and the chaotic nature of MD, the  
> error will likely not occur at the same point in your continuation.  
> Still, the source of the error persists and is what demands  
> attention I'd say.
Well starting from a cpt file, I would expect reproducibility at the  
difference that the
decomposition over CPUs might be refreshed.

It might be but I am still favoring a communication issue.

I am now testing the effect of using -rdd and -rcon on the LINCS  
warnings.
>> XAvier.
>>
>>> Ran.
>>>
>>> XAvier Periole wrote:
>>>>
>>>> Dears,
>>>>
>>>> I am experiencing some problems running a few proteins in water
>>>> (GROMOS43a1/SPC)
>>>> with gmx-4.0.4 using 32 CPUs.
>>>>
>>>> After some ns lincs warnings appear and eventually the simulation
>>>> crashes dues
>>>> to too many lincs warnings.
>>>>
>>>> When restarting the simulation feeding "-cpi md.cpt" to mdrun the
>>>> simulation restarts
>>>> fine and no lincs warning are reported on the same portion of the
>>>> simulation. The
>>>> simulation runs again for some ns and the same happens: new lincs
>>>> warning appear
>>>> and eventually the simulation crashes again.
>>>>
>>>> Any explanation for this? Any solution?
>>>>
>>>> Best,
>>>> XAvier.
>>>>
>>>> _______________________________________________
>>>> gmx-users mailing list    gmx-users at gromacs.org
>>>> http://www.gromacs.org/mailman/listinfo/gmx-users
>>>> Please search the archive at http://www.gromacs.org/search before
>>>> posting!
>>>> Please don't post (un)subscribe requests to the list. Use thewww
>>>> interface or send it to gmx-users-request at gromacs.org.
>>>> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>>>>
>>>
>>> _______________________________________________
>>> gmx-users mailing list    gmx-users at gromacs.org
>>> http://www.gromacs.org/mailman/listinfo/gmx-users
>>> Please search the archive at http://www.gromacs.org/search before  
>>> posting!
>>> Please don't post (un)subscribe requests to the list. Use the
>>> www interface or send it to gmx-users-request at gromacs.org.
>>> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>>
>> _______________________________________________
>> gmx-users mailing list    gmx-users at gromacs.org
>> http://www.gromacs.org/mailman/listinfo/gmx-users
>> Please search the archive at http://www.gromacs.org/search before  
>> posting!
>> Please don't post (un)subscribe requests to the list. Use thewww  
>> interface or send it to gmx-users-request at gromacs.org.
>> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>
>
> -- 
> -----------------------------------------------
> Erik Marklund, PhD student
> Laboratory of Molecular Biophysics,
> Dept. of Cell and Molecular Biology, Uppsala University.
> Husargatan 3, Box 596,    75124 Uppsala, Sweden
> phone:    +46 18 471 4537        fax: +46 18 511 755
> erikm at xray.bmc.uu.se    http://xray.bmc.uu.se/molbiophys
>
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before  
> posting!
> Please don't post (un)subscribe requests to the list. Use the www  
> interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php