[gmx-users] g_msd bug

[Justin A. Lemkul]

Dommert Florian wrote:
> Hello,
> 
> I can also confirm this behaviour. Furthermore if I use an index group,
> that just contains 1 molecule and compare the results from an analysis
> with the flag -mol and without, then I obtain different results which
> should not be the case.
> 

It sounds like it's time to file one or more redmine issues for g_msd.  Memory 
usage might be separate from the inconsistency in the results with or without -mol.

-Justin

> /Flo
> 
> 
> On Thu, 2011-07-07 at 09:28 +0200, Ivan Gladich wrote: 
>> Dear all,
>>   I found the same problem that Sławomir  pointed out 10 days ago about 
>> the memory usage in the computation of g_msd.
>>
>>
>> I have 100 ns simulation for 2880 water molecules. The trajectory is 
>> savede every 1 ps: this means that I have 100000 frame
>>
>> I think that there is a strange memory problem only when the option -mol 
>> is used
>> Indeed, if I use
>>
>> g_msd_d -f ../t230.xtc -s ../t230.tpr -n index.ndx -o msdt230.xvg -b 
>> 4000 -e 100000 -beginfit 5000 -endfit 25000 -trestart 20 -type x
>>
>> the g_msd does not have any problem
>>
>> by the way using the option -mol (as Sławomir did in the last e-mails)
>>
>> g_msd_d -f ../t230.xtc -s ../t230.tpr -n index.ndx -o msdt230.xvg -b 
>> 4000 -e 100000 -beginfit 5000 -endfit 25000 -trestart 20 -type x -mol
>>
>> g_msd crashes and memory information (see below) points out that this 
>> crash is due by the memory overflow.
>>
>> Thanks a lot
>> Ivan
>>
>> Tasks: 478 total,   5 running, 473 sleeping,   0 stopped,   0 zombie
>> Cpu(s):  0.0%us,  0.5%sy, 12.7%ni, 77.7%id,  9.1%wa,  0.0%hi,  0.0%si,  
>> 0.0%st
>> Mem:  65459256k total, 65382420k used,    76836k free,        0k buffers
>> Swap: 32009508k total,  5360348k used, 26649160k free,     6312k cached
>>
>>
>>
>> On 06/27/11 15:40, Tsjerk Wassenaar wrote:
>>> Hi Slawomir,
>>>
>>> That's quite a usage of memory! Can you provide more information? Like
>>> the number of frames in the trajectory, the command line you used, and
>>> the system you ran on?
>>>
>>> Cheers,
>>>
>>> Tsjerk
>>>
>>> 2011/6/27 Sławomir Stachura<stachura.slawomir at gmail.com>:
>>>> Hi GMX Users,
>>>> I am writting this email, beacause I think the g_msd program in Gromacs 4.5.4 bears a problem. I was calculating the MSD od center of mass of POPC in membrane (system contains 274 POPC lipid molecules in all-atom force field) from 50 ns trajectory and it seems to consume great amount of memory. With  time of calculations the memory reserves are gradually devoured to the extent, in my case,  of over 600 GB (than my administrator of cluster killed the process). It seems that it does not release memory and it's pilling results up with steps  in memory. Have you heard of such case?
>>>> Best wishes,
>>>>    Slawomir--
>>>> gmx-users mailing list    gmx-users at gromacs.org
>>>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>>>> Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/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/Support/Mailing_Lists
>>>>
>>>
>>
>> -- 
>> ------
>> Ivan Gladich, Ph.D.
>> Postdoctoral Fellow
>> Academy of Sciences of the Czech Republic
>> Institute of Organic Chemistry and Biochemistry AS CR, v.v.i.
>> Flemingovo nám. 2.
>> 166 10 Praha 6
>> Czech Republic
>>
>> Tel: +420775504164
>> e-mail: ivan.gladich at uochb.cas.cz
>> web page:http://www.molecular.cz/~gladich/
>> -----
>>
> 
> 

-- 
========================================

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

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