[gmx-users] grompp very slow generating .tpr when excluded bonded neighbours is large

[Dallas Warren]

Thanks for the reply Justin.

I am just going to use the largest exclusion bond distance I can, then
ignore the RDF of those beyond that distance.

Seems curious to me (not actually understanding what grommp is
generating) that the list is so large.  These are linear molecules, 38
atoms, 60 molecules in total.
Catch ya,

Dr. Dallas Warren
Drug Delivery, Disposition and Dynamics
Monash Institute of Pharmaceutical Sciences, Monash University
381 Royal Parade, Parkville VIC 3052
dallas.warren at monash.edu
---------------------------------
When the only tool you own is a hammer, every problem begins to resemble a nail.


On 28 September 2017 at 21:49, Justin Lemkul <jalemkul at vt.edu> wrote:
>
>
> On 9/28/17 12:25 AM, Dallas Warren wrote:
>>
>> Justin,
>>
>> No, I have separate index groups for each of the atom types.
>>
>> Wouldn't that still have the pair of atoms that are located on the
>> same molecule, contributing to the RDF?
>>
>> Molecule is like this:
>> C1 -------C38
>>
>> Index file would be something like this:
>> [ C1 ]
>> 1 39 77 .....
>>
>> [ C38 ]
>> 38 76 114 ....
>>
>> So doesn't it still include these atom pairs to generate the RDF? Atom
>> pair list below.  From the RDFs that are generating using this method,
>> it appears it does as there is a massive peak around the intra-atomic
>> distance.
>>
>> 1-38, 1-76, 1-114 ..... 39-38, 39-76, 39-114 ....
>>
>> First and fifth pair in this list are atoms that are contained within
>> the same molecule.
>>
>> Wouldn't be surprised if I am missing something simple here .....
>
>
> No, I misunderstood what you were actually after. What you are asking for
> does require a huge intramolecular nrexcl and you are indeed seemingly
> running out of memory trying to create such a huge exclusion list. Sorry to
> say I don't have much more to suggest here.
>
>
> -Justin
>
>> Catch ya,
>>
>> Dr. Dallas Warren
>> Drug Delivery, Disposition and Dynamics
>> Monash Institute of Pharmaceutical Sciences, Monash University
>> 381 Royal Parade, Parkville VIC 3052
>> dallas.warren at monash.edu
>> ---------------------------------
>> When the only tool you own is a hammer, every problem begins to resemble a
>> nail.
>>
>>
>> On 28 September 2017 at 10:27, Justin Lemkul <jalemkul at vt.edu> wrote:
>>>
>>>
>>> On 9/27/17 7:54 PM, Dallas Warren wrote:
>>>>
>>>> Left the 37 one running, and it ended up being "Killed". Was
>>>> distracted, so didn't really see how long it ran etc, but suspect it
>>>> was due to amount of memory being used up.
>>>>
>>>> I am going about this the right way? i.e. want the RDF between atoms
>>>> at either end of the molecule (38 atoms long), and don't want the
>>>> atoms within the same molecule included in the calculation then need
>>>> to set nrexcl to 37?
>>>
>>>
>>> I'm assuming you're trying to get the RDF simply by selecting the whole
>>> molecule? Why not just use index groups with the starting and ending
>>> atoms
>>> in separate groups and compute the RDF of one around the other?
>>>
>>> -Justin
>>>
>>>
>>>> Catch ya,
>>>>
>>>> Dr. Dallas Warren
>>>> Drug Delivery, Disposition and Dynamics
>>>> Monash Institute of Pharmaceutical Sciences, Monash University
>>>> 381 Royal Parade, Parkville VIC 3052
>>>> dallas.warren at monash.edu
>>>> ---------------------------------
>>>> When the only tool you own is a hammer, every problem begins to resemble
>>>> a
>>>> nail.
>>>>
>>>>
>>>> On 28 September 2017 at 09:02, Dallas Warren <dallas.warren at monash.edu>
>>>> wrote:
>>>>>
>>>>> Just want to check that this should be taking as long as it is
>>>>> (expected behaviour etc), for grompp to generate the tpr when nrexcl
>>>>> is large.
>>>>>
>>>>> I have a long straight chain molecule with 38 atoms that are
>>>>> generating RDFs for, so need to change the [ moleculetype ] nrexcl to
>>>>> exclude atoms within the same molecule from the generated RDFs.
>>>>> Initially set nrexcl to 37, ran grompp to get the tpr, and it was
>>>>> taking a loooong time and consuming significant amounts of memory.
>>>>> Killed it and ran a few tests with increasing values and it is pretty
>>>>> much instantaneous up until about 25.  Each increase after that
>>>>> increases the time it takes in the " excluding ## bonded neighbours
>>>>> molecule type '####' " step.
>>>>>
>>>>> Catch ya,
>>>>>
>>>>> Dr. Dallas Warren
>>>>> Drug Delivery, Disposition and Dynamics
>>>>> Monash Institute of Pharmaceutical Sciences, Monash University
>>>>> 381 Royal Parade, Parkville VIC 3052
>>>>> dallas.warren at monash.edu
>>>>> ---------------------------------
>>>>> When the only tool you own is a hammer, every problem begins to
>>>>> resemble
>>>>> a nail.
>>>
>>>
>>> --
>>> ==================================================
>>>
>>> Justin A. Lemkul, Ph.D.
>>> Assistant Professor
>>> Virginia Tech Department of Biochemistry
>>>
>>> 303 Engel Hall
>>> 340 West Campus Dr.
>>> Blacksburg, VA 24061
>>>
>>> jalemkul at vt.edu | (540) 231-3129
>>> http://www.biochem.vt.edu/people/faculty/JustinLemkul.html
>>>
>>> ==================================================
>>> --
>>> Gromacs Users mailing list
>>>
>>> * Please search the archive at
>>> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
>>> posting!
>>>
>>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>>
>>> * For (un)subscribe requests visit
>>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>>> send a
>>> mail to gmx-users-request at gromacs.org.
>
>
> --
> ==================================================
>
> Justin A. Lemkul, Ph.D.
> Assistant Professor
> Virginia Tech Department of Biochemistry
>
> 303 Engel Hall
> 340 West Campus Dr.
> Blacksburg, VA 24061
>
> jalemkul at vt.edu | (540) 231-3129
> http://www.biochem.vt.edu/people/faculty/JustinLemkul.html
>
> ==================================================
>
> --
> Gromacs Users mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a
> mail to gmx-users-request at gromacs.org.