[gmx-users] Constraints and efficency

[Elton Carvalho]

On Tue, Jan 4, 2011 at 7:38 PM, Mark Abraham <Mark.Abraham at anu.edu.au> wrote:
> On 5/01/2011 5:53 AM, Elton Carvalho wrote:
>> [...] Does the larger timestep allowed by eliminating
>> high-frequency O-H vibrations outweigh the time needed by SETTLE?
>
> Most definitely. Typical GROMACS simulations report up to a few percent of
> the total simulation time is spent in dealing with constraints. If the use
> of constraints permitted a 2fs timestep rather than 0.5fs, then it's nearly
> a four-fold speed-up.
>

Thanks :)

So, the larger ts allowed would be indeed the main reason to apply
constraints...

>> Any other reason I failed to notice?
>
> I seem to recall that bond constraints are held to model the essential QM
> nature of interatomic bonds better than a Morse or harmonic potential
> (particularly for bonds to hydrogen). I don't recall a reference for that
> offhand, unfortunately. Perhaps the original SHAKE paper.
>

Ok, I'll take a look at those papers, thank you

>> The only reason I see to constrain movement other than that fraquency
>> shaving (and, of course, isolating some behaviour of interest, but
>> this is case-specific) would be to reduce the number of dimensions of
>> the phase space and save computatuon time, but, as implemented, this
>> doesn't seem possible with the code as is.
>
> Various coarse-graining approaches exist to take advantage of that kind of
> saving, but they also gain from the higher maximum time step.

Oh, yes, coarse-graining methods definitely reduce the complexity of
the system, but I was focusing on all-atom simulations in my message.

Again, thank you for your response,

Best regards,
-- 
Elton Carvalho
Tel.: +55 11 3091-6985/6922
Dept Física dos Materiais e Mecânica
Instituto de Física
Universidade de São Paulo
P.O. Box 66318 - 05314-970 São Paulo-SP, Brazil