While the subject has minimization in it, I have a quick question: Is it
possible to use more than 2 CPUs in Thermodynamic Integration
minimization? I could never use 4 CPUs when I wanted to do minimization
with icfe set to either 1 or 2. Everything works fine with 2 CPUs, but not
with 4 CPUs in the minimization process. I use mpich2 and amber9.
On Tue, 11 Nov 2008, Lachele Foley wrote:
> Yep... Even the mdout file agrees:
>
>
> Number of triangulated 3-point waters found: 623
> | Atom division among processors:
> | 0 797 1592 2387 3182
>
> Sum of charges from parm topology file = 0.00000000
> Forcing neutrality...
> | Running AMBER/MPI version on 4 nodes
>
>
> I'll try running it on 2 nodes now and see if the output looks better.
>
>
> :-) Lachele
> --
> B. Lachele Foley, PhD '92,'02
> Assistant Research Scientist
> Complex Carbohydrate Research Center, UGA
> 706-542-0263
> lfoley.ccrc.uga.edu
>
>
> ----- Original Message -----
> From: Ilyas Yildirim
> [mailto:yildirim.pas.rochester.edu]
> To: amber-developers.scripps.edu
> Sent:
> Tue, 11 Nov 2008 13:41:50 -0500
> Subject: Re: amber-developers: Extra Points
> Calculation
>
>
> > Can you do minimization using 4 CPUs? I could only use 2 CPUs in
> > minimization processes.
> >
> > On Tue, 11 Nov 2008, Lachele Foley wrote:
> >
> > > PS to the previous... the minimization was run on four processors...
> > Sorry for any confusion.
> > >
> > >
> > > :-) Lachele
> > > --
> > > B. Lachele Foley, PhD '92,'02
> > > Assistant Research Scientist
> > > Complex Carbohydrate Research Center, UGA
> > > 706-542-0263
> > > lfoley.ccrc.uga.edu
> > >
> > >
> > > ----- Original Message -----
> > > From: Lachele Foley
> > > [mailto:lfoley.ccrc.uga.edu]
> > > To: amber-developers.scripps.edu
> > > Sent: Tue, 11
> > > Nov 2008 13:27:09 -0500
> > > Subject: Re: amber-developers: Extra Points
> > > Calculation
> > >
> > >
> > > > > We thought it does, but Lachele discovered more problems yesterday
> > > > > evening; he was going to take a look today.
> > > >
> > > > I kept having jobs fail... I could only get them to behave on 2
> > processors.
> > > > I hadn't even looked, graphically, at the output. Silly me.
> > > >
> > > > The minimization isn't doing quite what I expect. Load the attached
> > > > topology and restart into vmd or something and you will find, if you
> > twirl
> > > > the system the right way, a lovely rift between two halves of the
> > solvent,
> > > > with the sucrose molecule spanning the gap, more or less. The starting
> > > > coordinates look like a standard array of water molecules dropped around
> > a
> > > > solvent, so this is something the minimization is doing. Also attached
> > is
> > > > my MIN.in file in case there is some EP-relevant keyword I didn't set
> > > > properly. If I had to make a blind guess, I'd say the two processors
> > aren't
> > > > talking to each other in quite the way they should -- like half the
> > solvent
> > > > went to one, and half went to the other, and neither knows exactly where
> > the
> > > > other is. Or something.
> > > >
> > > > And, for what it's worth, The CP running on two processors seems to be
> > > > slowly correcting the problem. It's over 100,000 steps and the
> > coordinates
> > > > look pretty good now. At first, it was what you'd expect. I can send
> > some
> > > > or all of that output if anyone wants to see it.
> > > >
> > > > :-) Lachele
> > > > --
> > > > B. Lachele Foley, PhD '92,'02
> > > > Assistant Research Scientist
> > > > Complex Carbohydrate Research Center, UGA
> > > > 706-542-0263
> > > > lfoley.ccrc.uga.edu
> > > >
> > > >
> > > > ----- Original Message -----
> > > > From: Volodymyr Babin
> > > > [mailto:vbabin.ncsu.edu]
> > > > To: amber-developers.scripps.edu
> > > > Sent: Tue, 11 Nov
> > > > 2008 07:20:07 -0500
> > > > Subject: Re: amber-developers: Extra Points
> > > > Calculation
> > > >
> > > >
> > > > > > Does it solve Lachele's problem (the sucrose_EP test case)?
> > > > >
> > > > > We thought it does, but Lachele discovered more problems yesterday
> > > > > evening; he was going to take a look today.
> > > > >
> > > > > > As far as I can see, that example does not include any "LP" atoms,
> > > > > > and I don't think we should be expanding to that...my recollection
> > > > > > was that "LP" atoms (still found in
> > > > > > some old parameter/topology files) were *not* treated as extra
> > points,
> > > > but
> > > > > > as real atoms.
> > > > >
> > > > > I see. Then probably the 'LP ' part is unneeded indeed. I am not a
> > > > > right person to judge here.
> > > > >
> > > > > > I tried to expand the size of max14 (not as much as you did), and
> > was
> > > > > > still getting segfaults on Lachele's example.
> > > > >
> > > > > This is because of too short nb14_list and i14 (their sizes are
> > > > proportional
> > > > > to the max14, but max14 is not used in the corresponding allocate()).
> > > > >
> > > > > Volodymyr
> > > > >
> > > > >
> > >
> > >
> >
> > --
> > Ilyas Yildirim, Ph.D.
> > ---------------------------------------------------------------
> > = Hutchison Hall B#10 - Department of Chemistry =
> > = - University of Rochester =
> > = 585-275-6766 (office) - =
> > = http://www.pas.rochester.edu/~yildirim/ =
> > ---------------------------------------------------------------
> >
> >
>
>
--
Ilyas Yildirim, Ph.D.
---------------------------------------------------------------
= Hutchison Hall B#10 - Department of Chemistry =
= - University of Rochester =
= 585-275-6766 (office) - =
= http://www.pas.rochester.edu/~yildirim/ =
---------------------------------------------------------------
Received on Fri Dec 05 2008 - 14:29:05 PST
