Re: [AMBER-Developers] infinite ptraj.MPI, was: First AmberTools release candidate

From: Lachele Foley <lfoley.ccrc.uga.edu>
Date: Tue, 16 Mar 2010 20:05:15 -0400

I might just owe whoever wrote the cause of this... er... misfeature (*) a beer.

It is related to the filesystem... Yay for every little piece of information...

> I am seeing a slowdown for ptraj.MPI compared to ptraj, not sure why though.

I tried it direct, one processor, from the command line, e.g.:
$ $AMBERHOME/bin/ptraj.MPI compound.prmtop ptraj.in >& ptraj.out

1. In the usual $AMBERHOME space ...this is on Filesystem 1, type lustre: gave up after 2-3 minutes.

2. cp -r ptraj_comprehensive to some place in /users on Filesystem 2. type lustre: 3 minutes and give up

3. cp -r to some place in /tmp, the root filesystem, type ext3, and:

/usr/bin/time $AMBERHOME/bin/ptraj.MPI compound.prmtop ptraj.in >& ptraj.out
1.84user 0.04system 0:01.89elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k

...been meaning to do this test for a few days, actually.

Is there any chance you could hazard a guess what could cause this?

(*) Term "misfeature" by Volodymyr. Original illustration by unknown artist, enhanced by Matt Tessier:
http://lh5.ggpht.com/_7CIiCZypTWo/S6AbnyxtbBI/AAAAAAAAFqQ/y_0FAUZdFxw/s720/Bug%20v%20Feature%20v%20Enhanced%20v%20Misfeature.png

:-) 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: Daniel Roe
[mailto:daniel.r.roe.gmail.com]
To: AMBER Developers Mailing List
[mailto:amber-developers.ambermd.org]
Sent: Tue, 16 Mar 2010 19:18:54
-0400
Subject: Re: [AMBER-Developers] First AmberTools release candidate
> On Tue, Mar 16, 2010 at 4:26 PM, Lachele Foley <lfoley.ccrc.uga.edu> wrote:
> 
> > How long should the parallel ptraj test take?   I'm at 30+ minutes on 4
> > cpu's (45+ by the time I finished this...).  I'm running on the "bugfix
> 126"
> > machine (head node for simplicity).  Top says 100% CPU use and 0% memory. 
> I
> > just killed it and restarted it, and same behavior now, for about 2
> minutes.
> >
> 
> I am seeing a slowdown for ptraj.MPI compared to ptraj, not sure why though.
> However, the test cases complete fine for me with no failures (it just takes
> longer for some reason). I am looking into the slowdown now.
> 
> -Dan
> _______________________________________________
> AMBER-Developers mailing list
> AMBER-Developers.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber-developers
> 
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers

Received on Tue Mar 16 2010 - 17:30:02 PDT