On Tue, Jul 08, 2014, Blake Mertz wrote:
>
> I'm compiled the parallel version of Amber14 against gcc-4.7.4 and
> openmpi-1.6.5-8 on a debian testing box (kernel 3.11), but when I ran the
> tests, two groups failed, pmemdTI and emil_pmemd_tip3p. Here's some of the
> difs:
For what it may be worth to others: these are both failures to compute the
components of DV/DL (in the section "DV/DL, AVERAGES").
For Blake: what was your setting for DO_PARALLEL? Are there any other tests
with similar diffs (i.e. a bunch of zeroes, as below)? Did you install
openmpi yourself, or use something like a package manager?
I don't think that anyone has reported failures in this test before. I can't
get it to fail on any of my machines.
....dac
>
> possible FAILURE: check campTI.out.dif
> /usr/local/src/amber14/test/pmemdTI/campTI
> 339,342c339,342
> < Etot = 0. EKtot = 0. EPtot = 4.3929
> < BOND = 36.5819 ANGLE = -2.5879 DIHED =
> -0.5307
> < 1-4 NB = -0.0292 1-4 EEL = -85.1242 VDWAALS =
> 2.5131
> < EELEC = 53.5699 EHBOND = 0. RESTRAINT = 0.
> ---
> > Etot = 0. EKtot = 0. EPtot = 0.
> > BOND = 0. ANGLE = 0. DIHED = 0.
> > 1-4 NB = 0. 1-4 EEL = 0. VDWAALS = 0.
> > EELEC = 0. EHBOND = 0. RESTRAINT = 0.
> 345d344
> < DV/DL = 4.3929
>
> and
>
> possible FAILURE: check mdout.emil.0.1.dif
> /usr/local/src/amber14/test/emil/emil_pmemd_tip3p
> 196,200c196,199
> < Etot = 0. EKtot = 0. EPtot = 35484.36
> < BOND = -248.06 ANGLE = -306.01 DIHED = -480.71
> < 1-4 NB = -192.19 1-4 EEL = 4543.51 VDWAALS = -6625.23
> < EELEC = 38793.07 EHBOND = 0. RESTRAINT = 0.
> < DV/DL = 35484.36
> ---
> > Etot = 0. EKtot = 0. EPtot = 0.
> > BOND = 0. ANGLE = 0. DIHED = 0.
> > 1-4 NB = 0. 1-4 EEL = 0. VDWAALS = 0.
> > EELEC = 0. EHBOND = 0. RESTRAINT = 0.
>
> I didn't find anything specific to these tests on the mailing list, so I
> wasn't sure if how concerned I should be, since both of these tests deal
> with FE calculations, which I don't need at the present moment. Thanks for
> any help!
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jul 08 2014 - 12:00:02 PDT
