[gmx-users] Simulation runs on iMac but explodes on cluster
Mark Abraham
Mark.Abraham at anu.edu.au
Wed Jul 13 11:31:45 CEST 2011
On 13/07/2011 7:01 PM, Luke Goodsell wrote:
> Hi Mark,
>
> Perhaps your simulation is intrinsically unstable, and you haven't
> gotten unlucky enough yet on the iMac. Check out
> http://www.gromacs.org/Documentation/Terminology/Blowing_Up
>
> The simulation on the iMac reaches a stable equilibrium.
OK, but that can be just lucky - how lucky depends on your preparation
protocol, which you haven't told us about. You are not actually doing
the same calculation on both machines - the processes are chaotic even
if everything seems equivalent.
> On the cluster, it appears stable until step 14555 where I get LINCS
> warnings such as
Those warnings look pretty localized to a region of your structure.
Firstly, some kind of linking or GROMACS failure would be likely to be
manifest all over the place - and cause problems even if you did a run
with different velocity generation. Secondly, have you looked at the
resulting PDB files? You may wish to run the simulation to step 14500,
and then restart with nstxout=1 so you can watch it explode. That may
highlight a bad contact suddenly being able to relax.
Mark
>
> Step 14555, time 29.11 (ps) LINCS WARNING
>
> relative constraint deviation after LINCS:
>
> rms 0.078151, max 2.416183 (between atoms 1720 and 1721)
>
> bonds that rotated more than 30 degrees:
>
> atom 1 atom 2 angle previous, current, constraint length
>
> 1720 1721 151.3 0.1080 0.3689 0.1080
>
> 1707 1708 152.8 0.1090 0.3674 0.1090
>
> Wrote pdb files with previous and current coordinates
>
>
> Step 14555, time 29.11 (ps) LINCS WARNING
>
> relative constraint deviation after LINCS:
>
> rms 0.078245, max 2.420384 (between atoms 1720 and 1721)
>
> bonds that rotated more than 30 degrees:
>
> atom 1 atom 2 angle previous, current, constraint length
>
> 1720 1721 151.3 0.1080 0.3694 0.1080
>
> 1707 1708 152.8 0.1090 0.3676 0.1090
>
>
> Back Off! I just backed up step14555b.pdb to ./#step14555b.pdb.1#
>
>
> Back Off! I just backed up step14555c.pdb to ./#step14555c.pdb.1#
>
> Wrote pdb files with previous and current coordinates
>
>
> Step 14556, time 29.112 (ps) LINCS WARNING
>
> relative constraint deviation after LINCS:
>
> rms 133.212808, max 5757.494141 (between atoms 1720 and 1721)
>
> bonds that rotated more than 30 degrees:
>
> atom 1 atom 2 angle previous, current, constraint length
>
> 1722 1724 36.7 0.1336 0.1716 0.1335
>
> 1722 1723 35.5 0.1230 0.1553 0.1229
>
> 1720 1721 90.0 0.3694 621.9174 0.1080
>
> 1718 1720 144.9 0.1407 0.5335 0.1400
>
> 1718 1719 105.6 0.1085 1.0804 0.1080
>
> 1716 1717 90.6 0.0960 24.8478 0.0960
>
> 1715 1718 136.5 0.1414 0.4281 0.1409
>
> 1715 1716 90.6 0.1366 37.5311 0.1364
>
> 1713 1715 111.6 0.1412 1.0479 0.1409
>
> 1713 1714 120.7 0.1082 0.3603 0.1080
>
> 1711 1713 118.7 0.1402 0.3129 0.1400
>
> 1711 1712 53.9 0.1082 0.1451 0.1080
>
> 1710 1720 101.0 0.1407 1.7692 0.1400
>
> 1710 1711 57.8 0.1404 0.2418 0.1400
>
> 1707 1710 112.0 0.1520 0.5543 0.1510
>
> 1707 1709 107.4 0.1098 0.5075 0.1090
>
> 1707 1708 102.7 0.3676 1.4700 0.1090
>
> 1705 1722 75.1 0.1525 0.2535 0.1522
>
> 1705 1707 106.7 0.1533 0.6142 0.1526
>
> 1705 1706 56.4 0.1093 0.1136 0.1090
>
> 1703 1705 89.0 0.1452 0.2683 0.1449
>
> 1703 1704 42.3 0.1010 0.1384 0.1010
>
> 1701 1703 38.9 0.1336 0.1768 0.1335
>
> 1673 1674 35.7 0.1229 0.1230 0.1229
>
> 1670 1673 31.2 0.1522 0.1523 0.1522
>
> Wrote pdb files with previous and current coordinates
>
>
> Step 14556, time 29.112 (ps) LINCS WARNING
>
> relative constraint deviation after LINCS:
>
> rms 133.228686, max 5758.184082 (between atoms 1720 and 1721)
>
> bonds that rotated more than 30 degrees:
>
> atom 1 atom 2 angle previous, current, constraint length
>
> 1722 1724 47.6 0.1336 0.1995 0.1335
>
> 1722 1723 45.8 0.1230 0.1781 0.1229
>
> 1720 1721 90.0 0.3694 621.9919 0.1080
>
> 1718 1720 141.7 0.1407 0.5243 0.1400
>
> 1718 1719 106.2 0.1085 1.1444 0.1080
>
> 1716 1717 90.7 0.0960 24.8269 0.0960
>
> 1715 1718 144.6 0.1414 0.5103 0.1409
>
> 1715 1716 90.6 0.1366 37.5424 0.1364
>
> 1713 1715 112.9 0.1412 1.0642 0.1409
>
> 1713 1714 120.9 0.1082 0.3765 0.1080
>
> 1711 1713 124.0 0.1402 0.2954 0.1400
>
> 1711 1712 52.5 0.1082 0.1399 0.1080
>
> 1710 1720 103.1 0.1407 1.8233 0.1400
>
> 1710 1711 75.5 0.1404 0.2584 0.1400
>
> 1707 1710 117.5 0.1520 0.6407 0.1510
>
> 1707 1709 109.9 0.1098 0.6153 0.1090
>
> 1707 1708 102.2 0.3676 1.5528 0.1090
>
> 1705 1722 92.4 0.1525 0.2917 0.1522
>
> 1705 1707 107.2 0.1533 0.7484 0.1526
>
> 1705 1706 57.0 0.1093 0.0958 0.1090
>
> 1703 1705 94.4 0.1452 0.3185 0.1449
>
> 1703 1704 49.7 0.1010 0.1528 0.1010
>
> 1701 1703 46.6 0.1336 0.1937 0.1335
>
> 1673 1674 35.7 0.1229 0.1230 0.1229
>
> 1670 1673 31.2 0.1522 0.1523 0.1522
>
>
> Back Off! I just backed up step14556b.pdb to ./#step14556b.pdb.1#
>
>
> Back Off! I just backed up step14556c.pdb to ./#step14556c.pdb.1#
>
> Wrote pdb files with previous and current coordinates
>
>
> The system is 'blowing up', no (rms ~575.8nm between 1720 and 1721)?
>
> It appears to me that there is some significant difference in the
> operation of gromacs in the two environments.
>
> Thanks,
> Luke
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20110713/d03076bc/attachment.html>
More information about the gromacs.org_gmx-users
mailing list