Hi Ross,
Thanks for your quick response and sorry for not specifying the executable I was using. I am running the simulation with pmemd.cuda. For your consideration, I pasted the initial part of the output file at the end of this email.
The Run.vrand test using pmemd.cuda seems to be passed, although with numerical differences (output attached).
This problem seems to happen only with pmemd.cuda. With pmemd the temperature is kept constant at 300 K, as happened in pmemd.cuda in AMBER12.
Just changing ntt=2 to ntt=3 keeps the temperature constant at 300 K, so I guess only Andersen thermostat (which is faster than Langevin in our hands) is failing here.
I wonder if other users experienced the same issues.
Thanks a lot,
Gonzalo
***********************************************************************************************************************************************
-------------------------------------------------------
Amber 14 SANDER 2014
-------------------------------------------------------
| PMEMD implementation of SANDER, Release 14
| Run on 10/23/2014 at 04:42:19
| Executable path: /usr/local/amber14/amber14//bin/pmemd.cuda
| Working directory: /nfs/home/gonzalo/sim56
| Hostname: seaquest2
[-O]verwriting output
File Assignments:
| MDIN: md2.in
| MDOUT: md2.out
| INPCRD: md.rst
| PARM: sim56.prmtop
| RESTRT: md2.rst
| REFC: md.rst
| MDVEL: mdvel
| MDEN: mden
| MDCRD: mdcrd_MD2
| MDINFO: mdinfo
| MDFRC: mdfrc
Here is the input file:
production
&cntrl
imin = 0,
irest = 1,
ntx = 7,
ntb = 1,
cut = 8.0,
ntr = 0,
ntc = 2,
ntf = 2,
tempi = 300.0,
temp0 = 300.0,
ntt = 2,
ioutfm = 1,
iwrap = 1,
nstlim = 150000000, dt = 0.002,
ntpr = 10000, ntwx = 10000, ntwr = -10000,
/
|--------------------- INFORMATION ----------------------
| GPU (CUDA) Version of PMEMD in use: NVIDIA GPU IN USE.
| Version 14.0
|
| 01/31/2014
|
| Implementation by:
| Ross C. Walker (SDSC)
| Scott Le Grand (nVIDIA)
|
| CAUTION: The CUDA code is currently experimental.
| You use it at your own risk. Be sure to
| check ALL results carefully.
|
| Precision model in use:
| [SPFP] - Mixed Single/Double/Fixed Point Precision.
| (Default)
|
|--------------------------------------------------------
|----------------- CITATION INFORMATION -----------------
|
| When publishing work that utilized the CUDA version
| of AMBER, please cite the following in addition to
| the regular AMBER citations:
|
| - Romelia Salomon-Ferrer; Andreas W. Goetz; Duncan
| Poole; Scott Le Grand; Ross C. Walker "Routine
| microsecond molecular dynamics simulations with
| AMBER - Part II: Particle Mesh Ewald", J. Chem.
| Theory Comput., 2013, 9 (9), pp3878-3888,
| DOI: 10.1021/ct400314y.
|
| - Andreas W. Goetz; Mark J. Williamson; Dong Xu;
| Duncan Poole; Scott Le Grand; Ross C. Walker
| "Routine microsecond molecular dynamics simulations
| with AMBER - Part I: Generalized Born", J. Chem.
| Theory Comput., 2012, 8 (5), pp1542-1555.
|
| - Scott Le Grand; Andreas W. Goetz; Ross C. Walker
| "SPFP: Speed without compromise - a mixed precision
| model for GPU accelerated molecular dynamics
| simulations.", Comp. Phys. Comm., 2013, 184
| pp374-380, DOI: 10.1016/j.cpc.2012.09.022
|
|--------------------------------------------------------
|------------------- GPU DEVICE INFO --------------------
|
| CUDA_VISIBLE_DEVICES: not set
| CUDA Capable Devices Detected: 2
| CUDA Device ID in use: 1
| CUDA Device Name: GeForce GTX 780
| CUDA Device Global Mem Size: 3071 MB
| CUDA Device Num Multiprocessors: 12
| CUDA Device Core Freq: 0.90 GHz
|
|--------------------------------------------------------
| Conditional Compilation Defines Used:
| PUBFFT
| BINTRAJ
| CUDA
| EMIL
| Largest sphere to fit in unit cell has radius = 34.952
| New format PARM file being parsed.
| Version = 1.000 Date = 10/15/14 Time = 13:47:27
| Note: 1-4 EEL scale factors are being read from the topology file.
| Note: 1-4 VDW scale factors are being read from the topology file.
| Duplicated 0 dihedrals
| Duplicated 0 dihedrals
--------------------------------------------------------------------------------
1. RESOURCE USE:
--------------------------------------------------------------------------------
getting new box info from bottom of inpcrd
NATOM = 48510 NTYPES = 19 NBONH = 44770 MBONA = 3818
NTHETH = 8484 MTHETA = 5174 NPHIH = 16487 MPHIA = 13426
NHPARM = 0 NPARM = 0 NNB = 95981 NRES = 14157
NBONA = 3818 NTHETA = 5174 NPHIA = 13426 NUMBND = 82
NUMANG = 172 NPTRA = 77 NATYP = 50 NPHB = 1
IFBOX = 2 NMXRS = 85 IFCAP = 0 NEXTRA = 0
NCOPY = 0
| Coordinate Index Table dimensions: 16 16 16
| Direct force subcell size = 5.3509 5.3509 5.3509
BOX TYPE: TRUNCATED OCTAHEDRON
--------------------------------------------------------------------------------
2. CONTROL DATA FOR THE RUN
--------------------------------------------------------------------------------
default_name
General flags:
imin = 0, nmropt = 0
Nature and format of input:
ntx = 7, irest = 1, ntrx = 1
Nature and format of output:
ntxo = 1, ntpr = 10000, ntrx = 1, ntwr = -10000
iwrap = 1, ntwx = 10000, ntwv = 0, ntwe = 0
ioutfm = 1, ntwprt = 0, idecomp = 0, rbornstat= 0
Potential function:
ntf = 2, ntb = 1, igb = 0, nsnb = 25
ipol = 0, gbsa = 0, iesp = 0
dielc = 1.00000, cut = 8.00000, intdiel = 1.00000
Frozen or restrained atoms:
ibelly = 0, ntr = 0
Molecular dynamics:
nstlim = 150000000, nscm = 1000, nrespa = 1
t = 0.00000, dt = 0.00200, vlimit = -1.00000
Anderson (strong collision) temperature regulation:
ig = 71277, vrand = 1000
temp0 = 300.00000, tempi = 300.00000
SHAKE:
ntc = 2, jfastw = 0
tol = 0.00001
| Intermolecular bonds treatment:
| no_intermolecular_bonds = 1
| Energy averages sample interval:
| ene_avg_sampling = 10000
Ewald parameters:
verbose = 0, ew_type = 0, nbflag = 1, use_pme = 1
vdwmeth = 1, eedmeth = 1, netfrc = 1
Box X = 85.615 Box Y = 85.615 Box Z = 85.615
Alpha = 109.471 Beta = 109.471 Gamma = 109.471
NFFT1 = 96 NFFT2 = 96 NFFT3 = 96
Cutoff= 8.000 Tol =0.100E-04
Ewald Coefficient = 0.34864
Interpolation order = 4
--------------------------------------------------------------------------------
3. ATOMIC COORDINATES AND VELOCITIES
--------------------------------------------------------------------------------
default_name
begin time read from input coords =202100.000 ps
Number of triangulated 3-point waters found: 13687
Sum of charges from parm topology file = -0.00000621
Forcing neutrality...
| Dynamic Memory, Types Used:
| Reals 1341211
| Integers 1839531
| Nonbonded Pairs Initial Allocation: 8102382
| GPU memory information (estimate):
| KB of GPU memory in use: 294369
| KB of CPU memory in use: 62817
--------------------------------------------------------------------------------
4. RESULTS
--------------------------------------------------------------------------------
---------------------------------------------------
APPROXIMATING switch and d/dx switch using CUBIC SPLINE INTERPOLATION
using 5000.0 points per unit in tabled values
TESTING RELATIVE ERROR over r ranging from 0.0 to cutoff
| CHECK switch(x): max rel err = 0.2738E-14 at 2.422500
| CHECK d/dx switch(x): max rel err = 0.8332E-11 at 2.782960
---------------------------------------------------
|---------------------------------------------------
| APPROXIMATING direct energy using CUBIC SPLINE INTERPOLATION
| with 50.0 points per unit in tabled values
| Relative Error Limit not exceeded for r .gt. 2.47
| APPROXIMATING direct force using CUBIC SPLINE INTERPOLATION
| with 50.0 points per unit in tabled values
| Relative Error Limit not exceeded for r .gt. 2.89
|---------------------------------------------------
Setting new random velocities at step 1000
check COM velocity, temp: 0.001271 0.00(Removed)
Setting new random velocities at step 2000
check COM velocity, temp: 0.002907 0.01(Removed)
Setting new random velocities at step 3000
check COM velocity, temp: 0.001948 0.01(Removed)
Setting new random velocities at step 4000
check COM velocity, temp: 0.001399 0.00(Removed)
Setting new random velocities at step 5000
check COM velocity, temp: 0.003220 0.02(Removed)
Setting new random velocities at step 6000
check COM velocity, temp: 0.003343 0.02(Removed)
Setting new random velocities at step 7000
check COM velocity, temp: 0.003447 0.02(Removed)
Setting new random velocities at step 8000
check COM velocity, temp: 0.001476 0.00(Removed)
Setting new random velocities at step 9000
check COM velocity, temp: 0.000991 0.00(Removed)
Setting new random velocities at step 10000
check COM velocity, temp: 0.001625 0.00(Removed)
wrapping first mol.: -57.07654 40.35921 69.90421
writing md2.rst_10000
wrapping first mol.: -57.07654 40.35921 69.90421
NSTEP = 10000 TIME(PS) = 202120.000 TEMP(K) = 0.91 PRESS = 0.0
Etot = -145953.8202 EKtot = 91.3135 EPtot = -146045.1337
BOND = 1381.5973 ANGLE = 3836.8153 DIHED = 5299.1263
1-4 NB = 1694.1473 1-4 EEL = 19271.0073 VDWAALS = 15393.3846
EELEC = -192921.2118 EHBOND = 0.0000 RESTRAINT = 0.0000
------------------------------------------------------------------------------
Setting new random velocities at step 11000
check COM velocity, temp: 0.001552 0.00(Removed)
Setting new random velocities at step 12000
check COM velocity, temp: 0.001903 0.01(Removed)
Setting new random velocities at step 13000
check COM velocity, temp: 0.002350 0.01(Removed)
Setting new random velocities at step 14000
check COM velocity, temp: 0.002269 0.01(Removed)
Setting new random velocities at step 15000
check COM velocity, temp: 0.000604 0.00(Removed)
Setting new random velocities at step 16000
check COM velocity, temp: 0.002395 0.01(Removed)
Setting new random velocities at step 17000
check COM velocity, temp: 0.002963 0.01(Removed)
Setting new random velocities at step 18000
check COM velocity, temp: 0.003513 0.02(Removed)
Setting new random velocities at step 19000
check COM velocity, temp: 0.002791 0.01(Removed)
Setting new random velocities at step 20000
check COM velocity, temp: 0.002330 0.01(Removed)
wrapping first mol.: -57.07654 40.35921 69.90421
writing md2.rst_20000
-----Mensaje original-----
From: Ross Walker
Sent: Saturday, October 25, 2014 6:44 PM
To: AMBER Mailing List
Subject: Re: [AMBER] Temperature not correctly maintained using AMBER14
Hi Gonzalo,
Is this sander, sander.MPI, pmemd, pmemd.MPI, pmemd.cuda or pmemd.cuda.MPI?
Do the test cases pass - in particular Run.vrand in 4096wat and
cuda/4096wat?
And is there a difference if you use sander vs pmemd vs pmemd.cuda?
All the best
Ross
On 10/25/14, 6:07 PM, "Gonzalo Jimenez" <gonzalojimenezoses.gmail.com>
wrote:
>Hi,
>
>I am experiencing weird issues with temperature control in AMBER14. This
>is my input file:
>
>production
>&cntrl
> imin = 0,
> irest = 1,
> ntx = 7,
> ntb = 1,
> cut = 8.0,
> ntr = 0,
> ntc = 2,
> ntf = 2,
> tempi = 300.0,
> temp0 = 300.0,
> ntt = 2,
> ioutfm = 1,
> iwrap = 1,
> nstlim = 500000000, dt = 0.002,
> ntpr = 10000, ntwx = 10000, ntwr = -10000,
>/
>
>
>
>It works nicely in AMBER12, and keeps temperature at 300+-2 K, but in
>AMBER14, the temperature is constant, but 40K above!:
>
>md8.out: NSTEP = 33630000 TIME(PS) = 2069360.003 TEMP(K) = 339.36
>PRESS = 0.0
>md8.out: NSTEP = 33640000 TIME(PS) = 2069380.003 TEMP(K) = 341.18
>PRESS = 0.0
>md8.out: NSTEP = 33650000 TIME(PS) = 2069400.003 TEMP(K) = 340.15
>PRESS = 0.0
>md8.out: NSTEP = 33660000 TIME(PS) = 2069420.003 TEMP(K) = 340.09
>PRESS = 0.0
>md8.out: NSTEP = 33670000 TIME(PS) = 2069440.003 TEMP(K) = 339.43
>PRESS = 0.0
>md8.out: NSTEP = 33680000 TIME(PS) = 2069460.003 TEMP(K) = 339.60
>PRESS = 0.0
>md8.out: NSTEP = 33690000 TIME(PS) = 2069480.003 TEMP(K) = 339.53
>PRESS = 0.0
>md8.out: NSTEP = 33700000 TIME(PS) = 2069500.003 TEMP(K) = 339.43
>PRESS = 0.0
>md8.out: NSTEP = 33710000 TIME(PS) = 2069520.003 TEMP(K) = 339.41
>PRESS = 0.0
>md8.out: NSTEP = 33720000 TIME(PS) = 2069540.003 TEMP(K) = 339.10
>PRESS = 0.0
>md8.out: NSTEP = 33730000 TIME(PS) = 2069560.003 TEMP(K) = 342.46
>PRESS = 0.0
>md8.out: NSTEP = 33740000 TIME(PS) = 2069580.003 TEMP(K) = 339.59
>PRESS = 0.0
>md8.out: NSTEP = 33750000 TIME(PS) = 2069600.003 TEMP(K) = 339.71
>PRESS = 0.0
>
>
>Why is that? Is it a bug or just a printing problem?
>
>Thanks a lot,
>Gonzalo
>_______________________________________________
>AMBER mailing list
>AMBER.ambermd.org
>http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sat Oct 25 2014 - 20:00:02 PDT
