4. Quick Start Guide

This chapter describes how to perform a single-point calculation of amino acid (glycine) using a single process.

4.1. Creating calculation directories

ProteinDF writes a large number of large intermediate files with matrices/vectors on disk storage. Before starting actual calculations, create the following calculation directories:

  • fl_Input
  • fl_Table
  • fl_Work

.

Warning

The program may abnormally terminate if the intermediate files cannot be written properly. Before executing MPI parallel computation, make sure that the files can be properly written from all nodes.

Note

The user can execute ${PDF_HOME}/bin/pdfsetup to perform the operation above.

4.2. Creating input files

Create a text file with the following content, and save it under the name fl_Userinput:

>>>>MAIN
      step-control    = [create integral guess scf]
      cut-value       = 1.0e-10
      scf-start-guess = harris
      max-iteration   = 100
      method  = nsp
      method/nsp/electron-number      = 40
      method/nsp/occlevel     = [ 1 - 20 ]
      orbital-independence-threshold  = 0.007
      convergence/type        = density
      convergence/threshold   = 1e-4
      convergence/threshold-energy    = 1e-5
      scf-acceleration        = damping
      scf-acceleration/damping/damping-factor = 0.65
      xc-potential    = b3lyp
      scf-acceleration/damping/damping-type = density_matrix

>>>>MOLECULE
      geometry/cartesian/unit = angstrom
      geometry/cartesian/input        = {
              N        -1.888000        0.035000       -0.211000
              H        -1.766000        0.945000        0.189000
              H        -1.817000        0.099000       -1.205000
              C        -0.758000       -0.730000        0.287000
              H        -0.893000       -0.915000        1.372000
              H        -0.720000       -1.725000       -0.200000
              C         0.529000        0.065000        0.064000
              O         0.520000        1.294000        0.114000
              O         1.742000       -0.451000       -0.186000
              H         1.692000       -1.400000       -0.203000
      }end

      basis-set/orbital       = {
              H = "O-HYDROGEN (41) DZVP"
              O = "O-OXYGEN (621/41) by FS"
              C = "O-CARBON (621/41) by FS"
              N = "O-NITROGEN (621/41) by FS"
      }end

      basis-set/density-auxiliary     = {
              H = "A-HYDROGEN (4,1;4,1) from deMon"
              O = "A-OXYGEN (7/2;7/2) by FS"
              C = "A-CARBON (7/2;7/2) by FS"
              N = "A-NITROGEN (7/2;7/2) by FS"
      }end

      basis-set/exchange-auxiliary    = {
              H = "A-HYDROGEN (4,1;4,1) from deMon"
              O = "A-OXYGEN (7/2;7/2) by FS"
              C = "A-CARBON (7/2;7/2) by FS"
              N = "A-NITROGEN (7/2;7/2) by FS"
      }end

Note

Input files can be changed with a startup option of ProteinDF.

4.3. Executing ProteinDF

Configure the environment variable PDF_HOME appropriately, and execute ProteinDF (in serial mode).

% ${PDF_HOME}/bin/PDF.x

If the program terminates properly, the system returns to the command prompt.

4.4. Outputting results

The calculation results are output in files. The following shows an example of the output:

Note

The output file location can be changed with a startup option of ProteinDF.

The beginning of the output file shows the version of ProteinDF and the number of parallel processes (MPI process count, the number of OpenMP threads). Make sure that the calculation was performed as intended.

[0:2012/**/07 17:17:02:INFO] **************************************
[0:2012/**/07 17:17:02:INFO] ProteinDF version 20xx.x:xxxx (serial)
[0:2012/**/07 17:17:02:INFO]
[0:2012/**/07 17:17:02:INFO]  OpenMP threads: 12
[0:2012/**/07 17:17:02:INFO]

The calculation is performed according to the procedure described in step_control. The output date is indicated at left in the log

===============================================
 >>>> INTEGRAL
===============================================
 >>>> Hpq

 ...

===============================================
 >>>> GUESS
===============================================

 ...

===============================================
 >>>> SCF
===============================================

Energy data are output in the Total Energy block.

------------------------------------------------
 >>>> Total Energy
------------------------------------------------
 Ts+Vn          =        -745.7264230071891689
 E_J[Rho, Rho~] =         629.4907739256434525
 E_J[Rho~,Rho~] =        -311.1519936178789294
 E_xc(pure)     =         -27.8222685895087842
 E_K            =          -6.8830892822011522
 E_nuclei       =         179.8930288412503558
 TE             =        -282.1999717298841688
------------------------------------------------

Other information, such as calculation size or cut off data, is output as needed.

When the calculation completes properly, the CPU time and elapsed time are output.

************************************************
 ProteinDF successful completion
 CPU_TIME:         3454 sec
 ELAPS_TIME:        542 sec
************************************************

Table Of Contents

Previous topic

3. Installation

Next topic

5. Input files

This Page