Aborted default .OPTIMIZE (max DIIS iterations reached)

Find answers or ask questions regarding Dalton calculations.
Please upload an output file showing the problem, if applicable.
(It is not necessary to upload input files, they can be found in the output file.)

Post Reply
parh
Posts: 2
Joined: 08 Oct 2015, 11:11
First name(s): Par
Last name(s): Hakansson
Affiliation: University of Southampton
Country: United Kingdom

Aborted default .OPTIMIZE (max DIIS iterations reached)

Post by parh » 08 Oct 2015, 12:27

Hi,
I see a problem using the default .OPTIMIZE settings. (using DALTON2015.2)
In this case two iterations are performed before
"DIIS aborted because max DIIS iterations reached"

There is also an report: "QC-SCF not implemented/tested for DFT"
Is this the reason the optimization stopped?
Does "QC-SCF" add some additional criteria or iterations in SCF?

**DALTON INPUT
.OPTIMIZE !Request default geom. opt
.RUN PROPERTIES
**WAVE FUNCTION
.DFT
B3PW91
*END OF INPUT

I have also used the second order in energy optimization adding
*OPTIMIZE
.2NDORD
In this case the optimization completes.
However, since ".2NDORD" require more time, the default ".OPTIMIZE" would be preferred.

If I understand the "QC-SCF not implemented/tested for DFT" correct as the reason the optimization stopped,
does the default setting for ".OPTIMIZE" work better with HF instead of DFT?

Regards,
Pär

My standard output is:
*****************************************
**** OUTPUT FROM DALTON SHELL SCRIPT ****
*****************************************

DALTON release 2015.1

Invocation: /local/software/dalton/2015.2/dalton/dalton -N 16 -t /scratch/ph1q07/DALTON_scratch/ -b /local/software/dalton/2015.2/dalton/basis/ optRB3PW91.dal TTF001.mol

Thu Oct 8 04:10:54 BST 2015

Calculation: optRB3PW91_TTF001 (input files: optRB3PW91.dal and TTF001.mol)
PID : 1137
Input dir : /home/ph1q07/usr/DALTON/13C2DMHY_EXAMPLE
Scratch dir: /scratch/ph1q07/DALTON_scratch/DALTON_scratch_ph1q07/optRB3PW91_TTF001_1137


INFO : OMP_NUM_THREADS set to 1 because it was not defined by user
INFO : and the cores are probably used by MPI


output from the communication group generator:
1 intra-node group has been built.


DALTON: default work memory size used. 64000000

DALTON: master work memory size also used for slaves. 64000000


Work memory size (LMWORK+2): 64000002 = 488.28 megabytes; node 0

0: Directories for basis set searches:
/local/software/dalton/2015.2/dalton/basis/:/home/ph1q07/usr/DALTON/13C2DMHY_EXAMPLE:/home/local/software/dalton/2015.2/dalton/basis
Work memory size (LMWORK+2): 64000002 = 488.28 megabytes; node 1

1: Directories for basis set searches:
/local/software/dalton/2015.2/dalton/basis/:/home/ph1q07/usr/DALTON/13C2DMHY_EXAMPLE:/home/local/software/dalton/2015.2/dalton/basis
Error in mpirun -np 16 /home/local/software/dalton/2015.2/dalton/dalton.x, exit code 100
Backup: renaming /home/ph1q07/usr/DALTON/13C2DMHY_EXAMPLE/optRB3PW91_TTF001.tar.gz to /home/ph1q07/usr/DALTON/13C2DMHY_EXAMPLE/optRB3PW91_TTF001.tar.gz.0
optRB3PW91_TTF001.tar.gz has been copied to /home/ph1q07/usr/DALTON/13C2DMHY_EXAMPLE
Backup: renaming /home/ph1q07/usr/DALTON/13C2DMHY_EXAMPLE/optRB3PW91_TTF001.out to /home/ph1q07/usr/DALTON/13C2DMHY_EXAMPLE/optRB3PW91_TTF001.out.0
----------------------------------------------------------
WARNING for ERROR (Dalton program stopped with exit code 100)
- THEREFORE scratch directory /scratch/ph1q07/DALTON_scratch/DALTON_scratch_ph1q07/optRB3PW91_TTF001_1137 is not deleted
by this script so you may restart the calculation, if possible
( /scratch/ph1q07/DALTON_scratch/DALTON_scratch_ph1q07/optRB3PW91_TTF001_1137 may be deleted by your queueing system )
- REMEMBER to delete /scratch/ph1q07/DALTON_scratch/DALTON_scratch_ph1q07/optRB3PW91_TTF001_1137 yourself !!!

Output from 'grep -n ERROR' :
2223: --- SEVERE ERROR, PROGRAM WILL BE ABORTED ---
----------------------------------------------------------
1 WARNINGS have been issued by Dalton,
output from 'grep -n -i WARNING' (max 10 warnings):
2221:WARNING !!! DIIS aborted because max DIIS iterations reached !

Thu Oct 8 04:12:22 BST 2015
*****************************************
Attachments
optRB3PW91_TTF001.out
My optimization *out file
(145.69 KiB) Downloaded 289 times

hjaaj
Posts: 337
Joined: 27 Jun 2013, 18:44
First name(s): Hans Jørgen
Middle name(s): Aagaard
Last name(s): Jensen
Affiliation: Universith of Southern Denmark
Country: Denmark

Re: Aborted default .OPTIMIZE (max DIIS iterations reached)

Post by hjaaj » 08 Oct 2015, 16:30

The problem is not really DIIS or QC-SCF, the problem is that the second geometry is ridiculous. Some of the coordinates:

Code: Select all

  C         9.5915236508         -289.5048402884          172.3619826316
  C       -13.8843317627         -298.3977024655          122.2042802263
  C        37.7802280595         -278.8355389087          232.7359652228
  C       -41.9562486219         -309.1469863429           61.8232326161
  C       -32.7761289020          -42.2229831321           63.2953809315
  C        82.0368310591           13.4474569500           16.0609440536
  C       -46.0832233247          175.3415307881           23.3974726014
  C      -173.0792291975          132.5087162921         -281.7464391259
  C       151.3618037489          260.1267800054          -84.7845866651
  O         1.1135479365          -55.1916522919          137.4066183567
  O        43.8384886100          -24.5715014945          -39.4933978808
 
Definitely a bug. I have a feeling that it would be OK if you run a sequential Dalton calculation instead of a parallel with 16 processors. Could you try that ("dalton -N 1 ..."), and report the result here?

parh
Posts: 2
Joined: 08 Oct 2015, 11:11
First name(s): Par
Last name(s): Hakansson
Affiliation: University of Southampton
Country: United Kingdom

Re: Aborted default .OPTIMIZE (max DIIS iterations reached)

Post by parh » 08 Oct 2015, 21:44

I get almost identical unrealistic coordinates as second geometry when
one processor is used.

regards,
Pär

Code: Select all

                                Next geometry (au)
                                ------------------

  C         9.5906474422         -289.5343785394          172.1655403234
  C       -13.8869375208         -298.5301484906          122.0642370337
  C        37.7807508765         -278.7409515663          232.4718688902
  C       -41.9615393826         -309.4032398647           61.7515663757
  C       -32.7800713143          -42.2804609271           63.2226806301
  C        82.0450378902           13.5026488778           16.0436657209
  C       -46.0883056233          175.2832775918           23.3709610759
  C      -173.0928796453          132.5225842340         -281.4326147345
  C       151.3780357626          260.2971763429          -84.6826844229
  O         1.1114051584          -55.1091793666          137.2506943388
  O        43.8429608125          -24.6659431205          -39.4463789175
Attachments
optRB3PW91_TTF001.out
Single processor default .OPTIMIZE
(139.43 KiB) Downloaded 281 times

xiongyan21
Posts: 174
Joined: 24 Sep 2014, 08:36
First name(s): yan
Last name(s): xiong
Affiliation: CENTRAL CHINA NORMAL UNIVERSITY
Country: China

Re: Aborted default .OPTIMIZE (max DIIS iterations reached)

Post by xiongyan21 » 13 Oct 2015, 10:59

Prof. Jensen's comment is very insightful.
When I added MAX DIIS 500 into *SCF INPUT, the run got !!! Info: DIIS restarted because it was stalled ... for many times, and still failed because of WARNING !!! DIIS aborted because max DIIS iterations reached !
I think that the initial geometry is not so feasible that it produces a wrong second one, making the optimization unsuccessful.
Last edited by xiongyan21 on 13 Oct 2015, 12:59, edited 2 times in total.

Post Reply

Who is online

Users browsing this forum: No registered users and 4 guests