WebMO - Computational chemistry on the WWW
Recent news

WebMO 18.0 is now available for free download!

WebMO 18.0 Pro and Enterprise have a variety of additional features and is available for purchase.

The WebMO app is now available for iOS and Android.

August 23, 2019

Gamess issue with NMR calculation? Log Out | Topics | Search
Moderators | Edit Profile

WebMO Support Forum » Running WebMO Jobs » Other » Gamess issue with NMR calculation? « Previous Next »

Author Message
Dean Waldow
Unregistered guest
Posted on Tuesday, January 23, 2018 - 1:39 pm:   

I am running GAMESS (VERSION = 18 AUG 2016 (R1) ) with WebMO 17.0.010e) and when I submit a job for a molecule with 20 atoms to calculate NMR spectra initially I got a memory error. I changed the memory option to 1200000 (MW units I believe) and then the job didn't fail due to memory. But, it just endless says it is calculating GIAO TWO-ELECTRON INTEGRALS. For reference, gaussian finishes in about 7 minutes on this virtual box. Any ideas if this is a WebMO issue or a Gamess issue? Perhaps it is a WebMO setup / config issue? It would be interesting to be able to compare results.
JR Schmidt
Moderator
Username: Schmidt

Post Number: 589
Registered: 11-2006
Posted on Tuesday, January 23, 2018 - 2:12 pm:   

The best way to check this would be to run the same job / input file on the command line.
Dean Waldow
Unregistered guest
Posted on Tuesday, January 23, 2018 - 6:09 pm:   

I have never run a job directly from the command line. Do you have a reference for doing that for gamess? I'll see what I can find out for doing this. Thanks...
JR Schmidt
Moderator
Username: Schmidt

Post Number: 590
Registered: 11-2006
Posted on Wednesday, January 24, 2018 - 7:52 am:   

You can do it using the 'rungms' script provided with GAMESS.

See step 11 of the GAMESS installation instructions:
https://www.webmo.net/support/gamess_linux.html
Dean Waldow
Unregistered guest
Posted on Thursday, January 25, 2018 - 1:03 am:   

I followed step 11 and it appears to have finished well. So, I found the input.inp file for the job that hung before. I ran that input.inp manually and after 15 minutes the /tmp/input.dat file had not changed since the first minute. This same collection of settings except for memory ran on gaussian in 7 minutes or so.

input.inp from webmo

$CONTRL SCFTYP=RHF RUNTYP=NMR
ICHARG=0 MULT=1 COORD=CART $END
$SYSTEM MEMORY=185000000 $END
$BASIS GBASIS=N31 NGAUSS=6 NDFUNC=1 $END
$DATA
C13H10Br2
C1 1
C 6 0.00000000 0.00000000 0.00000000
C 6 -1.27630000 -0.82351100 0.00408900
C 6 -2.22983900 -0.65734300 0.99678300
C 6 -3.40845300 -1.39095000 0.99539500
C 6 -3.63087500 -2.30145400 -0.01635800
C 6 -2.69622200 -2.48925300 -1.02165500
C 6 -1.52820000 -1.75144600 -1.00324200
H 1 -0.80134100 -1.90530200 -1.78095200
H 1 -2.87743600 -3.20041000 -1.80472300
Br 35 -5.24214400 -3.31594100 -0.03328800
H 1 -4.13573000 -1.25057400 1.77210000
H 1 -2.06237200 0.05366800 1.78655100
C 6 1.27754900 -0.82097400 0.00216500
C 6 2.21719800 -0.68385400 -1.00800900
C 6 3.39536200 -1.41837200 -1.00204300
C 6 3.63166900 -2.29959800 0.03211300
C 6 2.71119300 -2.45741400 1.05577600
C 6 1.54337800 -1.71958600 1.03238800
H 1 0.82761900 -1.84974800 1.82455600
H 1 2.90348900 -3.14574900 1.85646300
Br 35 5.24266300 -3.31444700 0.05532700
H 1 4.11176100 -1.30083800 -1.79250800
H 1 2.03904100 0.00484000 -1.81514900
H 1 0.00088100 0.64911600 -0.87021200
H 1 -0.00257700 0.65528100 0.86551500
$END

A tail of the /tmp$ tail input.dat after a minute into the run and
it has not changes in another 15 minutes.

C 6.20115 -0.20115 6.15994 -0.15994
H 0.78353 0.21647 0.82692 0.17308
H 0.76717 0.23283 0.81701 0.18299
BR 35.04932 -0.04932 35.02023 -0.02023
H 0.76824 0.23176 0.81736 0.18264
H 0.79089 0.20911 0.83153 0.16847
H 0.80244 0.19756 0.83737 0.16263
H 0.80272 0.19728 0.83746 0.16254
MOMENTS AT POINT 1 X,Y,Z= 0.000000 0.000000 0.000000
DIPOLE -0.006488 -2.386617 -0.006254

When I ran this before from webmo it ran over night and didn't finish.

Is this helpful?
JR Schmidt
Moderator
Username: Schmidt

Post Number: 591
Registered: 11-2006
Posted on Thursday, January 25, 2018 - 12:56 pm:   

This calculation appears to be extremely slow. I don't know the details of the algorithm used by GAMESS, but the calculation of the relevant two-electron integrals is very slow. This doesn't appear to be a WebMO issue -- just GAMESS.

Try PC-GAMESS/Firefly. It is likely faster.
Dean Waldow
Unregistered guest
Posted on Thursday, January 25, 2018 - 12:59 pm:   

BTW... the cat /proc/sys/kernel/shmmax command essentially is set to infinity...

webmo@compute2$ cat /proc/sys/kernel/shmmax
18446744073692774399

I don't think this is an issue.
Dean Waldow
Unregistered guest
Posted on Thursday, January 25, 2018 - 1:08 pm:   

OK... I also let it run almost 24 hours with no progress. Will see what makes sense then. Might just avoid NMR on gamess.

Add Your Message Here
Post:
Username: Posting Information:
This is a public posting area. Enter your username and password if you have an account. Otherwise, enter your full name as your username and leave the password blank. Your e-mail address is optional.
Password:
E-mail:
Options: Post as "Anonymous"
Enable HTML code in message
Automatically activate URLs in message
Action:

Topics | Last Day | Last Week | Tree View | Search | Help/Instructions | Program Credits Administration