Gamess-US: automatic compilation on a Ubuntu Linux system

Procedure using Ubuntu 7.04 to 10.04, x86 or x86_64

Method 1: Live Linux: We start on a Windows or a Linux box with Ubuntu 10.04 running from CD. Download and burn the CD.
ubuntu-10.04-desktop-i386.iso ubuntu.com  699 MB
ubuntu-10.04-desktop-amd64.iso ubuntu.com  698 MB
Start Ubuntu from the CD.

Method 2: Using Wubi installer downloads and installs a permanent Ubuntu system in a ~18 GB file under Windows. You do not have to burn and run a live CD. Ubuntu automatically runs when the installation has finished.

Now, save your downloaded gamess-current.tar.gz or gamess-current.tar.Z in /home/$USER.

Download and save the following script also in /home/$USER:
for Gamess versions up to12 JAN 2009 (R3) ubuntugam, for targets linux64 and linux32
for Gamess version
 
 
 

25 MAR 2010 (R2)
 
 
 

ubugam10, for targets linux64 and linux32. Note that for this
new Gamess version the older script, ubuntugam, does not
work at all!

ubugamntv10, for targets linux64 and linux32, with enabled
NEO, Tinker, and VB2000 modules.

Overview

Running one of the scripts ubuntugam, ubugam10, or ubugamntv10 automatically updates all necessary programs and libraries, incl. libblas from Atlas. Then it starts a build script (gamessmake, buildgamess10, or buildgamntv10). This unpacks the Gamess tarball, edits all required scripts from the Gamess-US distribution to reflect your environment, and, then builds actvte.x, ddikick.x, and gamess.00.x. To test the execs all examXX.inp are run and results checked.

 

In about 10 to 15 min ddikick.x and gamess.00.x are made and Checktst tells you: All 44 test results are correct. Congratulations! your new Gamess works. Here is a summary of live sessions with gamessmake and buildgamess10 on 32- and 64-bit machines using Atlas math libraries for linking.

Other Mathlib than Atlas

If you want to use MKL or ACML Blaslibraries - both up to 30% faster than Atlas, but need registering and installing - download config which has been corrected for bugs and replace the original in ~/gamess. Run ./config composing a new install.info with the path to your mathlibrary (e.g. /opt/acml4.4.0). Then run lked again with a new name for gamess, e.g. ./lked gamess 01 >& lked1.log. Now you can compare gamess.00.x with Atlas to gamess.01.x with ACML.
Note for linux32 users: For linking to the MKL libraries lked (the linker script of gamess) has to be edited. Download lked_mkl32 to replace the original, faulty lked, run ./config and then
./lked_mkl32 gamess 02 >& lked_mkl32.log This is correct for mkl 10.2.5.035 or any other version 10.x.y.z.

Rungms on multicore node

rungms has been edited to make use of up to four cores of a dual or quad core CPU, running in parallel. Just call Gamess with:
rungms job(.inp) 00 2 (or 3, or 4) > job.log   to engage the processors of your SMP. With
rungms job(.inp) 00 1 job.out   or rungms job(.inp) > job.log   (with gamess.00.x) the normal single CPU run is started. If you use gameix, then you can select the number of CPU's in the Configuration file on the MasterMenu.
Make sure to check whether Gamess knows how to run your jobtype in parallel. There are many that are not (yet) parallel enabled.
Return to main page of Gamess compilation.