Linux Software Notes![]()
ABAQUS, ANSYS, and Fluent generate such a high volume of network traffic when run from the file server that they slow activity for everyone with accounts on that file server. To avoid dramatic network slowdowns when running ABAQUS, ANSYS, and Fluent, you must create a directory in /var/tmp, copy your input files to that directory, cd to that directory, and then run the application. The applications will not start until you do this. Read below the detailed instructions for ABAQUS, ANSYS, Fluent.
How to Run ABAQUS
1. Create a directory in your home directory for ABAQUS files and put ABAQUS-related files in that directory.
mkdir problem1 where problem1 is the directory you are creating2. Create a directory in /var/tmp to store your output ABAQUS files. ABAQUS creates very large scratch files which will be written to this directory. This is temporary space which is NOT backed up and is deleted at the end of every semester or sooner if the disk fills up. You must copy any files you want to keep back to your home directory.
mkdir /var/tmp/name where name is the directory you are creating. You can use your login name to name this directory.3. Copy your ABAQUS files to /var/tmp/name
cp ~/problem1/xxx /var/tmp/name where xxx are the files you want to copy from the directory you created in step 1 above.4. Change to the /var/tmp/name directory
cd /var/tmp/name
5. Run ABAQUS. Type abaqus cae at the command prompt. ABAQUS will automatically set your working directory to /var/tmp/name as long as you start ABAQUS from that directory.6. When you have ABAQUS results, copy those files to your home directory
cp xxx /path where path is where you want to store these files.
If you choose to store them in the directory created in step 1 above, use cp xxx ~/problem1From the start menu go to Search and type abaqusmenu or abaqus cae from a shell prompt. Temporary files must be stored in /var/tmp.
If you have questions, talk to the consultants in 1253 SC, 335-5055.
ANSYS 9.0 Notes
The ANSYS launcher does not work properly when displayed on a remote machine. For this reason, the launcher is run only if you are sitting directly on an Itanium workstation. If you run ANSYS using WinaXe from a Windows machine, ANSYS launches the main program graphical interface and skips the launcher.
How to Run ANSYS
1. Create a directory in your home directory for ANSYS files and put ANSYS-related files in that directory.
mkdir problem1where problem1 is the directory you are creating2. Create a directory in /var/tmp to store your output ANSYS file. ANSYS creates very large scratch files which will be written to this directory. This is temporary space which is NOT backed up and is deleted at the end of every semester or sooner if the disk fills up. You must copy any files you want to keep back to your home directory.
mkdir /var/tmp/name where name is the directory you are creating. You can use your login name to name this directory.3. Copy your ANSYS files to /var/tmp/name
cp ~/problem1/xxx /var/tmp/name where xxx are the files you want to copy from the directory you created in step 1 above.4. Change to the /var/tmp/name directory
cd /var/tmp/name
5. Run ANSYS. Type ansys at the command prompt. ANSYS will automatically set your working directory to /var/tmp/name as long as you start ANSYS from that directory.6. When you have ANSYS results, copy those files to your home directory
cp xxx /path where path is where you want to store these files.
If you choose to store them in the directory created in step 1 above, use cp xxx ~/problem1Run a batch job in the background
You can run ANSYS as a batch job, put it in the background and log out by using the following command. The & tells the system to run the job in the background; nohup tells the system to continue the job after you log out. The input file is your file of ANSYS commands; the output file will capture anything ANSYS normally writes to the screen. Name the output file anything you want.nohup ansys -b < inputfile > outputfile &
If you have questions, talk to the consultants in 1253 SC, 335-5055.
How to Run Fluent
1. Create a directory in your home directory for Fluent files and put Fluent-related files in that directory.
mkdir problem1where problem1 is the directory you are creating2. Create a directory in /var/tmp to store your output Fluent files. Fluent creates very large scratch files which will be written to this directory. This is temporary space which is NOT backed up and is deleted at the end of every semester or sooner if the disk fills up. You must copy any files you want to keep back to your home directory.
mkdir /var/tmp/name where name is the directory you are creating. You can use your login name to name this directory.3. Copy your Fluent files to /var/tmp/name
cp ~/problem1/xxx /var/tmp/name where xxx are the files you want to copy from the directory you created in step 1 above.4. Change to the /var/tmp/name directory
cd /var/tmp/name
5. Run Fluent. Type fluent at the command prompt. Fluent will automatically set your working directory to /var/tmp/name as long as you start Fluent from that directory.6. When you have Fluent results, copy those files to your home directory
cp xxx /path where path is where you want to store these files.
If you choose to store them in the directory created in step 1 above, use cp xxx ~/problem1If you have questions, talk to the consultants in 1253 SC, 335-5055.
Source the File
To run Fluent or Gambit, source the appropriate setup file for your shell. The setup files set your Path environment variable so that all the Fluent and associated programs will be available to you.
For a C shell, type
source /usr/ui/class/aux/fluent_setup.csh
For a Bash shell, type
. /usr/ui/class/aux/fluent_setup.sh
![]()
CSS | College of Engineering | © The University of Iowa 2008.
All rights reserved. |