Running Fluent on a remote server
If your simulation requires more processing power, you may need to remotely run the calculations on a larger server. This tutorial is for those who have access to the Swanson computer lab on Cornell University's Ithaca campus. Instructions are provided in the following video and summarized below:
- Append DNS suffix
- Go to Control Panel >> Network and Internet >> Local Area Connection
- Click Properties in the “Local Area Connection Status” window
- Select Internet Protocol Version 4 (TCP/IPv4)
- Click Properties >> Advanced
- Go to the “DNS” tab and under the window “Append these DNS suffixes”, click Add
- In the “TCP/IP Domain Suffix” window, type in: mae.cornell.edu
- Map Network drive
- Go to “My Computer” and select “Map network drive” on the top bar
- For “Folder”, type in: \\en-ma-simcafe1\workspace
- Click “Finish” and make note the drive letter (e.g., ‘Z:’)
- Set up FLUENT
- Select “Use Job Scheduler”
- Select “Parallel per Job Scheduler Solver”
- Choose # of processes. (Note that the academic version of FLUENT is limited to 2 processes.)
- Under the “General Options” tab
- For working directory, type in the name of mapped drive (e.g., ‘Z:\’)
- For “Fluent Root Path”, enter: \\en-ma-simcafe1\fluent
- Under the “Scheduler” tab
- For Computer Cluster Head Node Name, enter: en-ma-simcafe1.mae.cornell.edu
- If you are using UDF, check “Setup Compilation Environment for UDF” under the “Environment” tab
- Click OK to run FLUENT
- In the print-out console, make sure all licenses are available and simulation is running
- Check progress of Fluent solver
- Check that HPC Cluster Manager is installed on the computer
- Run HPC Job Manager
- Select "active" on the left tab to view active jobs