Ansys Fluent is a computational fluid dynamics application. Fluent version v19.3 (i.e. 2019r1) is available.<\/p>\n
fluent<\/code> command on the CSF4 login node to have it automatically submit a batch job. You must<\/strong> write your own jobscript and submit it using the sbatch<\/code> command. See below for example jobscripts.<\/div>\nRestrictions on Use<\/h2>\n
Only users who have been added to the Fluent group can run the application. Owing to licence restrictions, only users from the School of MACE and one specific CEAS Research Group can be added to this group. Requests to be added to the Fluent group should be submitted please via our form<\/a>.<\/p>\nThe maximum job size is 16 cores (this is a limitation imposed by the license).<\/p>\n
Fluent jobs must not be run on the login node. If you need to run an interactive job please use srun<\/a> as detailed below.<\/p>\nSet Up Procedure<\/h2>\n
Once you have been added to the Fluent group, you will be able to access the executables by using one<\/em> of the following module commands:<\/p>\n\r\nmodule load fluent\/2024R2\r\nmodule load fluent\/2022R1\r\nmodule load fluent\/19.3\r\nmodule load fluent\/19.2<\/pre>\nRequired Input Files<\/h2>\n
To run a fluent job you will need a fluent journal<\/em> file, a fluent case<\/em> file and usually a data<\/em> file which the case file will load.<\/p>\nJournal (.jou) files<\/h3>\n
The journal file can be created either by asking fluent to write it while using the fluent GUI (which can be run from the CSF or from a desktop PC installation). Alternatively, the journal file can be written by hand. This is common – they are often very simple files which load the case and data files, then start the simulation. Here are two examples:<\/p>\n
For transient cases:<\/p>\n
rc filename<\/em>.cas\r\nrd filename<\/em>.dat\r\nsolve dual-time-iterate 40000 200\r\n<\/pre>\nwhere 40000 is the number of time steps and 200 is the max number of iterations per time step.<\/p>\n
For steady-state cases. This also demonstrates how to use gzip<\/code> compressed .cas<\/code> and .dat<\/code> files:<\/p>\nrc filename<\/em>.cas.gz\r\nrd filename<\/em>.dat.gz\r\nit 200\r\n<\/pre>\nwhere 200 is the number of iterations. The case and data files were first compressed using the following commands to reduce the size of the files. When the job runs, Fluent will be able to uncompress the files to read them:<\/p>\n
gzip filename<\/em>.cas\r\ngzip filename<\/em>.dat\r\n<\/pre>\nUDF Files<\/h3>\n
If you need to compile a User Defined Function<\/em> (UDF) file, this should be done on the CSF. Fluent will usually do this automatically when your job runs provided you have saved the case<\/em> file with the UDF enabled in the GUI (See the Define \/ User-Defined \/ Function Hooks<\/em> menu. You can check your case<\/em> has the correct line in it as follows:<\/p>\n# For an normal (uncompressed) .cas file\r\ngrep libudf filename<\/em>.cas\r\n\r\n# For a compressed .cas.gz file\r\nzgrep libudf filename<\/em>.cas.gz\r\n<\/pre>\nThese commands should print something similar to the following if you are using a UDF<\/p>\n
(udf\/compile\/files ((\"libudf\" (source \"\/scratch\/username<\/em>\/my_project<\/em>\/my_udf<\/em>.c\") (header))))\r\n<\/pre>\nNote that the path to the .c<\/code> file will need to be correct for the CSF, otherwise fluent will not be able to find the file. If you saved your case<\/em> file on a Windows PC it could have a path that is only valid on that PC. You DO NOT need to edit your .cas file to add the above line. You can use the Fluent GUI to re-save your case<\/em> file on the CSF with the correct path.<\/p>\nAlternatively, you might be able to add some lines to the journal<\/em> file. However, there are many possible options – too many to list here – so we recommend using the Fluent GUI to make the settings. However, here is an example of lines added to the .jou<\/code> file for a specific type of UDF:<\/p>\n# Add the following to the .jou file to enable the UDF\r\n\/define user-defined compiled-functions compile libudf yes my_udf<\/em>.c\r\n\/define user-defined compiled-functions load\r\n\/define user-defined function-hooks DPM_BC<\/em> my_udf<\/em>::libudf\r\n #\r\n # This example uses a Discrete Phase Model\r\n # Boundary Condition UDF\r\n<\/pre>\nCompiling the UDF code<\/h3>\n
You UDF source code (e.g., my_udf.c<\/code>) must be compiled in to a library that Fluent can read. Please note, if you have compiled it on a Windows PC, the library will NOT work on the CSF. You must recompile it on the CSF.<\/p>\nYou can either let Fluent compile the UDF code when your job runs, or you can compile it on the CSF before your job runs then fluent won’t need to do it itself. This is also a good way to check your UDF code actually compiles! If you wait until fluent compiles it (when your job runs), any mistakes in your code will mean you have to fix the code then resubmit the job. It is much easier to try to compile the UDF on the login node before the job runs and make any corrections to the code if you need to do so.<\/p>\n
\nIf you have compiled the UDF on Windows before transferring the files to the CSF, please remove any automatically-generated Windows files from the CSF folder (e.g. udf_names.c<\/code>, ud_io1.h<\/code> and any existing libudf\/<\/code> folder.) These files will cause the compilation on the CSF to fail. You should only have the source files you wrote (e.g., my_udf.c<\/code>) in your CSF directory. All of the necessary files will then be generated by fluent when you compile the UDF on the CSF.<\/div>\nIf you wish to check your UDF code compiles on the CSF before<\/em> submitting a job to the batch system, you can run a helper script named csf_compile_udf<\/code> on the CSF to do the UDF compilation:<\/p>\ncsf_compile_udf simtype1<\/em> simtype1<\/em>_node simtype1<\/em>_host ...\r\n\r\n# Example 1: Compile for a 3ddp (3D double-precision) simulation:\r\ncsf_compile_udf 3ddp 3ddp_node 3ddp_host\r\n\r\n# Example 2: Compile for a 2d (2D single-precision) simulation:\r\ncsf_compile_udf 2d 2d_node 2d_host\r\n\r\n# For more info:\r\ncsf_compile_udf --help\r\n<\/pre>\nThis script will compile your my_udf.c<\/code> file (the script searches for .c<\/code> files in the current directory and checks whether they appear to be Fluent UDF files) in to a libudf.so<\/code> file. It will create a new sub-directory named libudf<\/code> in the current directory (where you run the script). Each simulation type (2d, 3ddp, …) will have its own libudf.so<\/code> file in directories inside the libudf\/<\/code> sub-dir.<\/p>\nRunning Fluent in Parallel in batch<\/h2>\n
Please note:<\/strong> the versions of fluent available do not currently support automatic job-submission to SLURM. Hence you must<\/strong> write a SLURM jobscript and submit it using the sbatch<\/code> command.<\/p>\nFor example, the following jobscript file named fluent-batch.slurm<\/code>:<\/p>\n#!\/bin\/bash --login\r\n# Job will run in the current directory by default\r\n#SBATCH -n 8 # 8 cores. Can also use --ntasks=8\r\n#SBATCH -p multicore # Single-node parallel job\r\n\r\n# The default output file will be slurm-JOBID.out. To use a SGE-style filename:\r\n#SBATCH -o %x.o%j # Output file similar to CSF3: jobscriptname.oJOBID\r\n\r\n# Choose your required version\r\nmodule load fluent\/19.3\r\n\r\nfluent 3d<\/strong> -g -t$SLURM_NTASKS -i input.jou<\/em>\r\n # #\r\n # #\r\n # # Replace input.jou with your input file.\r\n #\r\n # If using an UDF this must match the type used for compilation\r\n # (e.g., 2d or 3d or 3ddp). You should not use the _node or _host\r\n # type here.\r\n<\/pre>\nSubmit the job using:<\/p>\n
sbatch fluent-batch.slurm<\/pre>\nYou can then check on the status of the job using<\/p>\n
squeue --me\r\n<\/pre>\nThe maximum job size you can request is 16 due to licensing restrictions.<\/p>\n
Running Serial Interactive Jobs via srun<\/h2>\n
The running of serial<\/strong>, interactive jobs is tolerated. This is mainly used to set up a simulation which can then be saved to file and run in batch. You may need to use the GUI to set up a UDF. If you are setting up a parallel UDF please contact us via our form<\/a>.<\/p>\nYou may run fluent interactively using the srun<\/code> option which has a time limit of 1 hour.<\/p>\nStarting Fluent interactively<\/h3>\nsrun -n 1 --pty bash\r\n<\/pre>\nThen, once srun<\/code> has returned a commandline on a compute node, load the appropriate environment module and start Fluent:<\/p>\nmodule load fluent\/19.3 # Load the modulefile for the version you require\r\nfluent & ## The & ensures you get your command line and the GUI at the same time.\r\n<\/pre>\nNotes:<\/p>\n
\n- Linux: if the render window section of the GUI does not display your model try setting the following before running fluent:\n
export LIBGL_ALWAYS_INDIRECT=1<\/pre>\n<\/li>\n
- alternatively run fluent using:\n
fluent -driver x11<\/pre>\n<\/li>\n
- Windows: if the render window is slow to redraw or flashes try running fluent using:\n
fluent -driver x11<\/pre>\n<\/li>\n<\/ol>\nNote:<\/strong> For the GUI to work you must have an X server running on your PC. See the instructions on the using GUI based applications<\/a> page.<\/p>\nFurther Information<\/h2>\n
Further information on Fluent and other CFD applications may be found by visiting the MACE CFD Forum<\/a>.<\/p>\nMore information about srun<\/a> on the CSF.<\/p>\n","protected":false},"excerpt":{"rendered":"Overview Ansys Fluent is a computational fluid dynamics application. Fluent version v19.3 (i.e. 2019r1) is available. CSF3 users please note: it is NOT possible to run the fluent command on the CSF4 login node to have it automatically submit a batch job. You must write your own jobscript and submit it using the sbatch command. See below for example jobscripts. Restrictions on Use Only users who have been added to the Fluent group can run.. Read more »<\/a><\/p>\n","protected":false},"author":4,"featured_media":0,"parent":49,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-88","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/ri.itservices.manchester.ac.uk\/csf4\/wp-json\/wp\/v2\/pages\/88","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/ri.itservices.manchester.ac.uk\/csf4\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/ri.itservices.manchester.ac.uk\/csf4\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/ri.itservices.manchester.ac.uk\/csf4\/wp-json\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"https:\/\/ri.itservices.manchester.ac.uk\/csf4\/wp-json\/wp\/v2\/comments?post=88"}],"version-history":[{"count":21,"href":"https:\/\/ri.itservices.manchester.ac.uk\/csf4\/wp-json\/wp\/v2\/pages\/88\/revisions"}],"predecessor-version":[{"id":1462,"href":"https:\/\/ri.itservices.manchester.ac.uk\/csf4\/wp-json\/wp\/v2\/pages\/88\/revisions\/1462"}],"up":[{"embeddable":true,"href":"https:\/\/ri.itservices.manchester.ac.uk\/csf4\/wp-json\/wp\/v2\/pages\/49"}],"wp:attachment":[{"href":"https:\/\/ri.itservices.manchester.ac.uk\/csf4\/wp-json\/wp\/v2\/media?parent=88"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}