CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   Large Problem (https://www.cfd-online.com/Forums/fluent/198666-large-problem.html)

edmondlam February 14, 2018 10:35

Large Problem
 
Hi guys,

I have a large problem to solve with FLUENT. I did the mesh with ICEM CFD. I would like to ask if there are tricks to reduce the memory usage of FLUENT, i.e. is there a way to avoid loading the entire mesh into the RAM? It would be great if there is a way to specify the boundary conditions, solution methods, etc. "externally" without opening the FLUENT GUI and loading the mesh.

In the end I only need a FLUENT case file which I can solve with a cluster - where required size of memory won't be an issue.

Thanks in advance!

Edmond

LuckyTran February 14, 2018 17:28

There's no way to avoid loading the entire mesh in Fluent if you want to open the case file.

But you can always create a journal file (or less elegantly record a macro) to change all the boundary conditions and settings and this journal file can be run on the cluster. The difficulty here is that you need to know how to make this journal file (learn how to use the TUI instead of the GUI).

Of course being unable to load the mesh in your local machine does not help you to experiment. I would recommend to generate an extremely coarse mesh that you can load on your local machine and play with the TUI commands. You could even use this coarse mesh to do all the settings and then replace the mesh with the final mesh that you want to run (but be ready for lots of things going wrong if you don't know what you're doing).

An even more painful method is to open the .cas file with a text editor and manually change the text there. The .cas file is just a really big ascii text file (unless you save it as a binary) which contains all these settings.

In the end, you need some skills/experience dealing with Fluent.

edmondlam February 14, 2018 17:54

Thanks for your reply!! I think I will try with your methods now :)


All times are GMT -4. The time now is 12:33.