I am facing a problem during the simulation of a file mesh 6.7 millions , i simulated a file files smaller than it and didn't encounter this problem. I wanted to know what is the procedure?
Is there a way to overcome this problem?
My Computer: i7 4th generation with 8 gb Ram.
Thank you all.
Hi Djamel,
Parallel processing only takes away the computation from single core to multiple cores. As the solver is working with multiple iterations at the same time, it will take more memory allocation to work with different nodes. Remember that the only setback to simulations is the computing resources. I will recommend you to optimize your mesh to decrease the elements and check your solver and iteration settings if it is a transient study.
Dear Hassan Adnan Malik ,
Thank you for your answer.
Currently i'm working on the Grid adaption for different sizes: 1 million, 3 million, 6 million and 7 million. It's necessary to check mesh before begin the results.
The well-known solution is to change my computer with another highly efficient . But, I would like to change the processing method from parallel to serial, is this correct step or not and how is it done in Fluent 6.2.16 ?
Regards.
Hi Djamel,
It is important for engineers to optimize the resources available to them and my answer was based upon the same. Technically speaking, if your processes max out your memory in parallel processing, it depends on what stage the solver has reached. It may help to run a serial solver which would take more time but will optimize the memory.
The other thing you can do is decrease the number of iterations or break down the process, from let's say a First-Order Upwind at start to a Second-Order Upwind after let's say 40% iterations. Good luck.
- Badges
- Science topic
- Similar topics
- Filing