CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Phoenics (https://www.cfd-online.com/Forums/phoenics/)
-   -   Running Phoenics in Clusters. (https://www.cfd-online.com/Forums/phoenics/51139-running-phoenics-clusters.html)

Javier Larrondo January 8, 2001 16:11

Running Phoenics in Clusters.
 
Recently I read about Cluster technology, and as a user of phoenics, I'm wondering if it is posible to run phoenics in a cluster device?. I realize that it's not the same as parallel computing, but it seem to be more cheaper than parallel computing, in terms of implementation.

I need some coments, because in our company we want to improve our CFD devices.

Thanks for all.

Javier Larrondo CADE-IDEPE Chile

Note: When are you going to put the equations(mathematical background) of the models in POLIS? There is a lot of people who don't have access to the extensive references quoted.

Peter Spalding January 10, 2001 08:29

Re: Running Phoenics in Clusters.
 
PHOENICS is supported on both parallel-processing systems and pc clusters. The pc clusters (ie multiple pc's linked by fast Ethernet cabling) are supported on both Windows-based and LINUX-based configurations.

CHAM itself has clusters of both types, and there are several customer sites with larger implementations.

The only comprehensive document relating to the equations is TR99 available vie the cfdShop (see: http://www.simuserve.com/cfd-shop/tr99sab.htm). Although this is now quite old, and relates to what is now the Shareware version, it is nevertheless a good reference source.

Specific models, such as the various turbulence models, the scalar equation model, and so forth, ARE described in POLIS. If you can be a little more specific, we can no doubt help further.


Javier Larrondo January 10, 2001 13:32

Re: more about equations...
 
In the main forum I have found a large discuss of this issue and thanks for be more especific.

About the equations, I have found some in POLIS. But I'm intersted in the IMMERSOL model, because we have some doubts in the implementation with participating media. We have been testing our model using diferent values for RADIA from 0.15 to 1. Is this the value that I have to use to model a participating media? And what does it mean the "link between tem1 and t3" value? I have been using 1.0e+05, but I used it only because I found it in an example.

Thanks for answering.



All times are GMT -4. The time now is 00:23.