CFD Online Logo CFD Online URL
Home > Forums > CFX


Register Blogs Members List Search Today's Posts Mark Forums Read

LinkBack Thread Tools Display Modes
Old   January 16, 2009, 15:03
Default cfx5gtmconv
Posts: n/a

Does anyone know if it's possible to run the cfx5gtmconv.exe in parallel (on a multiple-processors linux operating system)?

Specifically, I'm trying to convert a .cdb file into a .def file using the cfx5gtmconv on a supercomputer (access to memory isn't an issue). However, the size of the .cdb file is >9gb (40million nodes) and requires a substantial amount of RAM from each node to convert the data.

Hence, as cfx5gtmconv is by default run on only one node, in my case with 8 processors, so there is a limit of 8gb of memory; CFX is simply using more memory than the node has available. Yet, if i can tell cfx5gtmconv to run in parallel and use multiple nodes, i can reduce the per node memory requirements and it wont be limited to 8gb. any suggestions on how to do this?

thanks for the help in advance
  Reply With Quote

Old   January 23, 2009, 18:59
Default Re: cfx5gtmconv
Posts: n/a
It doesn't run in parallel.

Can you split the mesh before saving it and convert each portion. You could then load each portion of the mesh into pre to reassemble it, though Pre is also serial, it may be irrelevant.

  Reply With Quote


Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On

All times are GMT -4. The time now is 23:28.