CFD Online Logo CFD Online URL
Home > Forums > Software User Forums > ANSYS > FLUENT

Parallel Processing

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

LinkBack Thread Tools Search this Thread Display Modes
Old   August 29, 2006, 07:52
Default Parallel Processing
Drew Abbott
Posts: n/a

I am using fluent to conduct my thesis on a UNIX system. I am unable to run the software on multiple processors in parallel. The File/Run task bar will not appear in fluent so I cannot use the select solver panel. I have created a run file so that I can type: fluent 3d -g ( MediumTest.out&

rc MediumTest.cas


solve/dti 500 50

wd MediumTestInter.dat

wc MediumTestInter.cas

solve/dti 500 50

wd MediumTestFinal.dat

wc MediumTestFinal.cas

I was wondering whether anyone could help me figure out where I need to instert the code to use parallel processors. Thanks for your time. If any further information is required please don't hesitate to ask.

  Reply With Quote

Old   August 29, 2006, 08:04
Default Re: Parallel Processing
Posts: n/a
You have to specify number of processors x

fluent 3d "tx "g

I have 2 processors so I write fluent 3d "t2 "g
  Reply With Quote

Old   August 29, 2006, 12:35
Default Re: Parallel Processing
Posts: n/a
That works for multiple processors on a single machine. For multiple machines (i.e. a parallel network) you have to specify which machines to run on. Read through the users guide, chapter 1.1 on starting Fluent and chapter 32.2 on running in Parallel for more info. It depends on the setup for your system. Sometimes you have to define the MPI (message passing interface?).

An example run line could be:

fluent 3d -tx -cnf=sample.hosts -g < inputfile >& outputfile &

where x (in -tx) is the number of compute nodes you want to use and sample.hosts is a text file that lists the computers you want to connect to.

If that doesn't work, you have to define a communicator:

fluent 3d -pcomm -tx -cnf=sample.hosts -g < inputfile >& outputfile &

where -pcomm is the communicator (-pnmpi, -ptmpi, -psmpi... there's a whole list of them in UG Chapter 32.2.1). Some communicators don't allow you to "spawn nodes" within Fluent, so you ahve to use a hosts file, but some communicators do allow you to, so you could potentially drop the -tx -cnf=sample.hosts from the runline and manually spawn the nodes within Fluent.

Hope this helps, and good luck, Jason
  Reply With Quote


Thread Tools Search this Thread
Search this Thread:

Advanced Search
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 Off
Pingbacks are On
Refbacks are On

Similar Threads
Thread Thread Starter Forum Replies Last Post
Parallel processing of OpenFOAM cases on multicore processor??? g.akbari OpenFOAM Running, Solving & CFD 31 November 1, 2017 09:25
multiphase solver - parallel processing - GAMG thibault_pringuey OpenFOAM Programming & Development 2 August 27, 2013 22:03
HP MPI warning...Distributed parallel processing Peter CFX 10 May 14, 2011 06:17
bubbly flow and parallel processing mvee FLUENT 0 September 12, 2007 05:08
About parallel processing in Linux tuks CFX 10 August 8, 2005 08:22

All times are GMT -4. The time now is 15:06.