CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Main CFD Forum

On Inavier

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

Reply
 
LinkBack Thread Tools Display Modes
Old   April 28, 2008, 09:16
Default On Inavier
  #1
kenneth
Guest
 
Posts: n/a
Is there someone playing Inavier?I tried to simulate fluid flow in a complex geometry by using the code, always have difficulty in obtaining convergence solution. I would appreciate your hints.

Regards,

Kenneth
  Reply With Quote

Old   April 29, 2008, 18:02
Default Re: On Inavier
  #2
Arjun
Guest
 
Posts: n/a
Hi, i have to update the solver, there was a small bug in AMG solver so I have replaced the linear solver with new version. Which is quite fast and better convergent. I have added quite a few things specially treatment of unmatched grids and grid rotation. This part have to be checked with turbulence models so it is taking little bit of time. In first week of june I will upload the newer version. Just little wait.
  Reply With Quote

Old   April 29, 2008, 19:31
Default Re: On Inavier
  #3
kenneth
Guest
 
Posts: n/a
That would be great! thanks. Have you considered to parallelize the code?

  Reply With Quote

Old   April 30, 2008, 21:31
Default Re: On Inavier
  #4
Arjun
Guest
 
Posts: n/a
About the parallelization, its almost time I started working on this issue. Parallelization has to come sooner or later, the main reason is in todays world usually the models are so large that they could not be handled by single processor.

The reasons why there is no parallelization so far:

1. I wished to write the code for students who could use it for their work. So wished that things be such that they do not need anything from outside. That is it should have collection of C++ file that you could compile anywhere. And the current code is easily readable (because each thing has a manager class , just go to it and read the relvant functions).

2. I am not smart enough. I do not have any idea how to implement parallel processing. OpenMP based shared memory parallization looks tempting but it puts the bar on the sizes of meshes we could use. And for MPI based parallezation, i do not have resource to even learn to write small programs. (But I will do something about this issue, just need some time).

The current version of code looks promising to it would be worth working further with it. But no matter how the code develops, the single processor version i wish to keep aside for students. Just for whom I put so much of efforts. (though it was mainly my selfish need for learning about what was going on behind the black box Fluent software).
  Reply With Quote

Old   April 30, 2008, 23:31
Default Re: On Inavier
  #5
kenneth
Guest
 
Posts: n/a
Thanks for your help.
  Reply With Quote

Old   May 7, 2008, 01:48
Default Re: On Inavier
  #6
javaid
Guest
 
Posts: n/a
The code looks intersting can i obtain source code.
  Reply With Quote

Old   May 7, 2008, 17:16
Default Re: On Inavier
  #7
Arjun
Guest
 
Posts: n/a
download it.
  Reply With Quote

Old   May 7, 2008, 17:17
Default Re: On Inavier
  #8
Arjun
Guest
 
Posts: n/a
I have updated the code. Download the new version.
  Reply With Quote

Reply

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


Similar Threads
Thread Thread Starter Forum Replies Last Post
Regarding iNavier Code Rajan Main CFD Forum 9 April 18, 2011 19:05
colver code in c or c++ phd_student Main CFD Forum 26 April 12, 2010 18:07
iNavier problem vanchisen Main CFD Forum 1 July 3, 2009 13:34
iNavier Arjun Main CFD Forum 2 May 8, 2008 06:13
iNavier - Download problems Arjun Main CFD Forum 0 June 1, 2007 08:50


All times are GMT -4. The time now is 12:24.