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

Parallelization of a CFD code

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

Reply
 
LinkBack Thread Tools Display Modes
Old   March 16, 2009, 07:29
Default Parallelization of a CFD code
  #1
New Member
 
ilias
Join Date: Mar 2009
Posts: 4
Rep Power: 8
mixanologos is on a distinguished road
Hallo CFD experts!

I'm trying to paralize a cfd code at algebraic system level. I have read a lot, and i concluded that Domain Decomposition and Alternating Schwarz method is a good choice.
I tried to use Alternating Schwarz method (with overlaping domains) as preconditioner for GMRES. I'm solving (in parallel) the preconditioned system in each iteration with GMRES again, but with low accuracy.
Nevertheless, results are'n got... I have no reduce in time. And the iterations of the outer GMRES doesn't reduce much.
I'm testing the method in a very simple problem: Poisson equation in a close cube.
Is the method i use the right one, or I'm doing something wrong?
What methods are being used mostly in commercial codes?

Many thanks

Last edited by mixanologos; March 16, 2009 at 09:24.
mixanologos is offline   Reply With Quote

Old   March 16, 2009, 11:36
Default
  #2
New Member
 
Join Date: Mar 2009
Posts: 4
Rep Power: 8
Marcus is on a distinguished road
A lot of questions come to mind... domain decomposition is certainly the way to go, but

What hardware and languages are you using? Overhead associated with parallelization and the granularity of your problem have a huge effect on your speedup. e.g., if you're using a distributed memory setup, and you're writing a very fine-grained code (lots of data exchange in a given iteration), you'll often get code that was slower than the original serial code.
Marcus is offline   Reply With Quote

Old   March 17, 2009, 02:50
Default
  #3
New Member
 
ilias
Join Date: Mar 2009
Posts: 4
Rep Power: 8
mixanologos is on a distinguished road
Quote:
Originally Posted by Marcus View Post
A lot of questions come to mind... domain decomposition is certainly the way to go, but

What hardware and languages are you using? Overhead associated with parallelization and the granularity of your problem have a huge effect on your speedup. e.g., if you're using a distributed memory setup, and you're writing a very fine-grained code (lots of data exchange in a given iteration), you'll often get code that was slower than the original serial code.
Well, i haven't yet try to write it in parallel
I haven't do it because i count the total time that precondition systems need to run, i divide it with the number of processes (=number of subdomains) and i calculate the time of parallel code (minimum time because no data exchange time included). But this minimum time is still above the normal time of serial GMRES/ILU solver

That's why i'm wondering if Domain Decomposition and additive Schwarz preconditioner is a way to go.

Thanks for the reply!
mixanologos is offline   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
Parallelization of a new solver jaswi OpenFOAM Running, Solving & CFD 19 August 27, 2012 02:45
A question on parallelization in OF ivan_cozza OpenFOAM 3 September 10, 2008 06:47
Parallelization Levi Main CFD Forum 1 May 25, 2003 02:03
About the parallelization ptyue Main CFD Forum 8 January 27, 2003 00:29
Parallelization efficiency.. karthik Main CFD Forum 1 January 4, 2000 12:20


All times are GMT -4. The time now is 14:21.