CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > ANSYS > CFX

Nodes per Core restriction?

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

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   February 7, 2012, 18:01
Default Nodes per Core restriction?
  #1
Member
 
Max
Join Date: Apr 2011
Posts: 47
Rep Power: 14
max3.2 is on a distinguished road
Is there a nodes per core restriction in CFX (educational)?
I have a 11mil cell mesh. If I run it on serial I get an error (something like to many integer values see below)
If I run it on more cores, it works.
I dont get that. Any ideas?

ERROR:

Quote:
ERROR #001100279 has occurred in subroutine ErrAction. |
| Message: |
| The total number of words needed for the Real stack exceeds the m- |
| aximum supported by the flow solver: 2,147,483,647. The solver w- |
| ill continue with the stack size clipped to that value.
max3.2 is offline   Reply With Quote

Old   February 7, 2012, 22:58
Default
  #2
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,665
Rep Power: 143
ghorrocks is just really niceghorrocks is just really niceghorrocks is just really niceghorrocks is just really nice
Wow, that is one big simulation - 2 billion for the real stack. Then clearly having a single simulation of this size is not possible, but if you split it up into some parallel runs it is.

I was not aware that this restriction existed.
ghorrocks is offline   Reply With Quote

Old   February 8, 2012, 10:03
Default
  #3
Member
 
Max
Join Date: Apr 2011
Posts: 47
Rep Power: 14
max3.2 is on a distinguished road
I honestly don"t know whats blowing up the real stack.
Well, the cell count is high, its a duct with a wedge, hexa meshed for y+ around 1. I will be probably using it for running high volume concentration particle tracking, comparing it to experiments and DEM.
But thats just the first mesh validation for water with SST. 7Eq.

I was just wondering if that's a known issue. But since it runs parallel, it must be some limit on the single core runs.

Thanks
Max
max3.2 is offline   Reply With Quote

Old   February 8, 2012, 11:03
Default
  #4
Senior Member
 
Join Date: Apr 2009
Posts: 531
Rep Power: 20
stumpy is on a distinguished road
There is a hard limit of that many words per stack per partition. It corresponds to about 16GB I think. Running in parallel reduces the memory required per partition, so you can drop below this limit.
stumpy is offline   Reply With Quote

Old   June 8, 2013, 06:46
Default
  #5
New Member
 
Vishal
Join Date: Feb 2011
Location: Bhopal, M.P. (India)
Posts: 18
Rep Power: 15
vishal.indomitable is on a distinguished road
Send a message via Yahoo to vishal.indomitable
No there is no restrictions for nodes per core in CFX (educational) as I know. The same problem I am also getting. The simulation will not show any error if it is run on parallel machines.
vishal.indomitable is offline   Reply With Quote

Old   June 10, 2013, 23:26
Default
  #6
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,665
Rep Power: 143
ghorrocks is just really niceghorrocks is just really niceghorrocks is just really niceghorrocks is just really nice
Stumpy's post answers the question - you need to run enough partitions to get all partitions below this stack limit. This stack limit will apply for the educational version of the software as well.
ghorrocks is offline   Reply With Quote

Reply

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
solving a conduction problem in FLUENT using UDF Avin2407 Fluent UDF and Scheme Programming 1 March 13, 2015 03:02
Superlinear speedup in OpenFOAM 13 msrinath80 OpenFOAM Running, Solving & CFD 18 March 3, 2015 06:36
Performance Improvements by core locking RobertB STAR-CCM+ 7 October 22, 2010 08:59
meshing F1 front wing Steve FLUENT 0 April 17, 2003 13:37
CFX4.3 -build analysis form Chie Min CFX 5 July 13, 2001 00:19


All times are GMT -4. The time now is 07:25.