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

parallel processing time+reconstruct time<serial processing?!

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

Like Tree6Likes
  • 1 Post By immortality
  • 1 Post By ngj
  • 1 Post By ngj
  • 3 Post By kwardle

Reply
 
LinkBack Thread Tools Display Modes
Old   August 9, 2013, 15:35
Default parallel processing time+reconstruct time<serial processing?!
  #1
Senior Member
 
immortality's Avatar
 
Ehsan
Join Date: Oct 2012
Location: Iran
Posts: 2,186
Rep Power: 16
immortality is on a distinguished road
Hi
have anyone tested that reconstructPar time does vanishes parallel processing benefits over serial run time on a local PC(with several threads) or not(parallel processing is better despite of reconstructPar time consumption)?
has done anyone any tests or have any idea?
cfdonline2mohsen likes this.
__________________
Injustice Anywhere is a Threat for Justice Everywhere.Martin Luther King.
To Be or Not To Be,Thats the Question!
The Only Stupid Question Is the One that Goes Unasked.
immortality is offline   Reply With Quote

Old   August 10, 2013, 02:32
Default
  #2
ngj
Senior Member
 
Niels Gjoel Jacobsen
Join Date: Mar 2009
Location: Deltares, Delft, The Netherlands
Posts: 1,607
Rep Power: 25
ngj will become famous soon enoughngj will become famous soon enough
Hi Ehsan,

The key point probably is that you do not need to reconstruct your results. paraFoam/paraView works on your decomposed case and your post-processing merely need to consider the fact that data is decomposed, once you work with them.

Previously, I also experienced that the reconstruction part took a lot of time, so I have not been doing that for a long time.

Kind regards

Niels
immortality likes this.
__________________
Please note that I do not use the Friend-feature, so do not be offended, if I do not accept a request.
ngj is offline   Reply With Quote

Old   August 10, 2013, 07:44
Default
  #3
Senior Member
 
immortality's Avatar
 
Ehsan
Join Date: Oct 2012
Location: Iran
Posts: 2,186
Rep Power: 16
immortality is on a distinguished road
Hi Niels
really there is no need to reconstructPar? I didn't know that! I'll test it,thank you so much.
__________________
Injustice Anywhere is a Threat for Justice Everywhere.Martin Luther King.
To Be or Not To Be,Thats the Question!
The Only Stupid Question Is the One that Goes Unasked.
immortality is offline   Reply With Quote

Old   August 10, 2013, 08:05
Default
  #4
Senior Member
 
cfdonline2mohsen's Avatar
 
Mohsen KiaMansouri
Join Date: Jan 2010
Location: CFD Lab
Posts: 117
Rep Power: 7
cfdonline2mohsen is on a distinguished road
Dear Ehsan
In most of the cases you do not need to reconstructPar all the time steps but only the results for the latestTime by using:
Code:
reconstructPar -latestTime
which does not take a lot of time.
and reconstructPar is definately required for postprocessing the results using Tecplot.

Dear Niels
Quote:
The key point probably is that you do not need to reconstruct your results. paraFoam/paraView works on your decomposed case
I remember that in the previous versions of Paraview when I did not reconstructPar my results, it did not show my whole geometry but only my decomposed geometries seperately! (in the processor0 , processor1 ... folders)
does it show your whole geometry or only your decomposed geometry separately?
__________________
“If you have an apple and I have an apple and we exchange these apples then you and I will still each have one apple. But if you have an idea and I have an idea and we exchange these ideas, then each of us will have two ideas.”
cfdonline2mohsen is offline   Reply With Quote

Old   August 10, 2013, 08:49
Default
  #5
ngj
Senior Member
 
Niels Gjoel Jacobsen
Join Date: Mar 2009
Location: Deltares, Delft, The Netherlands
Posts: 1,607
Rep Power: 25
ngj will become famous soon enoughngj will become famous soon enough
Hi Kia,

In your ParaView (my version: 3.98.1), there is a drop-down menu just below the "Refresh" bottom. I you change it from "Reconstructed Case" to "Decomposed Case", ParaView gathers the needed decomposed data by itself. It is no longer needed to manually load all N processor directories.

Kind regards

Niels
cfdonline2mohsen likes this.
__________________
Please note that I do not use the Friend-feature, so do not be offended, if I do not accept a request.
ngj is offline   Reply With Quote

Old   August 10, 2013, 12:40
Default
  #6
Senior Member
 
immortality's Avatar
 
Ehsan
Join Date: Oct 2012
Location: Iran
Posts: 2,186
Rep Power: 16
immortality is on a distinguished road
thanks Mohsen and Niels
I use 3.12.0 and think there is not such an option.and my case is unsteady then I need all time steps.never mind it's not a big issue!
__________________
Injustice Anywhere is a Threat for Justice Everywhere.Martin Luther King.
To Be or Not To Be,Thats the Question!
The Only Stupid Question Is the One that Goes Unasked.
immortality is offline   Reply With Quote

Old   August 12, 2013, 11:58
Default
  #7
Senior Member
 
Kent Wardle
Join Date: Mar 2009
Location: Illinois, USA
Posts: 195
Rep Power: 10
kwardle is on a distinguished road
How are you starting ParaView? If via paraFoam, then use the -builtin flag to use the parallel-aware reader in ParaView. If you are starting paraview on your own from the commandline or otherwise, if you make the controlDict softlink extension ".foam" (instead of .OpenFOAM) it will automatically use the built-in reader which will have the option to select reconstructed or decompose data sets. The "Refresh" button is not in 3.12 but is found in more recent versions (I have 4.0.1), but you could also record a python macro which switches from decomposed to reconstructed and back and this will refresh the times also.
kwardle is offline   Reply With Quote

Old   August 12, 2013, 12:39
Default
  #8
Senior Member
 
immortality's Avatar
 
Ehsan
Join Date: Oct 2012
Location: Iran
Posts: 2,186
Rep Power: 16
immortality is on a distinguished road
Hi Kent
that was very useful. -builtin option was excellent! thanks.
@Mohsen:
Hi Mohsen
Quote:
and reconstructPar is definately required for postprocessing the results using Tecplot.
for what type of postProcessing we may need tecPlot on top of paraView?and whats equivalent to tecPlot in ubuntu?
__________________
Injustice Anywhere is a Threat for Justice Everywhere.Martin Luther King.
To Be or Not To Be,Thats the Question!
The Only Stupid Question Is the One that Goes Unasked.
immortality is offline   Reply With Quote

Old   August 13, 2013, 09:15
Default
  #9
Senior Member
 
cfdonline2mohsen's Avatar
 
Mohsen KiaMansouri
Join Date: Jan 2010
Location: CFD Lab
Posts: 117
Rep Power: 7
cfdonline2mohsen is on a distinguished road
Quote:
for what type of postProcessing we may need tecPlot on top of paraView?
Well, actually I have used Tecplot from my B.Sc to PhD for more than 10 years and it's really hard to move to paraView completely and put all the experiences in using Tecplot into the trash can!
we have a proverb for that in Persian that says "old habits die hard"
I am really glad that Tecplot started to support OpenFoam directly without requiring any tools to convert the OF data.

Quote:
whats equivalent to tecPlot in ubuntu?
Actually there is tecplot-360 for ubuntu.find more information in here:
http://www.tecplot.com/products/tecplot-360/ in requirements section.
and of course it is not free like ParaView!
__________________
“If you have an apple and I have an apple and we exchange these apples then you and I will still each have one apple. But if you have an idea and I have an idea and we exchange these ideas, then each of us will have two ideas.”
cfdonline2mohsen 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
AMI speed performance danny123 OpenFOAM 19 October 24, 2012 07:44
calculation diverge after continue to run zhajingjing OpenFOAM 0 April 28, 2010 04:35
Differences between serial and parallel runs carsten OpenFOAM Bugs 11 September 12, 2008 11:16
IcoFoam parallel woes msrinath80 OpenFOAM Running, Solving & CFD 9 July 22, 2007 02:58
AMG versus ICCG msrinath80 OpenFOAM Running, Solving & CFD 2 November 7, 2006 16:15


All times are GMT -4. The time now is 23:01.