How to run ESTEL in parallel - Revision history

Jprenaud at 09:37, 7 September 2007

2007-09-07T09:37:10Z

2007-08-23T16:57:29Z

2007-08-23T16:55:57Z

2007-08-23T16:55:48Z

2007-08-22T14:29:22Z

Run a parallel job on several machines

2007-08-22T14:26:04Z

Note about ESTEL-3D

2007-08-22T14:24:08Z

Note about ESTEL-3D

2007-08-22T14:23:56Z

Note about estel2d

2007-08-22T14:23:27Z

Note about ESTEL-3D

2007-08-22T14:21:02Z

Note about parallel output

@@ Line 1: / Line 1: @@
 [[Category:Estel]]
-This article describes how to run parallel jobs in [[Estel | '''ESTEL''']] on "simple" networks of workstations.
+'''This article describes how to run parallel jobs in [[Estel | '''ESTEL''']] on "simple" networks of workstations.'''
 Note that the methodology differs slightly for real high performance facilities such as [http://www.acrc.bris.ac.uk Blue Crystal] or other [http://en.wikipedia.org/wiki/Beowulf_cluster Beowulf clusters]. Therefore, there is a [[Run ESTEL in parallel on a cluster | dedicated article for clusters]].

@@ Line 78: / Line 78: @@
 = Run a parallel job on several machines =
-If MPI has been setup properly, running '''[[ESTEL]]]''' on several machines is not very complicated.
+If MPI has been setup properly, running '''[[ESTEL]]''' on several machines is not very complicated.
 Start a <code>mpd</code> ring requesting the right number of processors:

@@ Line 122: / Line 122: @@
 For estel3d, you will also have:
 * series of <code>name-of-3d-results-fileN-M</code>
-* series of <code>name-of-mesh-results-fileN-M</code>
+* series of <code>name-of-mesh-results-fileN-M</code>   '''are empty at the moment, BUG'''
 This is because '''ESTEL-3D''' does not recompose the solution on one mesh due to limitation in the binary Tecplot library. You will need to load all these files at once in Tecplot (option "Load Multiple Files") to see the full solution (or the full mesh). Note that this creates interpolation artefacts for P0 variables. This will be "fixed" in the next version of ESTEL which will use a native format instead of the Tecplot format and therefore will be able to do domain recomposition. In Tecplot, be careful to change the default filter as the numbering of the files hides the typical <code>.plt</code> or <code>.dat</code> extension.

@@ Line 126: / Line 126: @@
 This is because '''ESTEL-3D''' does not recompose the solution on one mesh due to limitation in the binary Tecplot library. You will need to load all these files at once in Tecplot (option "Load Multiple Files") to see the full solution (or the full mesh). Note that this creates interpolation artefacts for P0 variables. This will be "fixed" in the next version of ESTEL which will use a native format instead of the Tecplot format and therefore will be able to do domain recomposition. In Tecplot, be careful to change the default filter as the numbering of the files hides the typical <code>.plt</code> or <code>.dat</code> extension.
-Also, as there is no domain recomposition, there is no <code>gretel.log<code> for '''ESTEL-3D'''.
+Also, as there is no domain recomposition, there is no <code>gretel.log</code> for '''ESTEL-3D'''.
 = Note about estel2d =

@@ Line 129: / Line 129: @@
 = Note about estel2d =
-* no "validation" possible, will crash with no warning. '''probably a bug'''.
+'''To finish, Fabien??'''
 * Problem with particle tracking
 ** second keyword required
 ** dictionary to be changed

← Older revision	Revision as of 16:57, 23 August 2007
(No difference)

← Older revision	Revision as of 16:55, 23 August 2007
(No difference)

@@ Line 107: / Line 107: @@
 = Note about parallel output =
 When you run '''ESTEL-2D''' or '''ESTEL-3D''' in parallel, you will obtain some new files in the directory where the simulation was run:
-* partel.log contains the log of the domain decomposition step at the very beginning of the simulation.
+* <code>partel.log</code> contains the log of the domain decomposition step at the very beginning of the simulation.
-* gretel.log contains the log of the domain recomposition step at the end of the simulation.
+* <code>gretel.log</code> contains the log of the domain recomposition step at the end of the simulation. Not for '''ESTEL-3D''', see note below.
-* mpirun.txt contains the hosts that have been used by MPI
+* <code>mpirun.txt</code> contains the hosts that have been used by MPI
-* series of pe00N-00M.log files. Each of these files is the listing output of ESTEL on the compute nodes. If
+* series of <code>peN-M.log</code> files. Each of these files is the listing output of ESTEL on the compute nodes, N is the total number of processors required and M the number of the host the log comes from. Note that numbering starts at zero and the log for the master node (M=0) is not kept as you see it on the screen. Therefore for 4 processors, there would be 3 files named:
 There are extra files for '''ESTEL-3D''', see below.