|
[Sponsors] |
April 1, 2015, 11:38 |
Parallel Simulation Data Write Failure
|
#1 |
New Member
N/A
Join Date: Jul 2010
Posts: 29
Rep Power: 16 |
After decompose a geometry with "scotch", using "decomposePar", running the simulation with "mpiexec -n 6 interFoam -parallel", then using "reconstructPar" I get the following error:
Code:
wrong token type - expected Scalar, found on line 29397 the punctuation token ',' file: /home/jesse/NotSynced/VOFCriticality/fill/processor5/74/omega at line 29397. From function operator>>(Istream&, Scalar&) in file lnInclude/Scalar.C at line 93. FOAM exiting I am using 2.3.1; however, this occurs with 2.3.0 as well. My decomposeParDict file is: Code:
FoamFile { version 2.0; format ascii; class dictionary; location "system"; object decomposeParDict; } // * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * // numberOfSubdomains 6; //number of cpus you'll use method scotch; //method of decomposition simpleCoeffs { n ( 2 2 2 ); delta 0.001; } hierarchicalCoeffs { n ( 1 1 1 ); delta 0.001; order xyz; } scotchCoeffs { processorWeights ( 1 1 1 1 1 1 ); } manualCoeffs { dataFile ""; } distributed no; roots ( ); |
|
April 3, 2015, 13:09 |
|
#2 |
New Member
N/A
Join Date: Jul 2010
Posts: 29
Rep Power: 16 |
Has anyone run into this issue at all? I'm wondering if it might be necessary that I use a server with ECC RAM or something, or if there is another trick to protect the writing.
I attempted to use the compression open, but it does not alleviate my issues. Any help would be appreciated. Thanks. |
|
April 16, 2015, 16:24 |
|
#3 |
New Member
N/A
Join Date: Jul 2010
Posts: 29
Rep Power: 16 |
In the event anyone has this kind of problem, I solved it by recompiling MPI and OpenFOAM. It seems likes something funny happened in the libraries. Luckily, it does not appear to be a hardware issue.
|
|
Tags |
decomposepar, error, reconstructpar |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Transfer of data from one simulation to another. | sreekargomatam | FLUENT | 2 | January 1, 2015 12:23 |
Understanding CFD Tracer Study Simulation Data | Tracer | FLUENT | 2 | October 15, 2014 14:20 |
Averaging in parallel - Simulation slowed down | fs82 | OpenFOAM Running, Solving & CFD | 0 | March 10, 2010 05:16 |
parallel mode failure in 3ddp but not in 2ddp | ak6g08 | FLUENT | 1 | September 22, 2009 07:56 |
parallel mode failure in 3ddp but not in 2ddp | ak6g08 | Fluent UDF and Scheme Programming | 0 | September 22, 2009 07:16 |