
[Sponsors] 
July 25, 2006, 20:43 
Hi,
I'm trying the same cav

#1 
Senior Member
Srinath Madhavan (a.k.a pUl)
Join Date: Mar 2009
Location: Edmonton, AB, Canada
Posts: 703
Rep Power: 17 
Hi,
I'm trying the same cavity case as explained in the user guide. But I've increased the mesh to 10000 cells and decreased the time step from 0.005 to 0.001. The case runs happily when in serial mode. When I decompose the domain and run in parallel, it seems to run fine for a while after which, suddenly a solution singluarity appears. The BCs and problem definition are OK as the case runs properly in serial. Any suggestions (please look at the time between 0.013 and 0.015 seconds). Thanks for your help. PS: At Time = 0.024, I hit Ctrl + C. Also, when running the case in parallel again (from scratch) sometimes it works fine up to 0.2 or 0.3 seconds, sometimes less, sometimes it just hangs there with no error messages as if waiting for something. Has anybody experienced such random behavior. I'm using OpenFOAM on a 64 bit IBM p360 (4CPU) machine running Suse Linux 10.1 and lam (compiled from sources that came with OpenFOAM). madhavan@varese:~/OpenFOAM/madhavan1.3/run/tutorials/icoFoam> ls Allclean cavity cavityGrade cavity_large_mesh_backup machines Allrun cavityClipped cavity_large_mesh elbow resetFixedWallsScr madhavan@varese:~/OpenFOAM/madhavan1.3/run/tutorials/icoFoam> decomposePar . cavity_large_mesh /**\  =========    \ / F ield  OpenFOAM: The Open Source CFD Toolbox   \ / O peration  Version: 1.3   \ / A nd  Web: http://www.openfoam.org   \/ M anipulation   \**/ Exec : decomposePar . cavity_large_mesh Date : Jul 25 2006 Time : 19:20:21 Host : varese PID : 32389 Root : /home/madhavan/OpenFOAM/madhavan1.3/run/tutorials/icoFoam Case : cavity_large_mesh Nprocs : 1 Create time Time = 0 Create mesh Calculating distribution of cells Selecting decompositionMethod simple Finished decomposition in 0.45 s Calculating original mesh data Distributing cells to processors Distributing faces to processors Calculating processor boundary addressing Distributing points to processors Constructing processor meshes Processor 0 Number of cells = 2500 Number of faces shared with processor 1 = 50 Number of faces shared with processor 2 = 50 Number of boundary faces = 5100 Processor 1 Number of cells = 2500 Number of faces shared with processor 0 = 50 Number of faces shared with processor 3 = 50 Number of boundary faces = 5100 Processor 2 Number of cells = 2500 Number of faces shared with processor 0 = 50 Number of faces shared with processor 3 = 50 Number of boundary faces = 5100 Processor 3 Number of cells = 2500 Number of faces shared with processor 1 = 50 Number of faces shared with processor 2 = 50 Number of boundary faces = 5100 Processor 0: field transfer Processor 1: field transfer Processor 2: field transfer Processor 3: field transfer End. madhavan@varese:~/OpenFOAM/madhavan1.3/run/tutorials/icoFoam> mpirun np 4 icoFoam . cavity_large_mesh parallel /**\  =========    \ / F ield  OpenFOAM: The Open Source CFD Toolbox   \ / O peration  Version: 1.3   \ / A nd  Web: http://www.openfoam.org   \/ M anipulation   \**/ /**\  =========    \ / F ield  OpenFOAM: The Open Source CFD Toolbox   \ / O peration  Version: 1.3   \ / A nd  Web: http://www.openfoam.org   \/ M anipulation   \**/ /**\  =========    \ / F ield  OpenFOAM: The Open Source CFD Toolbox   \ / O peration  Version: 1.3   \ / A nd  Web: http://www.openfoam.org   \/ M anipulation   \**/ /**\  =========    \ / F ield  OpenFOAM: The Open Source CFD Toolbox   \ / O peration  Version: 1.3   \ / A nd  Web: http://www.openfoam.org   \/ M anipulation   \**/ Exec : icoFoam . cavity_large_mesh parallel Exec : icoFoam . cavity_large_mesh parallel Exec : icoFoam . cavity_large_mesh parallel Exec : icoFoam . cavity_large_mesh parallel [2] Date : Jul 25 2006 [2] Time : 19:21:14 [0] Date : Jul 25 2006 [0] Time : 19:21:14 [0] Host : varese [2] Host : varese [2] PID : 32393 [3] Date : Jul 25 2006 [3] Time : 19:21:14 [0] PID : 32391 [3] Host : varese [3] PID : 32394 [1] Date : Jul 25 2006 [1] Time : 19:21:14 [1] Host : varese [1] PID : 32392 [1] Root : /home/madhavan/OpenFOAM/madhavan1.3/run/tutorials/icoFoam [1] Case : cavity_large_mesh [1] Nprocs : 4 [2] Root : /home/madhavan/OpenFOAM/madhavan1.3/run/tutorials/icoFoam [2] Case : cavity_large_mesh [2] Nprocs : 4 [3] Root : /home/madhavan/OpenFOAM/madhavan1.3/run/tutorials/icoFoam [3] Case : cavity_large_mesh [0] Root : /home/madhavan/OpenFOAM/madhavan1.3/run/tutorials/icoFoam [0] Case : cavity_large_mesh [0] Nprocs : 4 [0] Slaves : [0] 3 [3] Nprocs : 4 [0] ( [0] varese.32392 [0] varese.32393 [0] varese.32394 [0] ) [0] Create time Create mesh for time = 0 Reading transportProperties Reading field p Reading field U Reading/calculating face flux field phi Starting time loop Time = 0.001 Mean and max Courant Numbers = 0 0 BICCG: Solving for Ux, Initial residual = 1, Final residual = 9.04991e06, No Iterations 18 BICCG: Solving for Uy, Initial residual = 0, Final residual = 0, No Iterations 0 ICCG: Solving for p, Initial residual = 1, Final residual = 6.86206e07, No Iterations 201 time step continuity errors : sum local = 4.28884e10, global = 9.48776e21, cumulative = 9.48776e21 ICCG: Solving for p, Initial residual = 0.672536, Final residual = 8.34589e07, No Iterations 200 time step continuity errors : sum local = 7.54547e10, global = 5.38652e20, cumulative = 6.3353e20 ExecutionTime = 2.51 s ClockTime = 3 s Time = 0.002 Mean and max Courant Numbers = 0.0296832 0.810069 BICCG: Solving for Ux, Initial residual = 0.119078, Final residual = 6.90333e06, No Iterations 17 BICCG: Solving for Uy, Initial residual = 0.386927, Final residual = 6.9427e06, No Iterations 18 ICCG: Solving for p, Initial residual = 0.722155, Final residual = 9.66376e07, No Iterations 198 time step continuity errors : sum local = 9.31977e10, global = 1.34216e19, cumulative = 1.97569e19 ICCG: Solving for p, Initial residual = 0.633029, Final residual = 9.07363e07, No Iterations 197 time step continuity errors : sum local = 9.3323e10, global = 1.01922e19, cumulative = 9.56466e20 ExecutionTime = 4.04 s ClockTime = 4 s Time = 0.003 Mean and max Courant Numbers = 0.0449432 0.926016 BICCG: Solving for Ux, Initial residual = 0.047979, Final residual = 9.80318e06, No Iterations 15 BICCG: Solving for Uy, Initial residual = 0.146506, Final residual = 4.93936e06, No Iterations 17 ICCG: Solving for p, Initial residual = 0.711663, Final residual = 9.89994e07, No Iterations 197 time step continuity errors : sum local = 7.55206e10, global = 4.25957e20, cumulative = 5.30509e20 ICCG: Solving for p, Initial residual = 0.663622, Final residual = 8.84782e07, No Iterations 197 time step continuity errors : sum local = 6.7351e10, global = 8.68074e21, cumulative = 4.43702e20 ExecutionTime = 5.55 s ClockTime = 6 s Time = 0.004 Mean and max Courant Numbers = 0.053069 0.930456 BICCG: Solving for Ux, Initial residual = 0.026967, Final residual = 8.26492e06, No Iterations 14 BICCG: Solving for Uy, Initial residual = 0.104978, Final residual = 5.43694e06, No Iterations 16 ICCG: Solving for p, Initial residual = 0.496791, Final residual = 9.2743e07, No Iterations 196 time step continuity errors : sum local = 5.71826e10, global = 6.46623e22, cumulative = 4.50168e20 ICCG: Solving for p, Initial residual = 0.45407, Final residual = 8.34512e07, No Iterations 196 time step continuity errors : sum local = 5.17042e10, global = 1.43151e19, cumulative = 9.81339e20 ExecutionTime = 7.07 s ClockTime = 7 s Time = 0.005 Mean and max Courant Numbers = 0.0597545 0.939777 BICCG: Solving for Ux, Initial residual = 0.0206086, Final residual = 5.9467e06, No Iterations 14 BICCG: Solving for Uy, Initial residual = 0.0574585, Final residual = 8.81759e06, No Iterations 14 ICCG: Solving for p, Initial residual = 0.41837, Final residual = 7.66076e07, No Iterations 195 time step continuity errors : sum local = 3.66536e10, global = 1.93771e19, cumulative = 2.91905e19 ICCG: Solving for p, Initial residual = 0.38174, Final residual = 8.99953e07, No Iterations 194 time step continuity errors : sum local = 4.33183e10, global = 1.20382e19, cumulative = 1.71523e19 ExecutionTime = 8.55 s ClockTime = 9 s Time = 0.006 Mean and max Courant Numbers = 0.0648963 0.946156 BICCG: Solving for Ux, Initial residual = 0.0139187, Final residual = 7.52264e06, No Iterations 13 BICCG: Solving for Uy, Initial residual = 0.0455918, Final residual = 7.22159e06, No Iterations 14 ICCG: Solving for p, Initial residual = 0.253609, Final residual = 8.96376e07, No Iterations 192 time step continuity errors : sum local = 3.92679e10, global = 5.5118e20, cumulative = 2.26641e19 ICCG: Solving for p, Initial residual = 0.229742, Final residual = 8.75365e07, No Iterations 192 time step continuity errors : sum local = 3.8378e10, global = 8.76987e20, cumulative = 1.38943e19 ExecutionTime = 10.02 s ClockTime = 10 s Time = 0.007 Mean and max Courant Numbers = 0.0692539 0.949465 BICCG: Solving for Ux, Initial residual = 0.0116939, Final residual = 6.09785e06, No Iterations 13 BICCG: Solving for Uy, Initial residual = 0.0320625, Final residual = 7.43667e06, No Iterations 13 ICCG: Solving for p, Initial residual = 0.180605, Final residual = 9.52495e07, No Iterations 189 time step continuity errors : sum local = 3.83623e10, global = 1.58405e19, cumulative = 1.94624e20 ICCG: Solving for p, Initial residual = 0.163707, Final residual = 9.64875e07, No Iterations 188 time step continuity errors : sum local = 3.90012e10, global = 7.23862e20, cumulative = 9.18486e20 ExecutionTime = 11.47 s ClockTime = 12 s Time = 0.008 Mean and max Courant Numbers = 0.0729413 0.952808 BICCG: Solving for Ux, Initial residual = 0.00909364, Final residual = 8.65935e06, No Iterations 12 BICCG: Solving for Uy, Initial residual = 0.0269566, Final residual = 6.7032e06, No Iterations 13 ICCG: Solving for p, Initial residual = 0.109123, Final residual = 8.15401e07, No Iterations 188 time step continuity errors : sum local = 3.10857e10, global = 1.21822e20, cumulative = 7.96664e20 ICCG: Solving for p, Initial residual = 0.0988474, Final residual = 9.23239e07, No Iterations 187 time step continuity errors : sum local = 3.52258e10, global = 1.454e19, cumulative = 6.57332e20 ExecutionTime = 12.91 s ClockTime = 13 s Time = 0.009 Mean and max Courant Numbers = 0.0761604 0.954918 BICCG: Solving for Ux, Initial residual = 0.00788061, Final residual = 8.02051e06, No Iterations 12 BICCG: Solving for Uy, Initial residual = 0.0216089, Final residual = 7.38888e06, No Iterations 12 ICCG: Solving for p, Initial residual = 0.0751862, Final residual = 8.52566e07, No Iterations 184 time step continuity errors : sum local = 3.22218e10, global = 2.50286e19, cumulative = 1.84553e19 ICCG: Solving for p, Initial residual = 0.0683289, Final residual = 8.71176e07, No Iterations 184 time step continuity errors : sum local = 3.2978e10, global = 2.72231e20, cumulative = 1.5733e19 ExecutionTime = 14.32 s ClockTime = 14 s Time = 0.01 Mean and max Courant Numbers = 0.078984 0.956873 BICCG: Solving for Ux, Initial residual = 0.00663957, Final residual = 5.93408e06, No Iterations 12 BICCG: Solving for Uy, Initial residual = 0.0188549, Final residual = 6.51374e06, No Iterations 12 ICCG: Solving for p, Initial residual = 0.0476218, Final residual = 7.87959e07, No Iterations 183 time step continuity errors : sum local = 2.92796e10, global = 3.27487e20, cumulative = 1.90078e19 ICCG: Solving for p, Initial residual = 0.0435532, Final residual = 8.46472e07, No Iterations 182 time step continuity errors : sum local = 3.1477e10, global = 9.94609e20, cumulative = 9.06174e20 ExecutionTime = 15.72 s ClockTime = 16 s Time = 0.011 Mean and max Courant Numbers = 0.0815052 0.958373 BICCG: Solving for Ux, Initial residual = 0.00586825, Final residual = 9.88343e06, No Iterations 11 BICCG: Solving for Uy, Initial residual = 0.0161717, Final residual = 9.9482e06, No Iterations 11 ICCG: Solving for p, Initial residual = 0.0356458, Final residual = 9.13337e07, No Iterations 182 time step continuity errors : sum local = 3.37795e10, global = 2.71539e19, cumulative = 3.62156e19 ICCG: Solving for p, Initial residual = 0.0329001, Final residual = 7.4762e07, No Iterations 182 time step continuity errors : sum local = 2.7672e10, global = 1.66423e19, cumulative = 5.28579e19 ExecutionTime = 17.13 s ClockTime = 17 s Time = 0.012 Mean and max Courant Numbers = 0.0837613 0.959674 BICCG: Solving for Ux, Initial residual = 0.00515329, Final residual = 8.0906e06, No Iterations 11 BICCG: Solving for Uy, Initial residual = 0.0144256, Final residual = 7.93775e06, No Iterations 11 ICCG: Solving for p, Initial residual = 0.0248828, Final residual = 8.89308e07, No Iterations 179 time step continuity errors : sum local = 3.28367e10, global = 6.074e20, cumulative = 5.89319e19 ICCG: Solving for p, Initial residual = 0.0232373, Final residual = 9.85799e07, No Iterations 178 time step continuity errors : sum local = 3.64155e10, global = 1.95186e19, cumulative = 3.94132e19 ExecutionTime = 18.5 s ClockTime = 19 s Time = 0.013 Mean and max Courant Numbers = 0.0858044 0.960774 BICCG: Solving for Ux, Initial residual = 0.00462194, Final residual = 7.26935e06, No Iterations 11 BICCG: Solving for Uy, Initial residual = 0.012806, Final residual = 7.36052e06, No Iterations 11 ICCG: Solving for p, Initial residual = 0.0208117, Final residual = 8.85416e07, No Iterations 179 time step continuity errors : sum local = 3.25548e10, global = 9.66412e20, cumulative = 2.97491e19 ICCG: Solving for p, Initial residual = 0.0196253, Final residual = 8.99491e07, No Iterations 179 time step continuity errors : sum local = 3.30787e10, global = 7.37018e21, cumulative = 3.04861e19 ExecutionTime = 19.88 s ClockTime = 20 s Time = 0.014 Mean and max Courant Numbers = 0.0876578 0.961719 BICCG: Solving for Ux, Initial residual = 0.00415071, Final residual = 6.38942e06, No Iterations 11 BICCG: Solving for Uy, Initial residual = 0.0115789, Final residual = 1.44374e+95, No Iterations 1001 ICCG: Solving for p, Initial residual = 1, Final residual = 9.6264e07, No Iterations 189 time step continuity errors : sum local = 1.29818e+87, global = 5.93881e+76, cumulative = 5.93881e+76 ICCG: Solving for p, Initial residual = 0.934444, Final residual = 9.29122e07, No Iterations 189 time step continuity errors : sum local = 1.32389e+87, global = 5.63113e+75, cumulative = 6.50193e+76 ExecutionTime = 25.46 s ClockTime = 26 s Time = 0.015 Mean and max Courant Numbers = 7.04549e+93 3.68526e+94 BICCG: Solving for Ux: solution singularity BICCG: Solving for Uy: solution singularity ICCG: Solving for p, Initial residual = 1, Final residual = 1.40439, No Iterations 5001 time step continuity errors : sum local = 1.88237e+101, global = 1.44676e+86, cumulative = 1.44676e+86 ICCG: Solving for p: solution singularity time step continuity errors : sum local = 3.57117e+109, global = 6.875e+92, cumulative = 6.87499e+92 ExecutionTime = 39.12 s ClockTime = 39 s Time = 0.016 Mean and max Courant Numbers = 1.45147e+109 1.83744e+112 BICCG: Solving for Ux: solution singularity BICCG: Solving for Uy: solution singularity ICCG: Solving for p: solution singularity time step continuity errors : sum local = 1.67439e+110, global = 2.48722e+94, cumulative = 2.41847e+94 ICCG: Solving for p: solution singularity time step continuity errors : sum local = 1.26382e+111, global = 1.14441e+95, cumulative = 9.0256e+94 ExecutionTime = 39.47 s ClockTime = 40 s Time = 0.017 Mean and max Courant Numbers = 7.14503e+110 8.90101e+113 BICCG: Solving for Ux: solution singularity BICCG: Solving for Uy: solution singularity ICCG: Solving for p: solution singularity time step continuity errors : sum local = 4.72906e+112, global = 1.80693e+96, cumulative = 1.89719e+96 ICCG: Solving for p: solution singularity time step continuity errors : sum local = 2.80161e+113, global = 7.47804e+96, cumulative = 5.58085e+96 ExecutionTime = 39.82 s ClockTime = 40 s Time = 0.018 Mean and max Courant Numbers = 1.77155e+113 4.14651e+116 BICCG: Solving for Ux: solution singularity BICCG: Solving for Uy: solution singularity ICCG: Solving for p: solution singularity time step continuity errors : sum local = 6.67211e+114, global = 1.87091e+99, cumulative = 1.87649e+99 ICCG: Solving for p: solution singularity time step continuity errors : sum local = 2.64446e+115, global = 5.34057e+99, cumulative = 7.21706e+99 ExecutionTime = 40.16 s ClockTime = 40 s Time = 0.019 Mean and max Courant Numbers = 1.73067e+115 2.43263e+118 BICCG: Solving for Ux: solution singularity BICCG: Solving for Uy: solution singularity ICCG: Solving for p: solution singularity time step continuity errors : sum local = 1.03355e+116, global = 2.80081e+99, cumulative = 4.41625e+99 ICCG: Solving for p: solution singularity time step continuity errors : sum local = 8.83257e+116, global = 6.15057e+100, cumulative = 6.5922e+100 ExecutionTime = 40.51 s ClockTime = 41 s Time = 0.02 Mean and max Courant Numbers = 5.19846e+116 7.38682e+119 BICCG: Solving for Ux: solution singularity BICCG: Solving for Uy: solution singularity ICCG: Solving for p: solution singularity time step continuity errors : sum local = 3.63936e+117, global = 3.23517e+101, cumulative = 3.89439e+101 ICCG: Solving for p: solution singularity time step continuity errors : sum local = 8.63794e+118, global = 5.14892e+102, cumulative = 5.53836e+102 ExecutionTime = 40.85 s ClockTime = 41 s Time = 0.021 Mean and max Courant Numbers = 4.54631e+118 1.97763e+122 BICCG: Solving for Ux: solution singularity BICCG: Solving for Uy: solution singularity ICCG: Solving for p: solution singularity time step continuity errors : sum local = 2.1432e+120, global = 1.32031e+104, cumulative = 1.3757e+104 ICCG: Solving for p: solution singularity time step continuity errors : sum local = 4.05285e+121, global = 1.8166e+105, cumulative = 1.95417e+105 ExecutionTime = 41.2 s ClockTime = 41 s Time = 0.022 Mean and max Courant Numbers = 2.11858e+121 8.97658e+124 BICCG: Solving for Ux: solution singularity BICCG: Solving for Uy: solution singularity ICCG: Solving for p: solution singularity time step continuity errors : sum local = 1.7996e+122, global = 8.53168e+105, cumulative = 1.04858e+106 ICCG: Solving for p: solution singularity time step continuity errors : sum local = 2.79712e+124, global = 1.02687e+108, cumulative = 1.01639e+108 ExecutionTime = 41.55 s ClockTime = 42 s Time = 0.023 Mean and max Courant Numbers = 1.43656e+124 8.42775e+127 BICCG: Solving for Ux: solution singularity BICCG: Solving for Uy: solution singularity ICCG: Solving for p: solution singularity time step continuity errors : sum local = 1.90088e+125, global = 4.21083e+109, cumulative = 4.31247e+109 ICCG: Solving for p: solution singularity time step continuity errors : sum local = 1.54769e+126, global = 6.58994e+109, cumulative = 1.09024e+110 ExecutionTime = 41.89 s ClockTime = 42 s Time = 0.024 Mean and max Courant Numbers = 1.07014e+126 1.86495e+129  One of the processes started by mpirun has exited with a nonzero exit code. This typically indicates that the process finished in error. If your process did not finish in error, be sure to include a "return 0" or "exit(0)" in your C code before exiting the application. PID 32391 failed on node n0 (127.0.0.1) due to signal 15.  madhavan@varese:~/OpenFOAM/madhavan1.3/run/tutorials/icoFoam> 

July 26, 2006, 13:09 
Ok, I think I've solved the pr

#2 
Senior Member
Srinath Madhavan (a.k.a pUl)
Join Date: Mar 2009
Location: Edmonton, AB, Canada
Posts: 703
Rep Power: 17 
Ok, I think I've solved the problem. Earlier, I used the following command to run icoFoam in parallel:
mpirun np 4 icoFoam . cavity_large_mesh_parallel parallel and I faced those random problems... Now, I use the command below to run icoFoam in parallel: mpirun ssi rpi lamd C icoFoam . cavity_large_mesh_parallel parallel and it seems to run without problems. Note that I did not have to specify 'np 4'. Apparently lamd and the kernel (linux) take care of spawning 4 processes as discussed here[1]. In both cases, I first do a lamboot v machines The machines file (again in both cases) contains: localhost cpu=4 [1] http://www.lammpi.org/MailArchives/...01/04/2457.php 

July 27, 2006, 04:36 
Can you make it crash repeated

#3 
Senior Member
Eugene de Villiers
Join Date: Mar 2009
Posts: 726
Rep Power: 17 
Can you make it crash repeatedly and consistently when using the old command?


July 27, 2006, 13:13 
If 10 times can represent "rep

#4 
Senior Member
Srinath Madhavan (a.k.a pUl)
Join Date: Mar 2009
Location: Edmonton, AB, Canada
Posts: 703
Rep Power: 17 
If 10 times can represent "repeatedly", then yes. However, the crashes occur far apart in time steps. So with the old command, I could crash at the very beginning or even just before the last few time steps. Like I said, it appears random. Out of curiosity, why is that important? Thanks!


July 28, 2006, 03:30 
If the occurrance is random, i

#5 
Senior Member
Eugene de Villiers
Join Date: Mar 2009
Posts: 726
Rep Power: 17 
If the occurrance is random, it is likely that the problem is with your hardware/drivers (or very unlikely in your case an uninitialised pointer object). If it occurs consistently it is more likely a problem with Foam or your setup.
I have run many cases on 8cpu shared memory opterons using the "mpirun np 8" command without problems, so I doubt it is a Foam issue. 

July 28, 2006, 14:21 
Thanks Eugene. Like I said the

#6 
Senior Member
Srinath Madhavan (a.k.a pUl)
Join Date: Mar 2009
Location: Edmonton, AB, Canada
Posts: 703
Rep Power: 17 
Thanks Eugene. Like I said the thing works fine if I use:
mpirun ssi rpi lamd C icoFoam . cavity_large_mesh_parallel parallel 

July 19, 2007, 06:02 
FYI: I have discovered a bette

#7 
Senior Member
Srinath Madhavan (a.k.a pUl)
Join Date: Mar 2009
Location: Edmonton, AB, Canada
Posts: 703
Rep Power: 17 
FYI: I have discovered a better rpi for shared memory that works without any problems:
mpirun C ssi rpi usysv icoFoam_1 . unsteady_validation_refined parallel > unsteady_validation_refined/log 2>&1 & 

July 21, 2007, 04:58 
Update: On this ppc64 system,

#8 
Senior Member
Srinath Madhavan (a.k.a pUl)
Join Date: Mar 2009
Location: Edmonton, AB, Canada
Posts: 703
Rep Power: 17 
Update: On this ppc64 system, although LAM 7.1.3 compiles fine, the MPI process randomly fails at different time steps even if I choose a different RPI like usysv (basically the same problem as described in the beginning). An MPICH build fails with the following error message:
gcc m64 o serv_p4 serv_p4.o server_ssl.o lcrypt bin/mpicc o /home/madhavan/OpenFOAM/OpenFOAM1.4/src/mpich1.2.7p1/bin/mpichversion /home/madhavan/ OpenFOAM/OpenFOAM1.4/src/mpich1.2.7p1/util/mpichversion.c /home/madhavan/OpenFOAM/OpenFOAM1.4/src/mpich1.2.7p1/util/mpichversion.c: In function âmainâ: /home/madhavan/OpenFOAM/OpenFOAM1.4/src/mpich1.2.7p1/util/mpichversion.c:67: warning: incompatible implicit declaration of builtin function âexitâ collect2: ld terminated with signal 11 [Segmentation fault] /usr/bin/ld: warning: powerpc:common architecture of input file `mpichversion.o' is incompatible wit h powerpc:common64 output mpichversion.o: In function `main': mpichversion.c.text+0xe0): relocation truncated to fit: R_PPC_REL24 against `strcmp' mpichversion.c.text+0x124): relocation truncated to fit: R_PPC_REL24 against `strcmp' mpichversion.c.text+0x168): relocation truncated to fit: R_PPC_REL24 against `strcmp' mpichversion.c.text+0x1ac): relocation truncated to fit: R_PPC_REL24 against `strcmp' mpichversion.c.text+0x1f0): relocation truncated to fit: R_PPC_REL24 against `strcmp' make[1]: *** [mpiutils] Error 1 make: *** [mpi] Error 2 Luckily, OpenMPI 1.2.3 was built successfully and I am testing it now. 

July 21, 2007, 05:22 
I do not know if your problem

#9 
Guest
Posts: n/a

I do not know if your problem is the same as mine, but, in my case , on a ppc64  128 node machine i had compiled lam and openFoam and I was facing a very similar problem, that the run used to stop randomly. I was able to solve the problem by compiling openFoam with the mpich libraries which i found already existing on the system.
I really don't know why it used to stop, but the above solved the problem completely. 

July 22, 2007, 02:58 
Thanks for sharing your experi

#10 
Senior Member
Srinath Madhavan (a.k.a pUl)
Join Date: Mar 2009
Location: Edmonton, AB, Canada
Posts: 703
Rep Power: 17 
Thanks for sharing your experience. It appears that using OpenMPI 1.2.3 has solved this problem for good. By the way I easily get 2X speedup when using two processors. IBM sure did a nice job with their dualcores


Thread Tools  Search this Thread 
Display Modes  


Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
IcoFoam in parallel Issue with speed up  hjasak  OpenFOAM Running, Solving & CFD  19  October 11, 2011 17:07 
Density in icoFoam Densidad en icoFoam  manuel  OpenFOAM Running, Solving & CFD  8  September 22, 2010 04:10 
OF141dev installation Woes  chegdan  OpenFOAM Installation  13  July 18, 2008 17:16 
Problem with IcoFoam in parallel  skabilan  OpenFOAM Running, Solving & CFD  12  April 1, 2008 05:55 
SimpleFoam woes  msrinath80  OpenFOAM Bugs  2  April 13, 2007 10:15 