[CIG-SHORT] GeoFEST runs for strike-slip model

[Jiangning Lü]

I've done with the 250m resolution GeoFEST input file for strike-slip model,
when I executed command: gfmeshparse, tried to get the *.jpl file for parallel
calculation, error occured:

GeoFEST Mesh File Name:
bm4_uni0250_2.gf
Loading Data...
Activity Data...
Segmentation fault

Comparing to 500m resolution file, the code gfmeshparse collaps sooner fo 250m
resolution. I'm wondering is it because of the limited memory on my pc to deal
with the huge size of input files?

Jiangning

Quoting Jiangning Lü <Jiangning at MIT.EDU>:

> Hello,
>
> I'v made GeoFEST input files for strike-slip model, and I'm trying to 
> run them
> on parallel cluster. The 1km resolution file goes well, and I've put 
> the input
> file and output results on CIG website.
> There is an error when running 500m resoluton case.
> when I executed the command: gfmeshparse  to make the *.jpl file from geofest
> input file, an error occured:
> "cycling.  . .
>  cycling.  . .
> Displacement Data...
> Element Data...
> Material Data...
> Node Data...
> Data Loading Complete...
> Edge Numbering Assigned...
> Face Numbering Assigned...
> Segmentation fault
> "
>   I've checked the file many times, the 500m resolution file is made the same
> as 1km resolution file except resolution. The 1km file goes well, but 
> 500m file
> does not.  I've put the 500m input file on the website.
>
>   Because I couldn't run it on parallel cluster, I tried to run it on 
> my pc. It
> took more than 14 hours to stop at the messaage:" calling solver...", 
> I have to
> kill it...
>
> could someone knowing the geofest give me a hint what's the problem?
>
> For the 250m resolution case,  I'm also trying to made the GeoFEST 
> input files,
> I've using several matlab codes to made the input file from lagrit mesh and
> boundary conditions. Because of the huge node number, it usually took 
> more than
> 10 hours to run the matlab codes on my pc.
> If someone have the 250m resolution GeoFEST input files already, I could run
> them on the  Parallel cluster.
>
> Best,
> Jiangning
> EAPS,MIT
> _______________________________________________
> CIG-SHORT mailing list
> CIG-SHORT at geodynamics.org
> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>



---
Jiangning Lü
54-521     x3-6389
Geophysics Program
Earth, Atmospheric and Planetary Sciencs
Massachusetts Institute of Technology
johnnylu at mit.edu