[CIG-SHORT] BM Reverse Slip, no gravity...

[Eric Andreas Hetland]

Sorry for the delay:

bmrsnog_tet4_1000m results for my run of Pylith using Charles' latest mesh
and BCs are here:
http://www.gps.caltech.edu/~eah/CFEM/bmrsnog_tet4_1000m_EAH.tar.gz

bmrsnog_hex8_1000m I had a problem running. I could see nothing apparent
with the mesh, and in brief I got the following error:

Hello from pl3dscan.__init__ (end)!
terminate called after throwing an instance of 'ALE::Exception'

bmrsnog_tet4_0500m also resulted in an error:

Hello from pl3dscan.__init__ (end)!
Creating new bundle for dim 0
Creating new field material
Creating new field split
Creating new bundle for dim 3
Creating new field material
Creating new field split
terminate called after throwing an instance of 'std::bad_alloc'
  what():  St9bad_alloc

I am not sure how to interpret these error messages, but if it is a mesh
problem, I can look into it.

On a related note, I was not even sure if I was supposed to still be running
these in Pylith, since previous emails indicated Charles already did it
(since he made the meshes, etc). I am happy to run them, and I should be
less tardy now that we have Pylith installed on our linux machines and I am
not reliant on my iBook.

Concerning the problems with memory in the okada scripts, using these in
matlab has always ran into memory problems (I have not tried Brad's cleaned
up versions which fix the memory problem), but for what it is worth, I have
always split the input file up into small bite-size pieces instead of
passing matlab a gazilion points. Yes, this looses some efficiency, but for
as often as I needed to do this, it is not really a problem. I would imagine
even with sleeker code, there may still be a memory problem, especially with
120,000 points...

If there is anything the non-experts around here can do to help out, let me
know.

- Eric.