The code should import compressed bulk data files

[GoogleCodeExporter]

What is the new feature/capability that you'd like?
Real life bulk data files are large.

Why is it useful?
Therefore it would be useful to import compressed (*.gz) bulk data files.

Original issue reported on code.google.com by nils...@googlemail.com on 9 Feb 2012 at 8:01

[GoogleCodeExporter]

Define "large".  Most models I've dealt with in engineering have practical 
limits of 300k elements (150k is even large).  Beyond that you have 
post-processing issues.  Also, disk space is not costly, so I'm not convinced 
that supporting a .gz would be valuable.

and it would be helpful to have an example...maybe if you made a main BDF with 
lots of include files and emailed them to me.

Original comment by mesheb82 on 9 Feb 2012 at 9:24

• Changed state: Feedback

[GoogleCodeExporter]

Can you provide the manual for any version of Nastran that supports this 
feature.  I'm hesitant to add it because as far as I know, no version of 
Nastran does this so it wouldn't get used very often and it would slow down and 
complicate the file reading.  Also, once you add support for *.gz, a reasonable 
argument can be made to add support for *.zip, *.tar.gz, and *.rar.

The more I think about it, the more I wonder about things like INCLUDE files 
and the requirements for defining those.  A manual from a pre-processor/solver 
would greatly help in defining the scope and the interface format.

Original comment by mesheb82 on 11 Feb 2012 at 12:45

[GoogleCodeExporter]

A few unrelated questions:

Could you tell me about your intended use for this project, I'd like to make a 
library that is useful to people and need some feedback of what people want to 
do.  I'm having a tough time understanding what you're trying to do if you want 
to read the OP4 and the BDF with presumably millions of nodes.

So...
(1) What kinds of problems you're trying to solve (e.g. 
static/transient/acoustic)
(2) Are you looking to use an open-source solver or if you're going to stick 
with a paid version of Nastran?
(3) Are you doing MDAO (Multi-disciplinary Analysis and Optimization) so 
structural, thermal, modal, trajectory, aero, etc.?
(4) What scale of problems are you trying to solve (e.g. models with 100k nodes 
in 10 minutes) and do that 100 times in an optimizer?

Anything else you'd like to share.  Also, you can email me if you dont want to 
share here.

Original comment by mesheb82 on 11 Feb 2012 at 2:40

[GoogleCodeExporter]

closed due to inactivity

Original comment by mesheb82 on 4 Jul 2012 at 6:37

• Changed state: WontAddFeature