search

markovmodel / adaptivemd

A python framework to run adaptive Markov state model (MSM) simulation on HPC resources
GNU Lesser General Public License v2.1
18 stars 7 forks source link

Larger file storage #35

Closed jhprinz closed 7 years ago

jhprinz commented 7 years ago

resolves #26.

Add the possibility to store arbitrary large objects. So far this is only used by PyEMMAModel to save the returned results. This is handled by gridFS and it automatically divides large files into blocks.

I also store the result model.json in a folder in projects/proj-name/models for later processing.

jhprinz commented 7 years ago

Done.

jhprinz commented 7 years ago

If @thempel has more problems with file sizes we can use the same for storing file contents in general...

thempel commented 7 years ago

In general: Yes, it would be good to not be limited in the system size. The system I am working with right now is not very big (25,000 atoms) and the files that need to be moved around are about 11.5 MB, so already pretty close to the 16 MB. In the case mentioned in #28, I suspect another problem since it used to work before.

jhprinz commented 7 years ago

Done. This is implemented. This means (when it works) we have now all major features we talked about. The featurizer stuff would be icing on the cake. Who would have thought. I approach the pure 6000 lines of code... (not considering all that I added and deleted) :)

jhprinz commented 7 years ago

merging