Locally mirror nltk-data for compute cluster

[Mike Moore]

Hello everyone,

  I support a R&D org that needs to use nltk-data on a large compute cluster.  Rather than allow all node to access the internet to download the nltk-data models, etc., I would like to setup a local mirror or copy of the datasets and have the researchers use the mirror.

Has anyone setup a similar configuration before?  I did not see any information in the FAQ/documentation regarding this type of solution.

Thank you

-Mike Moore