I currently work for the small institutions that doesn't have a digital repository, so we begun the process of digitilising current research data: sensitive data (raw data in a form of pdf surveys, audio interviews, administrative data etc), as well as the data we would like to open for the general public (research papers linked with their anonymised datasets). We would like to store it on a local server and describe all data with the metadata and link DOI with the datasets, so DSpace seemed as a good solution.
However, I am not familiar with the DSpace paradigm - is DSpace made exclusively for storing open access data? Main goal is to make repository for every finished project that would be stored on a local server and available only to the research staff, where only a part of a project repository (datasets, papers, some pictures) would be open to general public.
Is DSpace good solution for storing sensitive data in general? If not, what do you recommended in this case - uploading the data we want to share on Zenodo or Figshare and making custom solution for the sensitive data? Can you recommend an open source solution for storing sensitive data (Pdf surveys, papers that are not yet to be shared, books etc?) that can be described with metadata (Dublin Core)?