Talk:PAT.File Storage

Enhancement of Description and Rationale Descriptions

• The description and rationale of File Engine should be revised. Suggestion:

This facility handles the actual storing/retrieving of semi-structured data in the form of files. It offers a file system to end-users and systems (amongst which application engine's and structured data stores) that need to store data. It may handle migration/replication to other instances of itself, encryption, compression, deduplication and other "standard" storage-related processing. It also can handle granular, file-based permissions. Note that it requires Raw Storage to physically store the bits and bytes that make up the pieces of unstructured data.

• Because of the file-based nature, archiving is supported at this level.

• A recursive flow should be drawn from and to the File Engine

--Daniel Jumelet, during meeting with Joost Bleijenberg and Rinus van der Kolk, 15:29, 16 November 2011 (CET)

Comments are integrated, although I do not want to focus solely on files and file systems as the mechanism underpinning the storage of the loosely structured data. There are other ways to store loosely structured data than just files, e.g. the Windows NT data streams, and perhaps in the future other concepts can also apply. So while mostly using the example of actual files, the text should never limit applicability to that single concept. --Jan Schoonderbeek 12:19, 17 November 2011 (CET)