A database file product is personal files system that's a database rather than a hierarchy. Much less complex a concept initially but I decided to request if anybody had the idea about how exactly they may make a move such as this? Do you know the problems that an easy plan will probably miss? My first guess in an implementation could be something similar to a filesystem to for any Linux platform (most likely atop a current file system) however i really aren't well versed about how exactly that might be began. Its a passing believed that I doubt I'd ever follow-through on but I am wishing to a minimum of satisfy my curiosity.

The simplest way is always to construct it using fuse, having a database back-finish.

A far more difficult factor to complete is to be like a kernel module (VFS).

On Home windows, you could utilize IFS.

I am not necessarily sure that which you mean with "A database file product is personal files system that's a database rather than a hierarchy".

Most likely, using "Filesystem in Userspace" (FUSE), as pointed out by Osama ALASSIRY, may be beneficial. The FUSE wiki lists a lot of existing projects about databased-backed filesystems in addition to filesystems that you can search by SQL-like queries.

Refer this.Its nice.He's implemented DBFS in linux.

It has been some time because you requested this. I am surprised nobody recommended the apparent. Take a look at mainframes and minis, especially iSeries-OS (now known as IBM-i had been known as iOS or OS/400).

How you can do an relational database like a mass data store is comparatively easy. Oracle and MySQL have these. The issue is it should be basically ubiquitous for consumer programs.

Therefore the steps to have an application conversion are:

1) My way through an ordinary hierarchical filesystem

2) Data in BLOBs with light metadata within the database. File with a few catalogue information.

3) Large data in BLOBs with extensive metadata and sophisticated structures within the database. File with substantial metadata connected by using it that may be basically to comprehending the structure.

4) Internal structures from the BLOB uncovered within an object <--> Relational map with extensive meta-data. While there might be an exportable form, the applying naturally works together with the database, the idea from the file because the repository sheds.

Maybe this is a great beginning point to get a concept the way it perform.

It is a basic overview from the Firebird architecture.

Firebird is definitely an opensource RDBMS, so you'll have a real deep insight look, too, if you are interested.