continuing with some history...
* database processing systems grew out of
those issues we discussed as file-processing
systems involve more and more files of
more and more data;
* one big idea: to HIDE/abstract file-level
details from applications, in
this case by using a DBMS
* potential advantages:
* more-integrated data
* reduced (unnecessary) data duplication
* decreased application program dependence
* easier (ad-hoc) representation of the
users' perspective
* ...and more (increased security,
tools for backup and recovery, etc.)
* there are also potential disadvantages
to database-processing systems:
* DBMSs can be complex
* they can be big (size)
* cost of DBMS
* cost of conversion
* performance (a DBMS is built to be more
general, satisfying the needs of many --
a custom-built application program
can be faster)
* central point of failure