Concept of File

The data needed to develop a file are gathered from different sources.  These facts are then logically organized and stored in storage media to create a file records.  The objectives of file organization are: (1) To provide a means for locating, processing, select or extracting and (2) Create and maintain it.  Some consideration for file design are file size, Item Design, Cost of file media, ease for file maintenance, file privacy etc.

The data processing organization are sequential, random, indexed random and indexed sequential.

Sequential File Organization:  In this, records are stored in serial order by record key and uses batch transaction.  Thus processing is easy.  It is termed as Sequential Access Method (SAM). The File design is simple.  Low cost due to magnetic tape being used.  However, the entire file must be processed and transaction must be sorted in the same order as the file.

Magnetic Tape - Image Source: www.wikipedia.org

Random File Organization:  In random, any location can be read without reference to its previous items.  The record key will then be transformed into a storage address.  This method is called ‘Randomizing’.  In randomizing, two or more record keys may produce identical disk address.  This is called a ‘Collision’.  In this event one of the records is stored in an over-flow location.  In situation when records are to be located, it will be done quickly and randomly.  it is termed as Direct Access Method (DAM) or Basic Direct Access Method (BDAM).

Indexed Random File Organization:  Separate indexes are maintained in which the records are found.  The index will be in order by record key, for searching sequentially or in binary form.

Indexed Sequential Access Method (ISAM): In this the records are stored sequentially by record key but indexes are also maintained to allow direct retrieval based on key value.

Next: Data Processing Concepts.