Evolution of File System
Evolution of File System
SYSTEM
Database Systems
Mariha Asad
EVOLUTION OF FILE SYSTEM
• A brief explanation of the evolution of file system data processing can be helpful
in understanding the data access limitation that database attempt to overcome
• Creating database design that avoid the pitfalls of earlier systems requires that
the designer understand what the problems of the earlier systems were and how
to avoid them
MANUAL FILE SYSTEM
• Generating reports from manual file systems was slow and cumbersome
• Therefore a computer based system was created that would track data and
produce required reports
• When business user wanted data from computerized file, they sent requests for
the data to data processing DP specialist
COMPUTERIZED FILE SYSTEMS
• For each request, the DP specialist had to create programs to retrieve the data
from the file, manipulate it in whatever manner user had requested and present it
as a printed report
• Gradually customer files are computerized then files of sales department
computerized and then agent file computerized
• As more and more files are developed, the problems of computerized file system
arises
COMPUTERIZED FILE SYSTEMS
• As writing the program was required even for generating the simple reports
• It makes ad-hoc queries impossible
• 1 week or 1 month was required to generate reports
COMPLEX SYSTEM ADMINISTRATION
• System administration become more difficult as the number of files in the system
expand
LACK OF SECURITY AND LIMITED DATA
SHARING
• Sharing data among multiple geographically dispersed users introduced a lot of
security risks
• Security features include effective password protection, the ability to lock out
parts of files etc.
• Such security and data sharing features are difficult to program and therefore
omitted
EXTENSIVE PROGRAMMING
• When we change the characteristic of data, such as changing the type of data
• It require changes in all the programs that access the file
• It is data dependency
• Conversely data independence exists when it is possible to make changes in the
data characteristics without affecting the application program
DATA REDUNDANCY
• Same file is given to different branches (for example record for a bank system)
• This will lead to islands of information
• The island of information often contain different version of same data
• The agent name and phone numbers occur in both the customer and agent file
• You only need one correct copy of the agent name and phone number
• Having them occur in mor than one place produces data redundancy
• Data redundancy occurs when same data is stored unnecessarily at different
places
DATA REDUNDANCY
• Data inconsistency
• Data inconsistency exists when different and conflicting version of the same data appear in different
places
DATA REDUNDANCY
• Data anomalies
• Defines as an abnormality
• A value changes in only one file and do not change in the other file at other location
• A data anomaly develops when not all of the required changes in the redundant data are made
successfully
• Update anomaly
• Insertion anomaly
• Deletion anomaly
THANKS