Data Redundancy

Data Redundancy

Learning Outcomes:

Explain the concepts of data redundancy and discuss methods or measures used to reduce data redundancy.

Relevant past paper:

DSE ICT Elec A(SP-2016): SP 2a. PP 2a, 4ai. 2012 3a, 4b. 2014 4aii. 2015 2di. 2016 2a.

AS CA(2000-2013): 2000 1a. 2002 4bi. 2005 6c. 2006 9g. 2012 9aii.

AL CS P2: 2004 1a. 2005 1.

A poorly designed database can lead to data redundancy.

Borrow

BookID	StuID	StuName
001	123	John
002	123	John
003	245	Peter

Suppose this is a record of student borrowing books. The attribute StuName is not necessary here because StuName depends on StuID. (This is known as functional dependency and will be covered in normalisation)

Problems occur when a student borrows more than 1 book. (more than 1 entry with StuID 123). StuName John is repeatedly stored.

The expected problems include:

• data redundancy: repeated data unnecessarily inserted.
• data inconsistency: when the repeated (redundant) data are not updated at the same time.
• storage wastage: the unnecessarily repeated data take up extra storage.
• overheads on database update: more fields, which are redundant, need to be updated when they are changed à affect database performance

 (These problems may also be referred to insertion anomaly, deletion anomaly and modification anomaly)

Derived attribute

Derived attribute is an attribute that is derived from other field(s).

Test

Test1	Test2	TestTotal

The TestTotal is a derived attribute because its content can be calculated from other fields (Test1 and Test2). In this sense, it is also redundant.

Sometimes, the derived attribute may still be included in a table.

• Advantage: more convenient to access the data.
• Disadvantage: refer to the 4 points in data redundancy.