Ch 2: Intro to the Relational Model

Question 1

True or False:

No two tuples in a Relation

can have exactly the same values

for all attributes

Answer 1

True!

The values of a tuple(row) must be such that they can

uniquely identify the tuple.

Therefore, we can always distinguish between tuples.

Question 2

How Seperate Tables

Can Be Related

Answer 2

Use Common Attribute Names:

Example:

The “Department” and “Teachers” table both have an attribute called “Dept_Name”

Can get a Department Name from the “Department” table,

then use it to find all “Teacher” tuples that have that Department

Question 3

Relation Schema

Answer 3

• A Relation is similar to a variable
• A Relation Schema is more like a Type Definition

The Relation Schema consists of a list of attribute names and their corresponding domains.

Structure: Relation_Name( attribute, attribute, … )

example: department( dept_name, building, budget)

Question 4

Relational Model:

Table Structure

Answer 4

• Each table is assigned a unique name
• Columns represent a certain Attribute and have header info
• Each Row is a set of related data,
  
  • It is an n-tuple of values
• If an attribute is unknown, or doesn’t exist for a particular relation instance, the null value is used

Question 5

Relational Model:

Types of Keys

Answer 5

• Super Key
  
  • A set of one or more attibutes which allow a tuple in a relation to be uniquely identified.
  • A good example would be (User_ID, Name)
• Candidate Key
  
  • A Super Key that can’t be reduced further
    
    • No proper subset is also a Super Key
• Primary Key
  
  • A Candidate Key chosen by the database designer to identify tuples within a relation.
  • Must be chosen with care
• Foreign Key

Question 6

What is a

Data Model ?

Answer 6

A notation for describing data or information.

The description generally consists of 3 parts:

• Structure of the data
• Operations on the data
  
  • Queries
  • Modifications
• Constraints on the data

Question 7

What is a Key

for a relation ?

Answer 7

• The most fundamental kind of constraint
• A set of attributes forms a Key for a relation when:
  
  • No two tuples in a relation instance may have the same values in all attributes of the key
• This makes it a unique identifier for each tuple

Question 8

Argument for

Why the Relational Model

is preferred in DBMS

Answer 8

• DBs are large, so access efficiency and modification efficiency are very important.
• Ease of Use is also important

Both goals can be achieved with the Relational Model:

• Simple, limited approach to structuring data,
  
  • yet reasonably versatile, so anything can be modeled
• Limited, yet useful collection of operations on data

Question 9

Two most important

Data Models for

Database Systems

Answer 9

• Relational Model
  
  • Tables or arrays
  • Including Object-Relational extensions
• Semistructured Data Model
  
  • trees or graphs
  • Includes XML and related standards