DatabaseSystemConcepts(SixthEdition) - Chapter 2 - Introduction to the Relational Model

Introduction to the Relational Model

The relational model is today the primary data model for commercial dataprocessing
applications. It attained its primary position because of its simplicity, which eases the job of the programmer, compared to earlier data models such as the network model or the hierarchical model.

In this chapter, we first study the fundamentals of the relational model. A substantial theory exists for relational databases.We study the part of this theory dealing with queries in Chapter 6. In Chapters 7 through 8, we shall examine aspects of database theory that help in the design of relational database schemas, while in Chapters 12 and 13we discuss aspects of the theory dealingwith efficient processing of queries.

2.1 Structure of Relational Databases

A relational database consists of a collection of tables, each of which is assigned a unique name. For example, consider the instructor table of Figure 2.1, which stores information about instructors. The table has four column headers: ID, name, dept name, and salary. Each row of this table records information about an instructor, consisting of the instructor’s ID, name, dept name, and salary. Similarly, the course table of Figure 2.2 stores information about courses, consisting of a course id, title, dept name, and credits, for each course. Note that each instructor is identified by the value of the column ID, while each course is identified by the value of the column course id.

Figure 2.3 shows a third table, prereq,which stores the prerequisite courses for each course. The table has two columns, course id and prereq id. Each rowconsists of a pair of course identifiers such that the second course is a prerequisite for the first course.

Thus, a row in the prereq table indicates that two courses are related in the sense that one course is a prerequisite for the other. As another example, we consider the table instructor, a row in the table can be thought of as representing the relationship between a specified ID and the corresponding values for name, dept name, and salary values.

Figure 2.1 The instructor relation.

In general, a row in a table represents a relationship among a set of values. Since a table is a collection of such relationships, there is a close correspondence between the concept of table and the mathematical concept of relation, fromwhich the relational data model takes its name. In mathematical terminology, a tuple is simply a sequence (or list) of values. A relationship between n values is represented mathematically by an n-tupleof values, i.e., a tuple with n values, which corresponds to a row in a table.

Figure 2.2 The course relation.

Figure 2.3 The prereq relation.
Thus, in the relationalmodel the term relation is used to refer to a table, while the term tupleis used to refer to a row. Similarly, the term attributerefers to a column of a table.

Examining Figure 2.1,we can see that the relation instructor has four attributes:ID, name, dept name, and salary.

We use the term relation instanceto refer to a specific instance of a relation, i.e., containing a specific set of rows. The instance of instructor shown in Figure 2.1 has 12 tuples, corresponding to 12 instructors.

In this chapter,weshall be using a number ofdifferent relations to illustrate the various concepts underlying the relational data model. These relations represent part of a university. They do not include all the data an actual university database would contain, in order to simplify our presentation.We shall discuss criteria for the appropriateness of relational structures in great detail in Chapters 7 and 8.

The order in which tuples appear in a relation is irrelevant, since a relation is a set of tuples. Thus, whether the tuples of a relation are listed in sorted order, as in Figure 2.1, or are unsorted, as in Figure 2.4, does not matter; the relations in the two figures are the same, since both contain the same set of tuples. For ease of exposition, we will mostly show the relations sorted by their first attribute.

Figure 2.4 Unsorted display of the instructor relation.

For each attribute of a relation, there is a set of permitted values, called the domainof that attribute. Thus, the domain of the salary attribute of the instructor relation is the set of all possible salary values, while the domain of the name attribute is the set of all possible instructor names.

We require that, for all relations r, the domains of all attributes of r be atomic. A domain is atomic if elements of the domain are considered to be indivisible units. For example, suppose the table instructor had an attribute phone number, which can store a set of phone numbers corresponding to the instructor. Then the domain of phone number would not be atomic, since an element of the domain is a set of phone numbers, and it has subparts, namely the individual phone numbers in the set.

The important issue is not what the domain itself is, but rather how we use domain elements in our database. Suppose now that the phone number attribute stores a single phone number. Even then, if we split the value from the phone number attribute into a country code, an area code and a local number, we would be treating it as a nonatomic value. If we treat each phone number as a single indivisible unit, then the attribute phone number would have an atomic domain.

In this chapter, as well as in Chapters 3 through 6, we assume that all attributes have atomic domains. In Chapter 22, we shall discuss extensions to the relational data model to permit nonatomic domains.

The null value is a special value that signifies that the value is unknown or does not exist. For example, suppose as before that we include the attribute phone number in the instructor relation. It may be that an instructor does not have a phone number at all, or that the telephone number is unlisted. We would then have to use the null value to signify that the value is unknown or does not exist. We shall see later that null values cause a number of difficulties when we access or update the database, and thus should be eliminated if at all possible.We shall assume null values are absent initially, and in Section 3.6 we describe the effect of nulls o