Chapter 3 The Relational Database Model Logical View

Chapter 3 The Relational Database Model

Logical View of Data l. Relational Database – Designer focuses on logical representation rather than physical – Use of table advantageous Structural and data independence l Related records stored in independent tables l Logical simplicity l

A Logical View of Data l Relational database model’s structural and data independence – allow us to ignore the structure’s physical complexity. – enables us to view data logically rather than physically. l The logical view allows a simpler file concept of data storage. l The use of logically independent tables is easier to understand. l Logical simplicity yields simpler and more effective database design methodologies.

A Logical View of Data l Entities and Attributes – An entity is a person, place, event, or thing for which we intend to collect data. l University -- Students, Faculty Members, Courses – Each entity has certain characteristics known as attributes. l Student -- Student Number, Name, GPA, Date of Enrollment, Data of Birth, Home Address, Phone Number, Major – A grouping of related entities becomes an entity set. l The STUDENT entity set contains all student entities. l The FACULTY entity set contains all faculty entities.

Table Characteristics • Two-dimensional structure with rows and columns • Row (tuple) represent a single entity occurrence within the • • entity set. Columns represent attributes Row/column intersection represents single value Column values all have same data format Each column has range of values called attribute domain • Order of the rows and columns is immaterial to the DBMS • Tables must have an attribute to uniquely identify each row

A Logical View of Data l Tables and Their Characteristics – A table contains a group of related entities -- i. e. an entity set. – The terms entity set and table are often used interchangeably. – A table is also called a relation. entity set ≒ table ≒ relation

Keys – A key Consists of one or more attributes that determine other attributes – Primary key (PK) is an attribute (or a combination of attributes) that uniquely identifies any given entity (row)

Keys – The key’s role is based on a concept known as determination, which is used in the definition of functional dependence. l l The attribute B is functionally dependent on A if A determines B. ( A → B ) An attribute that is part of a key is known as a key attribute. A multi-attribute key is known as a composite key. If the attribute (B) is functionally dependent on a composite key (A) but not on any subset of that composite key, the attribute (B) is fully functionally dependent on (A). (A is minimal !)

Freshman Sophomore Junior Senior STU_HRS → STU_CLASS

Keys l Superkey – Uniquely identifies each entity l Candidate key – Minimal superkey l Primary key – Candidate key selected to uniquely identify all other attributes in a given row l Secondary key – Used only for data retrieval l Foreign key – Values must match primary key in another table

l Super key – STU_NUM – (STU_NUM, STU_LNAME) l Candidate key – STU_NUM – (STU_LNAME, STU_FNAME, STU_INIT, STU_PHONE) l Primary key – STU_NUM l Secondary key not necessarily yield a unique outcome – CUSTOMER table: l Primary key : Customer number l Secondary key : Customer last name, Customer phone number – A secondary key’s effectiveness in narrowing down a search depends on how restrictive it is.

Null Values l No data entry l Not permitted in primary key l Should be avoided in other attributes l Can represent – An unknown attribute value – A known, but missing, attribute value – A “not applicable” condition l Can create problems in logic and using formulas (COUNT, AVERAGE, and SUM)

Keys l Controlled redundancy (shared common attributes) makes the relational database work. l The primary key of one table appears again as the link (foreign key) in another table. – l VEND_CODE in PRODUCT table (Figure 3. 2) referential integrity : if the foreign key contains a value, that value refers to an existing valid tuple in another relation.

Integrity Rules l Entity integrity – (unique and no null) All Primary key entries are unique , and no part of a Primary key may be null. CUS_CODE l AGENT_CODE l l Referential integrity – (null or matching) A Foreign key may have either a null entry or an entry that matches the primary key value in a table to which it is related. l AGENT_CODE

null or matching

Relational Database Operators l Relational algebra defines theoretical way of manipulating table contents using the eight relational functions: SELECT – PROJECT – JOIN – INTERSECT – UNION – DIFFERENCE – PRODUCT – DIVIDE Use of relational algebra operators on existing tables (relations) produces new relations – l

Relational Database Operators l UNION combines all rows from two tables. The two tables must be union compatible. (share the same columns and domains)

l INTERSECT produces a listing that contains only the rows that appear in both tables. The two tables must be union compatible.

l DIFFERENCE yields all rows in one table that are not found in the other table; i. e. , it subtracts one table from the other. The tables must be union compatible.

l PRODUCT produces a list of all possible pairs of rows from two tables. (Cartesian product)

l SELECT yields values for all rows found in a table. It yields a horizontal subset of a table. Figure 2. 9

l PROJECT produces a list of all values for selected attributes. It yields a vertical subset of a table.

l JOIN allows us to combine information from two or more tables. JOIN is the real power behind the relational database, allowing the use of independent tables linked by common attributes.

Relational Database Operators • Natural JOIN links tables by selecting only the rows with common values in their common attribute(s). It is the result of a 3 -stage process: 1. 2. 3. A PRODUCT of the tables is created. (Figure 3. 12) A SELECT is performed on the output of the first step to yield only the rows for which the common attribute values match. (Figure 3. 13) Common column(s) are called join column(s) A PROJECT is performed to yield a single copy of each attribute, thereby eliminating the duplicate column. (Figure 3. 14)

Relational Database Operators – Equi. JOIN links tables based on an equality condition that compares specified columns of each table. The outcome of the Equi. JOIN does not eliminate duplicate columns and the condition or criteria to join the tables must be explicitly defined. – Theta JOIN is an equi. JOIN that compares specified columns of each table using a comparison operator other than the equality(=) comparison operator. – In an Outer JOIN, the unmatched pairs would be retained and the values for the unmatched other tables would be left null.

equi. JOIN criteria(C=E) A 1 2 3 B X X y C 10 20 10 D m n p E 10 20 40 A B C D E 1 x 10 m 10 3 y 10 m 10 2 x 20 n 20

theta JOIN criteria(C<E) A 1 2 3 B X X y C 10 20 10 D m n p E 10 20 40 A B C D E 1 x 10 n 20 1 x 10 p 40 2 x p 40 3 y 20 10 n 3 y 10 p 20 40

CUSTOMER Full Outer JOINCUSTOMER. AGENT_CODE=AGENT_CODE AGENT

CUSTOMER Left Outer JOINCUSTOMER. AGENT_CODE=AGENT_CODE AGENT

CUSTOMER Right Outer JOINCUSTOMER. AGENT_CODE=AGENT_CODE AGENT

l DIVIDE requires the use of one single-column table and one two-column table.

This is not an uncommon situation, but the solution is not straightforward since the SQL SELECT statement does not directly support relational division. l There are numerous ways to achieve the same results as a relational division, however. The easiest method is to rephrase the request. Instead of "list the suppliers who provide all product categories, “ which is difficult to process, try "list all suppliers where the count of their product categories is equal to the count of all product categories. " l

The Data Dictionary and the System Catalog l Data dictionary contains metadata to provide detailed accounting of all tables within the database. l System catalog is a very detailed system data dictionary that describes all objects within the database. – Authorized users, access privileges … – System catalog is a system-created database whose tables store the database characteristics and contents. l Terms “system catalog” and “data dictionary” are often used interchangeably

Relationships within the Relational Database l E-R Diagram (ERD) – Rectangles are used to represent entities. – Entity names are nouns and capitalized. – Diamonds are used to represent the relationship(s) between the entities. – “ 1” side of relationship l Number 1 in Chen Model l Bar crossing line in Crow’s Feet Model – “Many” relationships l Letter “M” and “N” in Chen Model l Three pronged “Crow’s foot” in Crow’s Feet Model

The 1: 1 Relationship l One entity can be related to only one other entity, and vice versa l Often means that entity components were not defined properly l Could indicate that two entities actually belong in the same table l Sometimes 1: 1 relationships are appropriate

The 1: 1 Relationship Between PROFESSOR and DEPARTMENT

The Implemented 1: 1 Relationship Between PROFESSOR and DEPARTMENT

The 1: M Relationship Between PAINTER and PAINTING

Example 1: M Relationship

The Implemented 1: M Relationship Between PAINTER and PAINTING PAINTER (PAINTER_NUM) PAINTING (PAINTING_NUM) (PAINTER_NUM)

The 1: M Relationship Between COURSE and CLASS

The Implemented 1: M Relationship Between COURSE and CLASS COURSE (CRS_CODE) CLASS (CLASS_CODE) (CRS_CODE)

The M: N Relationship l Can be implemented by breaking it up to produce a set of 1: M relationships l Can avoid problems inherent to M: N relationship by creating a composite entity or bridge entity

The ERD’s M: N Relationship Between STUDENT and CLASS

Sample Student Enrollment Data

The M: N Relationship Between STUDENT and CLASS 321452 10014

Composite Entity l The tables (Figure 3. 25) creates many redundancies. l Composite Entity ( Bridge Entity) – Links the tables originally were related in a M: N relationship – Includes at least the primary keys of the tables that are to be linked. – In ERD, represented by a diamond within a rectangle

Converting the M: N Relationship into Two 1: M Relationships STUDENT (STU_NUM) ENROLL (CLASS_CODE+STU_NUM) (CLASS_CODE, STU_NUM) CLASS (CLASS_CODE) (CRS_CODE)

Changing the M: N Relationship to Two 1: M Relationships

The Expanded Entity Relationship Model

The Relational Schema for the Ch 03_Tiny. College Database

Data Redundancy Revisited l Data redundancy leads to data anomalies – Such anomalies can destroy database effectiveness l Foreign keys – Control data redundancies by using common attributes shared by tables – Foreign keys can reduce redundancy l Sometimes, data redundancy is necessary

Changing the M: N Relationship to Two 1: M Relationships PRODUCT INVOICE LINE PRODUCT

CUSTOMER (CUS_CODE) INVOICE (INV_NUM) (CUS_CODE) LINE (INV_NUM+LINE_NUM) (INV_NUM, PROD_CODE) PRODUCT (PROD_CODE) A Small Invoicing System

The Relational Schema for the Invoicing System

Why Data Redundancies ? § § Sometimes, data redundancy is necessary Product price (LINE_PRICE in LINE table) : Data redundancies ! § The redundancy is crucial to the system’s success. Copying the product price from PRODUCT table to the LINE table means that it is possible to maintain the historical accuracy of the transactions. Product price may change.

Indexes l An index is an orderly arrangement used to logically access rows in a table of keys and pointers. l An index is composed of – an index key (the index’s reference point) (PAINTER_NUM) – a set of pointers. l Each pointer points to the location of the data identified by the key. l Makes retrieval of data faster l Each index is associated with only one table

Indexes