Module 7: Implementing Data Integrity

Module 1: Course OverviewModule 4: Deriving a Logical Data Design Module 5: Normalizing the Logical Data Design Overview of DataIntegrity Implementing BusinessRules and Data Integrity Ac

Module 7: Implementing Data Integrity

Module 1: Course Overview

Module 4: Deriving a Logical Data Design Module 5: Normalizing the

Logical Data Design

Overview of DataIntegrity

Implementing BusinessRules and Data Integrity

Activity 7.1: Identifying Data Integrity Requirements

Activity 7.2: Determining

Business and Data Integrity

Implementations

Module 7: Implementing Data Integrity

! Overview

In this module

At the end of this module, you will be able to:

" Identify business rules that relate to data integrity

" Identify the data integrity requirement type to which a business rule applies

" Evaluate and determine a location for implementing data integrity

" List implementation techniques for a given data integrity requirement

In this module, you will learn

how to enforce business

rules and maintain data

integrity

! Overview of Data Integrity

This section discusses data

integrity rules and ways of

enforcing them

Data Integrity Categories

Entity Integrity (rows)

the allowable values for an attribute

an instance of an entity is unique

Referential Integrity

Data integrity refers to the consistency and accuracy of data An important step

in database planning is deciding how to ensure this integrity

Three categories of data integrity can be employed: domain, entity, and referential Here you will learn about domain and entity integrity (Referential integrity will be discussed in detail in the following topic.)

Domain integrity

Domain (or column) integrity specifies a set of legitimate data values for a column and determines whether null values are to be allowed Implementing validity checking can often enforce domain integrity Restricting the data type, format, or range of possible values allowed in a column can also enforce domain integrity

When used together, domain and entity integrity help to ensure that each entity

in a physical design maintains consistency When planning for databases in your design, you must consider consistency among entities and data validity within the entities as important additional aspects of data integrity

Slide Objective

To introduce the concept of

data integrity

Lead-in

Data integrity refers to the

consistency and accuracy of

data and falls into three

categories: domain, entity,

and referential

Referential Integrity

entity

The domain and entity integrity categories suffice for enforcing data integrity in

a single table; however, many databases contain multiple tables with parent and child entities that are related When both the entities in a relationship must exist and must have a valid link to one another, referential integrity makes sure that these conditions are met

Referential integrity ensures that the relationships between the primary keys (in the parent entity’s table) and foreign keys (in the child entity’s table) are always maintained

As a result of the restrictions that referential integrity imposes on tables, it is important for you to consider what to do when a primary key must be updated

or deleted If the primary key’s value in the parent entity is changed or deleted, then the instance of that entity has been modified, thus necessitating

corresponding modifications in all child entities that reference the parent By cascading, or walking, through all related child tables and updating or deleting data as necessary, the database can maintain referential integrity

Slide Objective

To explain the concept of

referential integrity

Lead-in

Whereas domain and entity

integrity help ensure

accuracy of data within a

table, referential integrity

helps ensure accuracy of

data between tables

Data Validation

It is important to have valid data Domain, entity, and referential integrity ensure that data exists where necessary and that entities follow a basic set of rules for existence None of these data integrity categories, however, can determine whether the data is valid You can ensure that the data is correct by checking it in the following ways:

" Range checking involves making sure the data’s value falls within a set of limits determined by the functional specification

" Data format checking involves making sure the data conforms to a set of format specifications, such as currency formatting, telephone number formatting, or length-of-character strings

" Data type checking involves making sure the data belongs to the proper data type and that an invalid data type is not used to represent the data when it is written to the database

To ensure that queries return the correct information, you must make sure that the information being stored is valid You can validate data in two ways:

" Client-side checks can ensure that invalid data is not posted to a database These checks can be coded as a script, as part of an executable program, or

as part of a Web page For example, if a field denotes a monetary value, a script can make sure that the information in that field is the proper data type and format

" Server-side checks, which are executed by the DBMS engine, can be implemented with field data types, default values, and constraints These checks can also ensure that invalid data is not stored in the database

Slide Objective

To introduce the concept of

data validation

Lead-in

Validity of data is of the

utmost importance This

topic discusses ways to help

achieve data validity

The methods or combination of methods that you use depends on the objectives and requirements of your solution; each method or combination has its value Although more difficult to maintain than server-side checks, client-side checks can help decrease the processing load on the DBMS server Although easier to maintain, server-side checks might present a bottleneck if the server has to validate many client requests

Identifying Data Integrity Requirements

must exist (produces domain and entity integrity)

requirements must exist (produces referential integrity)

Identifying data integrity requirements helps ensure that all logical and physical requirements for a solution are met and that the physical design supports the full intent of your specification

When attempting to identify the data integrity requirements of a solution, you must start with the data requirements specified during the solution’s design When examining the data requirements, you should identify areas within the requirements in which uniqueness, or limits and constraints, have been specified or deemed necessary to help ensure that entities can exist and be implemented correctly

If limits and constraints or limits have been specified or deemed necessary, you must determine whether the limits and constraints are bound to an entity or to a relationship between entities

For example, suppose you need to determine whether a particular task can exist without a related job number If it can’t, you must implement a data integrity requirement to make sure that no instance of that task is created without a corresponding job number Alternatively, suppose a consultant must submit a weekly timesheet for a client No timesheet can duplicate another timesheet, and no two timesheets can represent the same time or work performed by a consultant Therefore, a data integrity requirement exists for guaranteeing the uniqueness of timesheet information Each timesheet within the database must

be represented by a unique attribute or a combination of unique attributes Finally, you must implement referential integrity requirements to ensure that all relationships are maintained For example, if a customer wants to be removed from Ferguson and Bardell, Inc.’s database, a referential integrity requirement must ensure that the customer cannot be deleted as long as payment has not been received for any outstanding invoices

requirements helps validate

your physical design and

also ensures valid data

Activity 7.1: Identifying Data Integrity Requirements

In this activity, you will evaluate a set of data requirements from the Ferguson and Bardell, Inc case study to determine areas where possible data integrity issues might arise You will then formulate a set of data integrity requirements that must be followed by the development team for the Ferguson and Bardell, Inc solution

After completing this activity, you will be able to:

" Identify areas where data integrity and validity are required

" Specify data integrity requirements as they relate to data requirements

Slide Objective

To introduce this activity

Lead-in

In this activity, you will

determine data integrity

requirements for a solution

! Implementing Data Integrity Requirements

Slide Objective

To provide an overview of

the methods used to

implement data integrity

requirements

Lead-in

This section discusses ways

in which data integrity can

be implemented

Identifying Business Rules

thereby helping to determine solutions to events that can be predicted or automated

Business rules are the representation of logic specified within a solution When identifying business rules for a solution, you must reexamine the solution’s data requirements for the initial criteria that the rules need to satisfy You should also review any existing process documentation and use cases to determine what data-centric processes are currently being executed and what processes need to be executed in the future

It is important to consider the following three criteria when determining business rules:

" Conditions that must be satisfied for data to be considered valid Using the Ferguson and Bardell, Inc case study as an example, issues you might consider when determining applicable business rules include whether consultants can receive flat-level commissions even if they fail to bill for a period of time or whether a product can be shipped to a customer if no payment has been received

" Conditions that must be avoided Again using the Ferguson and Bardell, Inc case study as an example, issues you might consider include whether product pricing can fall below certain levels or whether inventory levels for products can drop below zero

" Sequences in which events must occur Using the Ferguson and Bardell, Inc case study as an example, issues you might consider include what to do when an inventory product falls below zero, when a new consultant is hired, or when payment for a contract is received

Slide Objective

To introduce methods for

identifying business rules

within data requirements

Lead-in

Business rules form a

foundation for encapsulating

logic from the data

requirements into the

solution

Deploying Business Rules and Data Integrity Requirements

of the database

required often determines which strategy is used

You can implement business rules and data integrity requirements by using two means of deployment:

" Directly in the database Most DBMSs provide automated processes that can be bound to tables and columns to enforce business rules and data integrity requirements

" Programmatically within or outside of the database You can code business rules and data integrity requirements into client applications, create components that can be deployed on an application server, or use programming logic and languages that are part of the database engine

The primary distinction between the two deployment methods is that the former involves rules that are automatically processed by the database engine, whereas the latter involves rules that must be called or invoked by an application’s logic Determining where to deploy rules is often based on the nature of the

programming logic required Most relational database engines now execute automatic methods that incorporate simple and commonly required application logic This logic applies to specific business rule tasks and resides in a single location It is also easily updated and efficient to execute If more complex logic

is required, it can be implemented as application code that can reside in any of the three service layers: user, business, or data You should choose your deployment strategy by using the same criteria with which you choose your development languages

Slide Objective

To introduce the different

ways in which business

rules can be deployed

Lead-in

Business rules can be

deployed in two ways: in the

database or

programmatically within or

outside of the database

Implementing in the Database

If you choose to enforce data integrity directly in the database, the DBMS will

do much of the work for you by using its automatic and built-in checks

Within the database engine, business rules and data integrity requirements can

be enforced through a set of criteria that data must either meet or not meet Usually, most of the data integrity requirements are implemented by using the automatic controls and properties of the database engine This automatic control involves the specification, or declaration, of the criteria at the time an object (such as a field, table, index, or key) is created

Enforcing data integrity through built-in database features has some distinct benefits: Integrity is enforced automatically, and the criteria do not have to be maintained or updated unless they change

You can use the following database features to enforce data integrity:

" Data types Setting appropriate data types for the fields in a database ensures that incorrect types of data can’t be entered into the table For example, if a data type of Date is specified, the database will reject any string or numeric value that is not in a date format Data types ensure that the format of the data is legitimate, but it cannot ensure that the value is valid

" Default values Default values specify the values to be used for a field when the values have not been explicitly supplied in INSERT statements By using default values

in your table definitions, you can ensure that a field value will be valid even when a user does not explicitly supply it

" Data validation rules Data validation rules encapsulate logic for specifying data values that are acceptable in a table’s fields They can specify the length of a field, an input mask, or a range of valid values

" Keys Most database engines can automatically monitor referential integrity between tables The primary and foreign key relationships from the physical model directly correspond to the database engine’s key settings These key settings automatically enforce referential integrity between linked tables

" Triggers Triggers are sets of programmatic statements that are explicitly defined for a particular table When a specific action (such as an insert, update, or delete) occurs in that table, the trigger causes the statements to automatically run For example, a trigger can verify data in other tables or perform an automatic update to a different table

Implementing Programmatically

You can use several methods to implement data integrity programmatically:

" Stored procedures Stored procedures are named collections of SQL statements that are stored

on the DBMS By using stored procedures, you can write one piece of code that performs an action, store it, and call it repeatedly as necessary

Stored procedures are useful for controlling changes to a table For example, instead of granting a user update rights to a specific table, you could allow updates to the table only through stored procedures By using the SQL code

in the stored procedure, you can determine whether the user’s change is appropriate and then either disregard or apply the change

The flow of a process can also be handled through the use of stored procedures For example, several tables could be changed or reordered by the code of a stored procedure

" Scripts Scripts can also be used to enforce data integrity within a database

Generally, you write a script to automate a process that is either inefficiently handled by the DBMS engine or impractical as a whole For example, you might write a script that maintains a database by importing a series of mainframe downloads, re-indexing the database, and then copying a report file to another file server

The script logic can be written in several languages Each script executes outside the memory space of the DBMS and might perform more tasks than simply manipulating data Scripts are generally used in a command-line environment or as part of a batch process that performs other tasks

Slide Objective

To introduce methods for

implementing data integrity