Because applications are often developed independently in file processing systems

Data: Facts, text, graphics, images, sound and video segments that have meaning in the users’ environment

Database: An organized collection of logically related data. By organized, we mean, “structured so as to be easily stored, manipulated, and retrieved by users. By related we mean that the data describe a domain of interest to a group of users and that the user can use the data to answer questions concerning that domain

Data Versus Information: Information is data that have been processed in such a way as to increase the knowledge of the person who uses the data. In practice, a database may contain either data or information, or both. Also, data are often preprocessed and stored in summarized form in databases that are used for support.

Metadata: Data that describe the properties or characteristics of other data. Some of these properties include data definition, data structures and rules or constraints. An example of metadata is the heading of a database that describes what each column (or field) holds. The metadata of a field that holds “Andrew”, “Bernard”, “Christopher”, “David”, et cetera could be “FirstName”, for example.

Traditional File Processing

To be useful for business applications, computers must be able to store, manipulate, and retrieve large files of data. Computer file processing systems were developed for this purpose. We explore file processing systems for two reasons:

1. they are still widely used today, especially for backing up database systems
2. understanding the problems and limitations inherent in file processing systems can help us avoid these same problems when designing database systems

Disadvantages of File Processing Systems

Program Data Dependence. File descriptions are store within each application program that accesses a given file. Consequently, any change to a file structure requires changes to the file descriptions for all programs that access that file.

Duplication of Data. Since applications are often developed independently in file processing systems, unplanned duplicate data files are the rule rather than the exception. This duplication is wasteful since it requires additional storage space and increase effort to keep the files up to date. Duplicate data files often result in loss of data and/or metadata integrity.

Limited Data Sharing. With the traditional file processing approach, each application has its own private files and users have little opportunity to share data outside their own application.

Lengthy Development Times. With the traditional file processing systems, there is little opportunity to leverage previous development efforts. Each new application required that the developer essentially design new files formats and descriptions, and then write the file access logic for each new program.

Excessive Program Maintenance. The preceding factors all combine to create a heavy program maintenance load in organizations that rely on traditional file processing systems. In fact, as much as 80% of the total IS budget may be devoted to program maintenance in such organizations.

The Database Approach

Emphasizes the integration and sharing of data throughout organizations.

Enterprise Data Model. A graphical model that shows the high-level entities for an organization and the associations or relationships among those entities. An entity is an object or concept that is important to the business. The entities of a business could include CUSTOMER, PRODUCT, EMPLOYEE, CUSTOMER ORDER, and DEPARTMENT.

The relationships between the entities are often captured in entity-relationship diagrams, and are explored extensively in Chapters 3 and 4.

Note the following characteristics of the enterprise data model:

1. It is a model of the organization that provides valuable information about how the organization functions, as well as important constraints
2. It stresses the integration of data and processes by focusing on entities, relationships, and business rules

Relational Databases

Views all data in the form of tables that represent the entities identified in an enterprise data model. Each column of the table represents an attribute (or characteristic) of an entity.

An important property of the relational model is that it represents relationships between entities by values stored in the columns of the corresponding tables.

Implementing Relational Databases

• Use a local area network (LAN) that workstations to a database server
• To improve intra-company communication and decision making, install an Intranet that uses Web-based Internet technology and is only accessible within the company
• Each workstation may then be used as a Web browser, gaining fast access to company information, or as a personal computer and connected to the database server as needed through the LAN.
• Add a Web interface to [company] business applications, such as order entry, so that more internal business activities can be conducted through the Intranet
• Become more accessible to customers through the Internet (these raise information security questions (Ch12) and platform issues (Ch10)
• Build a separate database that contains historical and summarized information. (such a database is usually called a data warehouse, or data mart; also analysts need specialized decision support tools to query and analyze database. One class of tools used for this purpose is called on-line analytical processing (or OLAP) tools (Ch11))

A Database Application

An application program (or set of related programs) that is used to perform a series of database activities (create, read, update or modify, and delete) on behalf of database users.

The Range of Database Applications

Personal Databases

• Residing on PCs (including laptops) and PDAs, personal databases are designed to support one user
• Widely used to improve personal productivity. Caveat: Data cannot easily be shared with other users. This illustrates a common problem: If data are of interest to one person, the probably are (or will soon become) of interest to others as well.

Workgroup Databases

A workgroup is a relatively small team of people who collaborate on the same project or application or on a group of similar projects or applications. A workgroup database is designed to support the collaborative efforts of such a team.

Here, each member of a workgroup could be linked by means of a LAN to a central database server. Different types of group members (e.g. developer or project manager) may have difference user views of the shared database. (This arrangement however introduces data management issues not present with personal—single-use—databases, such as data security and data integrity with concurrent user data updating. Also there is the question of which database processing operations should be performed at the workstation and which should occur at the server (that of client- versus server-side processing).

Department Databases

A department is a functional unit within an organization, and department databases are designed to support the various functions and activities of a department. Of the sleuth of questions that must be addressed when designing and implement department databases, a particularly important one is that of whether or not to create a distributed database (if the users of the database are geographically dispersed or the size of the database is so great that the data must be stored on several computer systems).

Enterprise Databases

An enterprise database is one whose scope is the entire organization or enterprise (or, at least, several departments), and are usually intended to support organization-wide operations and decision making. Two major developments have resulted from the evolution of enterprise databases:

1. Enterprise resource planning (ERP) systems
2. Data warehousing implementations

Enterprise resource planning (ERP): A business system that integrates all functions of the enterprise, such as manufacturing, sales, finance, marketing, inventory, accounting, and human resources. ERP systems are software applications that provide data necessary for the enterprise to examine and manage its activities. Evolved from manufacturing resource planning (MRP) and MRP-II systems of the 1970s and 1980s.

Data warehouse: An integrated decision support database whose content is derived from the various operational databases. Data warehouses (Ch11) provide users with the opportunity to work with historical data to identify patterns and trends and answers to strategic business questions.

Internet, Intranet, and Extranet Databases

...

Advantages of the Database Approach

Program-Data Independence. The separation of data descriptions (metadata) from the application program that use the data: data descriptions are stored in a central location called the repository. Hence, data database systems allow an organization' s data to change and evolve (within limits) without changing the application programs that process the data.

Minimal Data Redundancy. The design goal with the database approach is that previously separate (and redundant) data are integrated into a single logical structure.

There are time when it might be desirable to introduce some limited redundancy to improve database performance, as we will see.

Improved Data Consistency. By eliminating data redundancy, we greatly reduce the chances for inconsistency. Further, we avoid the wasted storage space resulting from redundant data storage.

Improved Data Sharing. A database is a shared corporate resource. Authorized users are granted permission to use the database, and each user (or group of users) is provided one more more user views.

A user view is a logical description of some portion of the database that is required by a user to perform some task. A user view is usually a form or report that comprises data from more than one table.

Increase Productivity on Application Development. There are two important reasons that database applications can often be developed more rapidly and less expensively that conventional file applications:

1. Assuming that the database and the related data capture and maintenance applications have already been designed and implemented, the programmer can concentrate on the specific functions required for the new application without having to worry about the file design or low-level implementation details.
2. The DBMS provides a number of high-level productivity tools such as forms and report generators and high-level languages that automate some of the activities of database design and implementation.

Enforcement of Standards. ...the database administration function should be granted single-point authority and responsibility for establishing and enforcing data standards, including naming conventions, data quality standards, and uniform procedures for accessing, updating, and protecting data...

Improved Data Quality. The database approach provides a number of tools and processes to improve data quality. Two of the most important are:

1. Database designers can specify integrity constraints that are enforced by the DBMS (Ch 3 & 4).

A constraint is a rule that cannot be violated by database users.

1. One of the objectives of a data warehouse environment is to clean up (or "scrub") operational data before they are placed in the data warehouse.

Improved Data Accessibility and Responsiveness. With a relational database, end users without programming experience can often retrieve and display data using queries (SQL, Ch7 & 8).

Reduced Program Maintenance. As data are independent of the application programs that use them, in the database approach, we can change either the data or the application programs without necessitating a change in the other factor.

Does the file processing programs are data independent?

What are the advantages of file processing system?

What are the two major weaknesses of file processing systems?

Is often developed by identifying a form or report that a user needs on a regular basis?