Introduction › Distributed Data Processing

Distributed Data Processing

As information requirements become increasingly complex, organizations need to find new ways to distribute processing power, application programs, and data. Distributed data processing allows distribution of application programs and data among interconnected sites to satisfy the information needs of the organization. Depending on its requirements, an organization may choose to centralize or decentralize its data processing systems.

Centralized data processing

In a centralized system, one machine controls all file access and updates.

A centralized system responds to the needs of the organization by permitting a high level of control over application programs and data.

A centralized data processing system is useful when:

• All data is shared across application programs.
• Many end users need access to the same data and also require the most current data available.
• Security has been established as the responsibility of the central site.
• Large volumes of data are involved. In this case, centralized control can help handle the volume of data efficiently.

In a centralized data processing environment, more attention typically is given to direct access storage devices (that is, their cost and reliability) than to data transmission.

The following illustration shows the way in which one machine controls all file access and updates in a centralized system.

Decentralized data processing

In a decentralized system, multiple machines control file access and updates to serve the varied needs of end users.

End users tend to have more control over application development and operations in a decentralized environment. A decentralized system is useful when:

• Data is primarily accessed at a single location. In this case, it is most efficient to store the data where it is used.
• The update rate is high. In this case, distribution of the data across different sites would permit more efficient update operations.
• Security has been established as a local responsibility.
• Data is used in highly specialized ways by end users.

In a decentralized data processing environment, more attention typically is given to data transmission costs and reliability than to direct access storage devices.

Two kinds of networks

Distributed data processing networks make it possible to combine the benefits of both centralized and decentralized systems. These networks can be viewed in two ways:

• Physical network—The combination of interconnected equipment (hardware) and programs (telecommunications access methods) used to transmit data between physical locations.
• Logical network—The combination of databases and application programs that handle an organization's information and processing requirements.

CA IDMS DDS allows distributed processing

CA IDMS DDS is designed to respond to the need for distributed processing. The remainder of this chapter presents an overview of concepts and network configuration.