Chapter 3 Distributed Data Processing Business Data Communications
Chapter 3 : Distributed Data Processing Business Data Communications, 4 e
Centralized Data Processing 8 Centralized computers, processing, data, control, support 8 What are the advantages? 8 Economies of scale (equipment and personnel) 8 Lack of duplication 8 Ease in enforcing standards, security ICSA 411: Data Communications & Computer Networking (Lawley) 2
Distributed Data Processing 8 Computers are dispersed throughout organization 8 Allows greater flexibility in meeting individual needs 8 More redundancy 8 More autonomy ICSA 411: Data Communications & Computer Networking (Lawley) 3
Why is DDP Increasing? 8 Dramatically reduced workstation costs 8 Improved user interfaces and desktop power 8 Ability to share data across multiple servers
DDP Pros & Cons 8 There are no “one-size-fits-all” solutions 8 Key issues 8 How does it affect end-users? 8 How does it affect management? 8 How does it affect productivity? 8 How does it affect bottom-line? ICSA 411: Data Communications & Computer Networking (Lawley) 5
Benefits of DDP 8 Responsiveness 8 Availability 8 Correspondence to Org. Patterns 8 Resource Sharing 8 Incremental Growth 8 Increased User Involvement & Control 8 End-user Productivity 8 Distance & location independence 8 Privacy and security 8 Vendor independence 8 Flexibility
Drawbacks of DDP 8 More difficulty test & failure diagnosis 8 More components and dependence on communication means more points of failure 8 Incompatibility of components 8 Incompatibility of data 8 More complex management & control 8 Difficulty controlling information resources 8 Suboptimal procurement 8 Duplication of effort
Reasons for DDP 8 Need for new applications 8 On large centralized systems, development can take years 8 On small distributed systems, development can be component-based and very fast 8 Need for short response time 8 Centralized systems result in contention among users and processes 8 Distributed systems provide dedicated resources
The DP “Pendulum” 8 Centralized systems (mainframes, etc) 8 Distributed systems (PCs) 8 Networked systems 8 Client-Server computing
Client/Server Architecture 8 Combines advantages of distributed and centralized computing 8 Cost-effective, achieves economies of scale 8 Flexible, scalable approach
Intranets 8 Uses Internet-based standards & TCP/IP 8 Content is accessible only to internal users 8 A specialized form of client/server architecture
Extranets 8 Similar to intranet, but provides access to controlled number of outside users 8 Vendors/suppliers 8 Customers
Distributed applications 8 Horizontal partitioning 8 Different applications on different systems 8 One application replicated on systems 8 Example: Office automation 8 Vertical partitioning 8 One application dispersed among systems 8 Example: Retail chain POS, inventory, analysis
Distributed data 8 Centralized database 8 Pro: No duplication of data 8 Con: Contention for access 8 Replicated database 8 Pro: No contention 8 Con: High storage and data reorg/update costs 8 Partitioned database 8 Pro: No duplication, limited contention 8 Con: Ad hoc reports more difficult to assemble
Networking Implications 8 Connectivity requirements 8 What links between components are necessary? 8 Availability requirements 8 Percentage of time application or data is available to users 8 Performance requirements 8 Response time requirements
- Slides: 15