Evolution of IT Architecture
The co-evolution of information technology architecture over the past 40 years has indeed dramatically influenced the concept of enterprises. IT is no longer simply a tool to support back-office transactions; it has become a strategic business component, enabling the reengineering of business processes, redesigning of the organizational structure, and redefinition of markets and industries
Strategies and policies that guide the development and deployment of IT within the organization must be defined, enforced, and continuously reviewed as part of an ongoing process of developing and monitoring business strategy. Also the organization must understand that IT architecture is only one of the many components of IS infrastructure. The other components of ISI are business processes, organizational structure and culture, people, and management controls. The organization must recognize the interrelationship between all these components while designing and implementing IS infrastructure.
Information Systems Architecture
The IS architecture has evolved from the mainframe era to the client– server and distributed component (Internet) era.
- Mainframe Era (1960s to 1970s)
The IS architecture in the mainframe era was based on centralized computing and data processing using proprietary hardware and software technology. The entire software is installed in the central host computer. Users interact with the host through a terminal that captures key strokes and sends that information to the host. The mainframe software architectures are not tied to a hardware platform.
- Client-Server Era (1980s to 1990s)
The client–server architecture may be defined as an organization of computer applications on a network in which one computer (the client) can request for a process or data from another computer (the server) attached to the network. The term is a functional, not physical, description. In other words, the two computers may alternate their roles as ‘client’ and ‘server’. Binding is the process whereby the association between the client and the server occurs. Binding may be dynamic, which means that the client may find the appropriate server through the network directory service if the server is registered. The typical client–server model divides an application into server-side and client-side components. Client–server applications typically distribute the components of an application so that the database would reside on the server, the user interface would reside on the client, and the business logic would reside on either of them.
- Web-based Client Server
Most of the enterprise application software such as SAP R/3, PeopleSoft, Oracle, Baan, I2 technology, SIBEL, etc. have been developed using client–server technology and are now translating to Web-based computing. In this case the application software resides entirely on a network server and the client computer only needs a standard browser. The browser is a ‘thin client’ unlike the ‘fat client’ of client–server computing.
The browser provides a window to the server-based applications. To upgrade or modify application software, customers need to make changes only on server-based applications rather than upgrading hundreds or thousands of copies on individual PCs. The benefit of Web-based computing is that users do not have to learn to navigate multiple interfaces and the cost and administrative burden of keeping each desktop in the firm updated are reduced because fewer applications need to be stored and maintained. Web-centric applications also allow larger number of clients to be served simultaneously.
- Distributed Compononet Era (2000 and beyond)
The next generation technology is the emergence of distributed component architecture, which promises to be the most significant shift in corporate computing environment since the move from monolithic enterprise systems. In particular, the shift towards componentized, packaged applications— large-scale suites that combine component-based software with highly integrated functionality—heralds a new era in the design, implementation, maintenance, and upgrade of corporate information systems.
These new application suites promise highly configurable application systems that match the distributed functionality of real-world business processes more closely than any previous architecture did. It is these real-world applications of componentization that make this new technology especially germane to the dynamic, fast growing companies. This architecture takes the concept of muti-tier client–server to its natural conclusion. Instead of differentiating between business logic and data access, the distributed component system model simply exposes all functionality of the application as objects, each of which can use any of the services provided by other objects in the system, or even objects in the other system.
