Networked Computing | Research & Encyclopedia Articles

Networked Computing

Networked computing has evolved from networks connecting computers and distributed computing through thin-client network computing in the 1980s and 1990s to Internet computing today. Centralized computing--which started with mainframes at the outset of the computing evolution, evolved to PCs, and lasted up to the 1980s--focused mainly on automating existing processes. This approach gave way as applications and systems evolved to the client-server model and the distributed computing model. The client-server model allowed for the creation of server applications responsible for the storage, analysis, and sorting of large amounts of data on central data servers, with connected workstations responsible only for front-end client applications capable of running queries, producing reports, and adding new records. Distributed computing meant that dedicated file servers also could be used throughout an internal Local-Area Network (LAN), each supporting a single application, such as e-mail, facsimile, data storage, graphics storage, or documentation storage. Computers, sometimes using different operating systems, could store different components of a single application, such as a spellchecker and thesaurus stored on different computers. Printers and other peripherals connected to the LAN offered users shared access over the network. Network monitoring and backups offered a high degree of data security and integrity. Personal computing entered a new era of productivity as spreadsheets, database programs, and word processors came increasingly into use at personal workstations.

The concept of network computing had its origins in 1995 with Oracle Corporation's concept of full-service client-server computing, where individual computers would rely on central servers on an internal network for all applications, databases, and information. Desktops would be minimally equipped with application-specific thin clients intended primarily to communicate with servers that would be responsible for all data processing. (In contrast a fat client would comprise a stand-alone PC workstation fully equipped with all requisite applications.) This model of network computing allows for lower costs of installation, security, and maintenance, since databases and applications would need to be maintained only on the central server, not at every workstation. This model has been found particularly useful for task-oriented professionals in engineering, medicine, accounting, and real-estate focused on content rather than technology, as well as in back-office operations supported by clerical and sales staff. One disadvantage to this model is that as clients increase in number, servers need to be upgraded given the additional computing burden placed on them, thus raising infrastructure and maintenance costs.

Networked computing has evolved with the advent and progress of the Internet to mean Internet computing. This means that computers linked over the Internet would access large central servers outfitted with pre-packaged server software and Internet database applications for all processing purposes. Oracle8i, developed by Oracle Corporation to replace thin-client computing in 1998, offers such a platform; it functions as an Internet database that supports Java, and it can consolidate data, Java objects, and Windows files. Java is important because it is being used as a foundation to develop interactive, real-time exchange of information by allowing dynamic querying of databases and managing logical workflow processing. Java applets and servlets, which are compact Java programs supported and run by larger applications such as the major Internet browsers (e.g., Netscape and Explorer) are increasingly being used to streamline online banking, interactive trading, investment planning, and online purchasing. Through the use of Java applets and servlets Web users need not download whole workflow applications and can thereby save space, memory, and hardware wear-and-tear at their desktops. In this model, computing is focused on facilitating the search and dissemination of information. Applications are distributed across the network, which means that various functions can be accessed through the network as though they were stand-alone components, dynamically loaded and discarded when used.

Internet computing also means that business data is consolidated onto large servers for global Internet access, easy management, and speedier business application usage. Remote access computing is supported through remote access servers and concentrators. The Internet computing model is valuable to businesses because it can lower computing costs without the complexity of general-purpose operating systems. Computers with different operating systems would utilize Internet protocols as they accessed, computed, and communicated over the Internet. This model is thought to combine the best of the mainframe and client-server worlds by centralizing backups and offering users an intuitive graphical interface. In effect, client-server computing would expand to allow computing via TCP/IP over the Internet, allowing the use of different systems and platforms as long as a Web browser with a graphical interface provided a point of entry for a Web user. Additionally, mobile and remote computing becomes possible in this model, with all users required only to have access to the Internet and a Web browser installed on their machines.

In fact, the trend toward mobile and other computing devices is expected to change the profile of Internet computing in the next 10 years, with an increase in the usage of handhelds, wireless, and miniature "wearable" computers. Increased reliance on Internet computing will lead to data storage, synchronization, and scheduling being consolidated at centralized server farms that connect the corporate office and remote workers through broadband links. The concept of building hard-wired applications for functions such as inventory and accounting will evolve into a more dynamic reliance on toolkits for a particular task, as is already evident in the proliferation of shareware, freeware, and discountware on the Internet. More than ever before, a menu of software tools is now available to complete any project. These tools, while occasionally downloaded to the desktop, are increasingly expected to reside on the network.

Futurists predict that PCs will soon be replaced by network computers, Internet appliances (from telephone-like devices to televisions), and non-desktop computers, such as enterprise servers and wearable computers. As business transactions increasingly take place online, issues such as security, authentication, and quality of service (QoS) are expected to be resolved. Constant wireless connectivity and virtual private networks will allow users to be connected anytime, anywhere to their corporate nets over the Internet. As wireless communications expands, the convergence of video, voice, and data will allow users to download books, movies, television, and radio signals to their portables over broadband wireless connections. Computers will not only be everywhere, they will be held centrally on a network and controlled remotely. You could go to work, connect to your home server and control devices such as security cameras, motion detectors, lights, alarm clocks, heating and cooling systems, refrigerators, microwaves, PCs, televisions, and VCRs remotely.