From Mainframes to Smartphones (9780674286559) by Campbell-Kelly Martin & Daniel D. Garcia-Swartz

Author:Campbell-Kelly, Martin & Daniel D. Garcia-Swartz
Language: eng
Format: epub
ISBN: 9780674729063
Publisher: Harvard Univ Pr

SCIENTIFIC WORKSTATIONS AND THE CLIENT-SERVER COMPUTER MODEL

By the early 1990s, the mainframe and minicomputer businesses of companies such as IBM and DEC were feeling the effects of a rise in alternative computing modes: the client-server paradigm and powerful scientific workstations. From the mid-1960s to the mid-1980s, typical computer networks were organized such that a central mainframe did all the computing and several “dumb” terminals facilitated input-output. Beginning in the early 1980s, the development of microcomputers and powerful workstations enabled the diffusion of more complex topologies. Among these was the client-server model, featuring personal computers or workstations (“clients”) that were attached to one or more powerful “server” computers within a local area network. The servers provided services—file storage, database access, and running enterprise applications, among others—for the clients.

Although the first client-server networks relied on the UNIX operating system and on high-performance servers, less expensive PC-based client-server schemes proliferated in the late 1980s. These were clusters of IBM-compatible personal computers connected to one or more low-cost Intel-based servers. At the software level, the clients typically used MS-DOS or Windows and the server a Novell Netware network operating system. Over time, the client-server model replaced the traditional mainframe-based computing scheme in many American corporations. In a 1994 Datamation survey, for example, 58 percent of corporations that had traditionally used mainframes were implementing client-server computing. A year later, the share had grown to 67 percent.7

Just as client-server networks competed with traditional mainframes, scientific workstations began to displace minicomputers. Workstations relied on a high-performance microprocessor to make the power of a scientific mainframe available within a desktop form factor. The first workstation company was Apollo, founded in Chelmsford, Massachusetts, by Bill Poduska, one of the cofounders of the Prime minicomputer company. In 1981, Apollo Computer delivered its first product, a machine that sold for $40,000 and up and used a proprietary operating system.8 Apollo soon faced competition from Sun Microsystems, a company established in Silicon Valley in 1982 by Stanford University graduate students Andy Bechtolsheim, Vinod Khosla, and Scott McNealy. Sun benefited from technology transfer from the academic world: the hardware was derived from the Stanford University Network workstation project, and the operating system was the University of California at Berkeley’s variant of UNIX, usually known as BSD UNIX. (Bill Joy, who had played a key role in the development of BSD UNIX, joined the company soon after its founding and is considered a cofounder.) Sun advocated open standards, a strategy that worked well against Apollo and HP, both of which remained committed to their proprietary operating systems. By the end of the 1980s, Sun Microsystems dominated the workstation market, and HP, which had acquired Apollo, remained far behind.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.