Mainframe computerMainframes (often colloquially referred to as "big iron") are large, powerful, and expensive computers used mainly by large companies for bulk data processing (such as bank transaction processing). The term arose during the early 1970s with the introduction of smaller computers such as the DEC PDP series, which became known as minicomputers, so users coined the term "mainframe" to describe larger, earlier types. Description Mainframe computers' abilities are not so much defined by their CPU speed as by their massive internal memory, large, high-capacity external storage, fast high-throughput I/O, high-quality internal engineering and resulting proven reliability, and expensive but high-quality technical support. These machines can and do run successfully for years without interruption, with repairs taking place whilst they continue to run. Mainframe vendors offer such services as off-site redundancy—if a machine does break down, the vendor offers the option to run customers' applications on their own machines (often without users even noticing the change) whilst repairs go on. The internal redundancy of these computers can be such that, in at least one reported case, technicians could move one from one site to another by disassembling it piece by piece, and reassembling it at the new site, whilst leaving the machines running. The switchover in this example took place entirely transparently. Often, mainframes support thousands of simultaneous users who gain access through "dumb" terminals. Some mainframes have the ability to run (or "host") multiple operating systems and thereby operate not as a single computer but as a number of "virtual machines". In this role, a single mainframe can replace dozens or hundreds of smaller PCs, reducing management and administrative costs while providing greatly improved scalability and reliability. The reliability is improved because of the hardware redundancy noted above, and the scalability is achieved because hardware resources can be reallocated among the "virtual machines" as needed. This is much harder to do with PCs, because adding or removing hardware resources often requires the machine to be taken offline, and the hardware limitations are much more restrictive. When running as the host for many "virtual machines" a mainframe can provide the raw power for which they have always been valued, but also the flexibility provided by PC networks. Currently, IBM mainframes are dominant in the market, with Hitachi, Amdahl, and Fujitsu also producing machines. Prices start at several hundred thousand dollars. History Several manufacturers produced mainframe computers in the 1960s and 1970s; in the 'glory days' it was "IBM and the Seven Dwarves": Burroughs, Control Data, General Electric, Honeywell, NCR, RCA, and Univac. But shrinking demand and tough competition caused a huge shakeout in the market—RCA sold out to Univac and GE also left; Honeywell was bought out by Bull, Univac merged with Sperry to form Sperry/Univac, which was later merged with Burroughs to form Unisys Corporation in 1986 (so-called "dinosaurs mating"). In 1991, AT&T briefly owned NCR. Companies found that servers based on lower-cost microcomputer designs could be deployed at a fraction of the cost and offer local users much greater control of their own systems, and "dumb terminals" used for interacting with mainframe systems were gradually replaced by personal computers. Consequently, demand plummeted and mainframe installations were restricted mainly to financial institutions with massive data processing requirements. For a while, there was a consensus among industry analysts that the mainframe was a dying market, as mainframe platforms were increasingly replaced by personal computer networks. This trend started to turn around in the late 1990s as corporations found new uses for their mainframes, since they can offer web server performance similar to that of hundreds of smaller machines, but with much lower power and administration costs. Another factor currently increasing mainframe use is the development of the Linux operating system, which is capable of running on many mainframe systems, either directly or, more commonly, in a virtual machine. This allows mainframes to take advantage of the software and development expertise and communities from the PC market. Comparison with supercomputers The distinction between supercomputers and mainframes is not a hard and fast one, but generally one can say that supercomputers focus on problems which are limited by calculation speed while mainframes focus on problems which are limited by Input/Output and reliability. As a consequence: * Supercomputers typically exploit massive parallelism, often with thousands of processors, while mainframes have a single or a small number (up to several dozen) of processors. * Because of the parallelism visible to the programmer, supercomputers are quite complicated to program; in mainframes, the limited parallelism (if present) is usually hidden from the programmer. * Supercomputers are optimized for complicated computations that take place largely in memory, while mainframes are optimized for simple computations involving huge amounts of external data accessed from databases. * Supercomputers tend to cater to science and the military, while mainframes tend to target business and civilian government applications.