Classes of computers
Minicomputers (Midrange computers)
A minicomputer (colloquially, mini) is a class of multi-user computers that lies in the middle range of the computing spectrum, in between the smallest multi-user systems (see "Mainframe computers") and the largest single-user systems (microcomputers or personal computers). The contemporary term for this class of system is midrange computer, such as the higher-end SPARC, POWER, and Itanium-based systems from Sun Microsystems, IBM and Hewlett-Packard.
The term mainframe computer was created to distinguish the traditional, large, institutional computer intended to service multiple users from the smaller, single user machines. These computers are capable of handling and processing very large amounts of data quickly. Mainframe computers are used in large institutions such as government, banks, and large corporations. These institutions were early adopters of computer use, long before personal computers were available to individuals. "Mainframe" often refers to computers compatible with the computer architectures established in the 1960s. Thus, the origin of the architecture also affects the classification, not just processing power.
Mainframes are measured in millions of instructions per second or MIPS. An example of integer operation is moving data around in memory or I/O devices. A more useful industrial benchmark is transaction processing as defined by the Transaction Processing Performance Council. Mainframes are built to be reliable for transaction processing as it is commonly understood in the business world: a commercial exchange of goods, services, or money. A typical transaction, as defined by the Transaction Processing Performance Council, would include the updating to a database system for such things as inventory control (goods), airline reservations (services), or banking (money). A transaction could refer to a set of operations including disk read/writes, operating system calls, or some form of data transfer from one subsystem to another.
Modern mainframe computers have abilities not so much defined by their single task computational speed as by their redundant internal engineering and resulting high reliability and security, extensive input-output facilities, strict backward compatibility with older software, and high utilization rates to support massive throughput. These machines often run for years without interruption, with repairs and hardware upgrades taking place during normal operation.
A supercomputer is focused on performing tasks involving intense numerical calculations such as weather forecasting, fluid dynamics, nuclear simulations, theoretical astrophysics, and complex scientific computations. A supercomputer is a computer that is at the frontline of current processing capacity, particularly speed of calculation. The term supercomputer itself is rather fluid, and the speed of today's supercomputers tends to become typical of tomorrow's ordinary computer. Supercomputer processing speeds are measured in floating point operations per second or FLOPS. Example of floating point operation is the calculation of mathematical equations in real numbers. In terms of computational capability, memory size and speed, I/O technology, and topological issues such as bandwidth and latency, supercomputers are the most powerful. Supercomputers are very expensive and not cost-effective just to perform batch or transaction processing. Transaction processing is handled by less powerful computer such as server computer or mainframe. The architecture of modern supercomputers tend to be built around a large number of "off the shelf" CPU's or GPU's, rather than those of the past that tended to be built around a few specialized chips.