Classification of Computers (According to Size)

According to Size

Micro Computers

A microcomputer is a complete computer on a smaller scale and is generally a synonym for 
the more common term, personal computer or PC, a computer designed for an individual. A 
microcomputer contains a microprocessor (a central p rocessing unit on a microchip), memory 
in the form of read-only memory and random access memory, I/O ports and a bus or system of 
interconnecting wires, housed in a unit that is usually called a motherboard. In an ascending 
hierarchy of general computer sizes, we find:
• An embedded systems programming computer, which is embedded in something and 
doesn't support direct human interaction but nevertheless meets all the other criteria of a 
microcomputer
• Microcomputer
• workstation, as used to mean a more powerful personal computer for special applications
• minicomputer, now restyled a "mid-range server"
• mainframe or mainframe computer, which is now usually referred to by its manufacturers 
as a "large server"
• Supercomputer, formerly almost a synonym for "Cray supercomputer" but now meaning a 
very large server and sometimes including a system of computers using parallel 
processing
• A parallel processing system is a system of interconnected computers that work on the 

same application together, sharing tasks that can be performed concurrently.

Mini Computers

A minicomputer, a term no longer much used, is a computer of a size intermediate

between a microcomputer and a mainframe. Typically, minicomputers have been standalone

computers (computer systems with attached terminals and other devices) sold to

small and mid-size businesses for general business applications and to large enterprises

for department-level operations. In recent years, the minicomputer has evolved into the

"mid-range server" and is part of a network. IBM's AS/400e is a good example.

Mainframes

Mainframe is an industry term for a large computer, typically manufactured by a large 

company such as IBM for the commercial applications of Fortune 1000 businesses and other 

large-scale computing purposes. Historically, a mainframe is associated with centralized rather 

than d istributed computing. Today, IBM refers to its larger processors as large servers and 

emphasizes that they can be used to serve distributed users and smaller servers in a 

computing network.

Super Computers

A supercomputer is a computer that performs at or near the currently highest operational rate 

for computers. A supercomputer is typically used for scientific and engineering applications 

that must handle very large databases or do a great amount of computation (or both). At any 

given time, there are usually a few well-publicized supercomputers that operate at the very 

latest and always incredible speeds. The term is also sometimes applied to far slower (but still 

impressively fast) computers. Most supercomputers are really multiple computers that perform 

parallel processing. In general, there are two parallel processing approaches: symmetric 

multiprocessing (SMP) and massively parallel processing (MPP).

Perhaps the best-known builder of supercomputers has been Cray Research, now a part of 

Silicon Graphics. Some supercomputers are at "supercomputer center," usually university 

research centers, some of which, in the United States, are interconnected on an Internet 

backbone known as vBNS or NSFNet. This network is the foundation for an evolving network 

infrastructure known as the National Technology Grid. Internet2 is a university-led proje ct that 

is part of this initiative.

At the high end of supercomputing are computers like IBM's "Blue Pacific," announced on 

October 29, 1998. Built in partnership with Lawrence Livermore National Laboratory in 

California. Blue Pacific is reported to operated at 3.9 teraflop (trillion operations per second), 

15,000 times faster than the average personal computer. It consists of 5,800 processors 

containing a total of 2.6 trillion bytes of memory and interconnected with five miles of cable. It 

was built to simulate the physics of a nuclear explosion. IBM is also building an academic 

supercomputer for the San Diego Supercomputer Center that will operate at 1 teraflop. 

It's based on IBM's RISC System/6000 and the AIX operating system and will have 1,000 

micropro cessors with IBM's own POWER3 chip. At the lower end of supercomputing, a new 

trend, called clustering, suggests more of a build-it-yourself approach to supercomputing. The 

Beowulf Project offers guidance on how to "strap together" a number of off-the-shelf personal 

computer processors, using Linux operating systems, and interconnecting the processors with 

Fast Ethernet.

Applications must be written to manage the parallel processing.