Super Computers
Super Computers are the largest and fastest computers available. Generally, they are not used for commercial data processing. Super Computers have huge amount of memory and high processing speed. They can process up to one billion instructions per second. Super Computers can process 64 bits or more at a time. Their processing speed ranges from 10,000 million instructions per second (MIPS) to 1.2 billion instructions per second. They can support up to 10,000 terminals at a time.
Super Computers can recover automatically from failures (fault tolerance). Conventional computers have single processor which can process one instruction at a time. But super computers have multiple processors (or CPUs) that process multiple instructions at a time. This is called parallel processing. In olden days high clock rate was considered as the main difference between ordinary computers and super computers. Now the distinguishing feature of super computers is their high degree or parallelism i.e. supercomputers consist of several processors, which can process simultaneously.
Milestones: The first super computer was the ILLIAC IV made by Burroughs. Other suppliers of super computers are CRAY, CDAC [PARAM 2000 FROM INDIA], Fujitsu, Intel Corporation, Thinking Machine Corporation, Hitachi, IBM and Sun Microsystems, etc.
Where Super Computers are used?
- These computers are used in specialized areas such as Defense, aircraft design, computer generated movies, weather research etc.
- In medical field, super computers are used to study the structure of viruses, impact of various drugs on human beings etc.
- Designing an aircraft involves simulating (=to create) and analyzing the airflow around the aircraft. This again requires a super computer.
Mainframe Computers
Mainframes are less powerful and cheaper than Super computers. However, they are big general-purpose computers and they can handle all kind of scientific and business applications. Mainframes can process several million instructions per second. Mainframes can be used for a variety of applications. A typical application of these computers is airline reservation or railway reservation system. It can support many powerful devices. (E.g. a main frame can support more than 1,000 remote terminals.)
Mainframes have large secondary storage capacity. Variety of peripheral devices like magnetic tape drives, hard disks, visual display units, plotters, printers and telecommunication devices can be connected to main-frame computers. They have high-speed cache memory which helps to process data at a faster rate than mini or microcomputers. They also offer the facility of multi-programming.
Prices of Mainframe computer ranges from 1 crore to 5 crores depending upon the configuration. Major suppliers of mainframe computers are IBM, HP, Sun Microsystems, Honey well, Burroughs, NCR, CDAC and Sperry etc. Large mainframes can have hundreds of MBs of main memory and hundreds of GBs of secondary storage.
Mini Computer
It is a relatively fast, small and inexpensive computer. The input/output capabilities are some what limited. Data processing is similar to that of mainframe but on a small scale. The cost of minis is low. Usually data is input by means of a keyboard. As the name implies, a minicomputer is small, compared to a mainframe. It is also called as scaled-down mainframe. The processor and the peripherals are physically smaller.
Minicomputers cost from Rs.5 lacs to Rs.50 lacs. The most popular minicomputers or minis are Data General Nova, IBM series/1. These computers can support multiple number of users. Small to medium sized organizations use minicomputers for their data processing activities. Primary storage capacity starts at about 640KB and can go up to few MBs. Programming languages include BASIC, PASCAL, COBOL, C and FORTRAN.
Minicomputer systems can support most of the input/output (I/O) devices and secondary storage devices that large mainframe systems can handle. They are also used to develop distributed data processing systems. Instead of having large mainframe computer, a company may install mini-computer at each remote location and connect them to each other through telecommunications.
Micro Computer
A microcomputer is a full-fledged computer system that uses a microprocessor as its CPU. Microprocessor is a small chip which may vary from fingernail size to postage-stamp size. Microcomputers were first available in 1970s. The first microcomputer designed for personal use was ‘Altair’.
PC Vs. Micro Computer
The term PC has a more specific meaning. In 1981, IBM named its first microcomputer as IBM PC. Within few years, many companies copied the IBM design, which can function like IBM PC’s and are called IBM clones or IBM compatible computers. Thus the term PC includes IBM PC and its compatibles. On the other hand Apple Macintosh computer is neither an IBM PC nor its compatible. It belongs to another family of microcomputers made by Apple computers.
Manufacturers: Currently IBM & Apple are the 2 most popular manufacturers of microcomputers.
Elements of Micro Computers: A microcomputer may consist of the following elements:
- 16, 32 or 64 bit processor,
- Internal memory 256 MB expandable to 512 MB and more
- Backing storage-cassette, floppy disc, microfloppy discs, micro-drive, silicon disc or hard disc, CD-ROMS, DVDs, pen drives etc
- Keyboard and screen (input and output)
- Interface (for the connection of peripherals)
- Bus (communication and control channels)
- Printer and/or plotter (multicolour text and graphics)
- Pulse generator (clock)
- Light pens, mouse, paddles/joysticks, Multimedia (graphics and games)
- Software (programs)
Usage: Microcomputers are useful even to the smallest businesses. However their primary market is the personal home computer market.
Workstation
Between minicomputer and microcomputers – in terms of processing power – there is a class of computers known as WORKSTATIONS. A workstation looks like a personal computer and is typically used by one person. Workstations are still more powerful than PC. There are 2 major differences between a workstation and a PC. Internally, workstations are constructed differently than microcomputers. They are based on different architecture of CPU called Reduced Instruction Set Computing (RISC), which results in faster processing of instructions.
On the other hand most microcomputers can run any of the four major operating systems – DOS, Unix, OS/2 and Microsoft Windows NT. But workstations run on Unix operating system. A disk less workstation does not have floppy disk drive. Data storage is provided by centralized hard disc. This provides control over data storage because it reduces the possibilities of data theft.
Manufacturers: The biggest manufacturer of workstations is Sun Microsystems. Other manufacturers include IBM, DEC, Hewlett Packard, Silicon Graphics, DELL etc.
Conclusion: Many people use the term workstation to refer to any computer or terminal that is connected to another computer. But this is an outdated term. These days, a workstation is powerful RISC based computer and is generally used by scientists and engineers.
Server
A server is a computer system that provides services to other computers in a network, called clients or workstations. Servers occupy a place, similar to that of mini computers. But today servers have largely replaced minicomputers. A typical server is a computer system that operates continuously in a network and serves the requests of other computers in the network. Servers may be broadly classified as dedicated and non-dedicated.
A dedicated server is completely reserved for the purpose of serving other computers and no other activity can be performed on such servers. These types of servers are preferred when there are large number of nodes on the network. On the other hand a non dedicated server is not completely reserved for this purpose i.e. it can also be used simultaneously for other purposes. This kind of servers are generally preferred by small organizations. For e.g., in a small office, a large desktop computer may act as both workstation and as a server.
Today, Servers are physically similar to most other general-purpose computers. But their hardware configuration is optimized to meet the requirements. In many cases the hardware will be similar to that of a standard desktop PC. However, servers run software that is very much different from that used on desktop computers and workstations. Generally servers are used to host (= to provide or to share) hardware resources on a controlled and shared basis to client computers such as printers (print servers) and file systems (file servers). This sharing permits better access control (and thus better security) and can reduce costs by reducing hardware duplication.
Servers should not be confused with mainframes. Mainframes are very large computers that centralize certain information processing activities in large organizations. They may or may not act as servers in addition to their other activities. Many large organizations have both mainframes and servers. Usually, servers are smaller and much more numerous and decentralized than mainframes.
