Mainframe computers are critical for some of the largest corporations in the world. Each mainframe has more than one modern processor, RAM ranging from a few megabytes to multiple-score gigabytes, and disk space and other storage beyond anything on a microcomputer. A mainframe can control multiple tasks and serve thousands of users every second without downtime.
The chief difference between mainframes and other computing systems is the level of processing that takes place. Mainframes are also different in terms of data bandwidth, organization, reliability, and control. Big organizations-banking, healthcare, insurance, and telecom companies, etc.-use mainframes for processing critical commercial data.
In this article, we discuss the evolution of mainframe computers and their components.
History of mainframe computers
IBM developed a critical part of mainframe computing, the Automatic Sequenced Controlled Calculator (ASCC) for arithmetic operations, in 1944. From the late 1950s through the 1970s, several companies manufactured mainframes: IBM, Burroughs, RCA, NCR, General Electric, and Sperry Rand, for example. Since then, System / 390 by IBM is the only kind of mainframe in use. It evolved from IBM's System / 360 in 1960.
An Early mainframe occupied a huge space. New technologies have drastically reduced the size and cost of the hardware. A current-generation mainframe can fit in a small closet.
Components of a modern mainframe computer
Like a PC, a mainframe has many components for processing data: operating system, motherboard or main board, processor, controllers, storage devices, and channels.
• Motherboard: The motherboard of a mainframe computer consists of a printed circuit that allows CPU, RAM, and other hardware components to function together through a concept called "Bus architecture". The motherboard has device slots for input cards and cable interfaces for various external devices. Where PC motherboards use 32- or 64-bit buses, mainframes use 128-bit buses. General instructions regarding the internal architecture help the motherboard connect to the other devices and retrieve data using binary computation.
• Processor: A CPU acts as the central processing point in mainframe architecture and includes an Arithmetic Logic Unit (ALU) for performing arithmetic calculations. It also works as a controller for the bus architecture and handles traffic and data requests. The processing power of mainframes is much higher compared to PCs, so that they can handle huge amounts of data.
• Storage devices: Storage devices are for entering, retrieving, storing, and recording data. Many are external devices, such as hard drives, tape drives, and punch card readers, all connected to terminals of the mainframe and controlled by the CPU. Their capacity for data storage can be hundred or even thousands of times that of a PC.
• Communication controllers: Communication controllers allow remote computers to access a mainframe. With the help of networks, LAN or WAN, communication controllers establish connections with various devices, perform data transmission over communication channels, and keep track of users at terminals.
• Channels: The "channels" are the cables used to connect the CPU and the main storage to other parts of the system and make sure that data is moved in a systematic way without losing its integrity.
Modern mainframes have advanced features such as expanded service management capabilities, cross-platform integration facilities, etc. And so are suitable for critical data center operations. The cost of maintaining modern mainframes is much less compared to older models.