You get a bonus - 1 coin for daily activity. Now you have 1 coin

Mainframe

Lecture



Mainframe The term is derived from two English words. Main - main or main. Frame - a frame, a skeleton the basis of something. To begin, we give a definition.

A mainframe is the main computer of a data center with a large amount of internal and external memory. It is designed for tasks requiring complex computational operations.

But initially, this term understood the rack where the processor is located. The first generation of computers, occupying entire halls, consisted of huge racks. Each of which implemented a specific component computers. Processor - took the whole rack. Today, the entire processor is located on a single integrated circuit. But at that time, the rack with the processor was called the main one, that is, the mainframe.

Mainframe specifications

  • Vertical scaling is formed by increasing the number of processors in one system.
  • Horizontal scaling is implemented by combining powerful computers into one logical computer. You can combine almost any number of computers. They can be, geographically, remote from each other.
  • Highly reliable data storage. Mainframe is provided by using RAID arrays. Hot-swappable. There is a round-the-clock data backup.
  • Hot swap, virtually any component. Disks, RAM and even processors.
  • Mainframe have high error control. Instant testing of data in memory when using them. Duplication of communication channels with peripheral equipment.
  • Cryptographic data protection at the hardware level.
  • Centralized data storage. Gives high reliability and relevance.


The development of mainframes began in the 60s. At that time, IBM, thinking about the creation of powerful and versatile computers, offered the System / 360 computer to the market. At that time, for IBM, it was one of the most expensive projects. The new computer itself has proven itself so well that some experts began to call it a mainframe. It can be argued that the very first idea of ​​mainframes, appeared together with the computer UNIVAC.

Until today, the undisputed leader in the production of mainframes is IBM. From the System / 360 series to the latest zSeries series. The first mainframes of this series are zSeries 800, 890, 900, 990.

In the summer of 2005. IBM announced the release of new zSeries family of “Z” family. Z9 - was a very good example of this family, which is still in use today. Z9 is a direct descendant of z990, and has high performance (up to 600 MIPS), memory (up to 512 Mb) and can work with 60 disk partitions. The machine has a multi-chip module which contains 16 processors. IBM spent about $ 1.2 billion on the Z9 project.

On the Z9 computer, so-called logical partitions LPARs are used. This allows you to simultaneously run multiple operating systems on one mainframe, including Z / OS, OS / 390, Linux, z / VM. All large database management systems are serviced by such machines. Including CISC, WebSphere Application Server, DB2 and Oracle. As one of IBM's leading analysts, John Abbott, points out, "mainframes are still the best platforms for customers that require 100 percent data integrity, but they need open source software."

The first network technologies began to be used on mainframes. SNA (Systems Network Architecture) - system network architecture. It was developed by IBM back in 1974. And allowed to integrate mainframes into global computer networks.

Especially for mainframes are created operating systems. That allows you to fully manage the entire computing power of these machines. z / OS is a 64 bit server operating system. Used exclusively on mainframes. Its monolithic core began to design in the 60s. At that time, in terms of scaling, it significantly surpassed all existing operating systems. Like many mainframe products, z / OS was developed by IBM. Today, z / OS remains a highly relevant proposition for the mainframe market. She already knows how to support Java and the Unix software specification.

By the early 1990s, interest in mainframes was significantly lower. They gradually began to crowd out cluster systems. They allow distributed to store and process data. That increases their reliability before mainframes. Improve the performance of distributed systems, much cheaper than centralized. These key factors have led to a significant reduction in the number of mainframes worldwide. However, for a certain class of tasks, building a centralized data warehouse, it justifies itself both in terms of performance and in economic terms. Therefore, some enterprises build their computing centers on the basis of the mainframe.


Comments


To leave a comment
If you have any suggestion, idea, thanks or comment, feel free to write. We really value feedback and are glad to hear your opinion.
To reply

History of computer technology and IT technology

Terms: History of computer technology and IT technology