Mainframe Computers

Mainframes are often described as big, powerful computers that are used in big organisations where they manage & process huge volumes of data. Mainframes are mostly used is government & big organizations for critical applications where continuous processing of huge amount of data is required. Some examples of these type of applications are enterprise resource planning (ERP), measuring census, banking & financial transactions, etc. Most of the original mainframes architecture & basic designs were developed around 1960's but still they are evolving today to suit the market needs. We often hear about another term called supercomputers that are used for determining weather patterns or in nuclear research. We often get confused between these two terms. But we need to remember that while supercomputers are required where complex analysis and more computational speeds & accuracies are required, mainframes are used for handling high volume/traffic of data. In technical terms mainframes performance are measured in millions of instructions per second (MIPS), whereas supercomputer performance are measured in floating point operations per second (FLOPS). An instruction may be read/write whereas floating point operations are addition, subtraction, multiplication or division.

Mainframes have higher reliability, availability & serviceability which are its defining characteristics. these three characteristics are very important because mainframes are used in those applications which should have very minimal or no downtime. In older mainframes, there was no interactive interface. But, today most of the mainframes support either fully or partially web based user interfaces. The new generation mainframes require much less cooling, power consumption & space. From business perspective, IBM dominates the mainframe markets. Some examples of mainframes are IBM zSeries, System z9 and System z10 servers.