When did mainframe computers come into being? Mainframe development occurred in a series of generations starting in the 1950s. In those days, mainframe computers were not just the largest computers; they were the only computers and few businesses could afford them.

First generation systems, such as the IBM® 705 in 1954 and the IBM 1401 in 1959, were a far cry from the enormously powerful machines that were to follow, but they clearly had characteristics of mainframe computers. These computers were sold as business machines and served then— as now— as the central data repository in a corporation's data processing center.

In the 1960s, the course of computing history changed dramatically when mainframe manufacturers began to standardize the hardware and software they offered to customers. The introduction of the IBM System/360™ (or S/360™) in 1964 signaled the start of the third generation: the first general purpose computers. Earlier systems such as the 1401 were dedicated as either commercial or scientific computers. The revolutionary S/360 could perform both types of computing, as long as the customer, a software company, or a consultant provided the programs to do so. In fact, the name S/360 refers to the architecture's wide scope: 360 degrees to cover the entire circle of possible uses.

The S/360 was also the first of these computers to use microcode to implement many of its machine instructions, as opposed to having all of its machine instructions hard-wired into its circuitry. Microcode (or firmware, as it is sometimes called) consists of stored microinstructions, not available to users, that provide a functional layer between hardware and software. The advantage of microcoding is flexibility, where any correction or new function can be implemented by just changing the existing microcode, rather than replacing the computer.

With standardized mainframe computers to run their workloads, customers could, in turn, write business applications that didn't need specialized hardware or software. Moreover, customers were free to upgrade to newer and more powerful processors without concern for compatibility problems with their existing applications. The first wave of customer business applications were mostly written in Assembler, COBOL, FORTRAN, or PL/1, and a substantial number of these older programs are still in use today.

In the decades since the 1960s, mainframe computers have steadily grown to achieve enormous processing capabilities. The New Mainframe has an unrivaled ability to serve end users by the tens of thousands, manage petabytes of data, and reconfigure hardware and software resources to accommodate changes in workload— all from a single point of control.