|Thomas J Watson Jr and the IBM 360 Fifty Years On
|Written by Harry Fairhead
Page 2 of 2
With the rising tide of computers and automation the first of the many panics about the threat posed by such machines was beginning.
Documentaries that explained how robots would replace workers and how huge mechanical brains would do the thinking were alerting the public to what was going on.
At the same time Tom Watson's company had embarked on its biggest project ever - System 360. IBM had decided to risk its all on the largest private venture in US history. The project needed over 60,000 new employees and five new manufacturing plants all over the world at a cost of $5 billion - easily more than double what it cost the US government to develop the atomic bomb.
System 360 was ambitious and it broke new ground. A family of six processors was planned. Each processor was modular and could use the same range of peripherals and run the same software.
With this one family of machines IBM was gambling that it could capture the market - all markets. The machine would be suitable for scientific and business computing - it was just a matter of which processor and how many peripherals.
Before the 360 project IBM's machines had been one-offs. For example, the 701 was specifically designed for the sort of calculations that the military did - it was initially called the Defense Calculator. Other machines grew from this line as technological advances were incorporated into their design - the 705 had core memory and so on.
Univac had better machines that IBM but it was usually late in delivering them and its sales force were ineffective compared to IBM's. System 360 was intended to give their salesmen something to sell that was better.
A great many people were involved in designing and making system 360 a reality and so you can hardly credit Tom Watson alone for it - but he must have been the decision maker who decided to risk the company on it!
In 1961 a technical committee proposed a policy for IBM's next, or should it be first generation of machines. It proposed the single compatible family concept. This had obvious advantages but even in these early days IBM was worried that compatibility would allow other manufacturers to cash in on their success by making other compatible system components.
Of course there was also the problem that the new range would anger existing customers by being incompatible with the existing IBM range. The committee also recommended that the first machines be ready by 1964 to avoid early returns of IBM's increasingly obsolete models. Top management accepted the report and System 360 was born. Four of the mainframes would be built at Ploughkeepsie, one at Endicott and one at the Hursley research lab in the UK to complete the set.
IBM 360 control panel
The size of the project worried many people. John Backus, the man who invented FORTRAN for IBM, said that the task was too ambitious. It very nearly was. IBM could easily have floundered at any point during the building of the 360.
The system software was a particular problem. As well as a unified hardware architecture the system was to have a single operating system. This was the start of IBM's dominance of the operating system market - OS360 was for many years the standard.
However the software project was the biggest of its kind. Fred Brooks, one of the men in charge, later went on to write the book "The Mythical Man Month" as a result of his experiences of the OS360 project. It was the first time that it was realized that if one programmer takes a week to write some code two programmers do not take half a week to write the same code - in fact they are more likely to take two weeks!
System 360 was well behind schedule in 1963 and other companies were beginning to launch good machines. IBM looked as if it might not make it. They were also under attack by emulators which would allow customers to move IBM software to their own processors. Although the 360 project was well behind IBM announced it on April 7 1964 - they simply could not wait. They told the world about six new processors and forty peripherals - few of which were near completion and even fewer had been tested.
As time went on it was clear that they had problems - early manufacturing versions were malfunctioning and the manufacturing problems were not solved until 1965 - when the magnitude of the software problems became clear. It was 1967 before all of the 360 software was delivered and another two years before it was debugged and running well. It is estimated that OS/360 cost $500 million.
IBM System/360 Model 40
The machine that nearly broke IBM, the System 360, was an astounding success. By the end of the 1970s it was the standard computer to be found everywhere from large corporations to universities and hospitals. Its success forced most other computer companies out of business, leaving IBM as the giant of the mainframe world. IBM made billions of dollars from it and its direct descendants.
It is difficult to overestimate the impact of System 360 on computer users and programmers. There was the 360 and then there were a few other interesting machines that had specific advantages - for scientific computing say. The 360 range was what you took as a benchmark to compare others with. Its range of languages and applications made it the de facto standard.
The 360 family made a lot money, partly because IBM priced their processors by power and not by production cost. The reasoning went that if you wanted a processor that was twice as powerful then you should pay twice as much. This gave rise to popular rumours or folklore - such as the processor up-grade kit that was installed by an IBM engineer who turned up and just flipped an internal switch to make the processor go faster!
Another big feature of the 360 family was its use of a standard bus to connect perhipherals together. This was the first of the standard bus systems that gave rise to the S100, ISA, EISA and PCI buses and eventually to external interconnects such as SCSI, USB and so on. The 360 Channel started it all and even though it wasn't documented by IBM it didn't take long for other companies to reverse enginer it and create third party add-ons.
The 360 Channel Based Architecture
The 360 marked the start of a new phase of maturity in the computer world. It also brought with it some oddities of its own. IBM installations were generally run as mini versions of the big corporation itself. They were bureaucratic and authoritarian. Even the software was like this! Users learned a certain ethos, a way of doing things that provided something to revolt against in the early years of the microcomputer revolution.
You can only fight for freedom if you have been first in chains.
To be informed about new articles on I Programmer, install the I Programmer Toolbar, subscribe to the RSS feed, follow us on, Twitter, Facebook, Google+ or Linkedin, or sign up for our weekly newsletter.
or email your comment to: email@example.com
|Last Updated ( Wednesday, 09 April 2014 )