IBM System/360

The IBM System/360 is a family of mainframe computer systems announced by IBM on April 7, 1964, and delivered between 1965 and 1978. System/360 was the first family of computers designed to cover both commercial and scientific uses and a complete range of applications from large to small. The design distinguishes between architecture and implementation, which allowed IBM to release a group of compatible designs at different prices. Most models use microcode to implement the instruction set, and feature 8-bit byte addressing and fixed-point binary, fixed-point decimal, and hexadecimal floating-point calculations. The System/360 family introduced IBM's Solid Logic Technology (SLT), which packs more transistors onto a circuit card than previous technologies, allowing more powerful but smaller computers to be built.

What is the aim of this site?

The Background

I was an Air Traffic Engineer with the UK’s National Air Traffic Services (NATS). NATS came into being as an Air Traffic Control Service provider as part of the UK Civil Aviation Authority (CAA) when the latter was formed in 1972. The CAA is broadly equivalent to the Federal Aviation Administration (FAA) in the US.

I was initially trained to operate and maintain such aviation-related equipment as radar, navigation aids, and communication systems. The basic training also included courses on “computer techniques”. I was posted to the London Air Traffic Control Centre, located on the RAF West Drayton site in West London – a few miles north of Heathrow Airport. This NATS centre was comparable to an ARTCC in the US.

In the early 1970s NATS purchased, via the FAA, an IBM 9020 air traffic control computer complex which was installed at the RAF West Drayton centre. This system consisted of various components drawn from the IBM System/360 family. In 1979 I was selected to attend a year-long 9020 engineers’ course at the FAA Academy in Oklahoma City. We were trained to IBM Customer Engineer level on all the constituent 9020 elements. As well as the computers and input/output channels the training covered peripherals such as disk and tape drives and high-speed printers.

On return from Oklahoma, I became a member of one of the “9020 Watches” which looked after the system on a 24/365 basis as part of the centre’s watchkeeping roster system. After several years I was promoted to Head of 9020 Engineering and moved to a management role in charge of all aspects of 9020 hardware support, alongside a colleague who was responsible for 9020 software support. In this role I managed the decommissioning and disposal of the 9020, and the introduction of its replacement system (another IBM mainframe complex) from an operational point-of-view. I stress “operational” as I was just a cog in the machine. There were tens of ‘project people’ far cleverer then me, but they needed someone with operational experience and guidance.

Because all hardware engineering support of the 9020 was in-house we had access to far more information than the usual IBM customer. For example, we had copies of hardware schematics and logic diagrams down to component level. On a day-to-day basis we were entirely independent of IBM, except that we ordered replenishment spare parts from them. Speaking of which, the initial procurement included a “recommended spares” package of thousands of individual items. Most of these remained on the shelves until they were scrapped along with the main system!

The Restoration

Well, the main system wasn’t entirely scrapped. The combined 9020 operational (triplex) and support (simplex) systems included a total of eight IBM System/360 mainframe computers. Seven of these were scrapped (“there’s gold in them there computers!”) but one was saved from the hands of the metal reclaimers.

That mainframe computer was an IBM System/360 Model 65. It differs from a standard Model 65 in that it has several 9020-specific features added. It has everything a base Model 65 has, but also some bells and whistles. All commercial S/360 mainframes had a range of additional features a customer could order. However, the 9020 features could only be ordered through government contract, as the 9020 was supplied by the IBM Federal Systems Division rather than the commercial IBM Data Processing Division.

After being safely cosseted in secure storage for over three decades I am in the process of restoring that Model 65 to working condition, a process that has only just commenced (late 2024). It is to my advantage that, as well as being CE-trained on the system, I was the one who personally dismantled it in 1990, making copious notes. I also have all the necessary documentation and many spare parts, should they be necessary.

The Objectives

This site will not try to be “all you wanted to know about System/360”. IBM’s S/360 family is so iconic and revolutionary that there are many existing excellent sources of information. Instead, it will focus on technical detail directly related to the restoration of this particular Model 65. The objectives are twofold:

1 – Out of historical interest, document the challenges and solutions encountered during the restoration. As well as (hopefully!) being of interest in themselves, they may also prove useful when maintaining the restored system in years to come.

2 – To enable future maintenance procedures to be successfully undertaken by others. This involves making all the relevant technical information available. But more than this, to show how the various S/360 maintenance procedures are actually put into practice. It is one thing to have a maintenance or theory of operation manual and understand the words. It is another to find one’s way around a hundred thousand (literally) ’scope points in practice and know what one is looking for at each point, for example. CE training wasn’t just ‘reading the books’. We were shown how to do things. For now there are still some CEs, like me, who have that knowledge. But not for long…

As the second objective implies, I hope the restored mainframe will live on in working condition well into the future. Indeed, given the technology (IBM's legendary 'built like a battleship' back in the day) and the component-level information available, I don’t see why ‘for ever’ isn’t a possibility. 

The Apology

Of necessity, the information on this site will evolve in a piecemeal fashion. I will try to develop a sensible page/article structure as I go along. I have several other projects on the go, and I think it is more important to ‘do it’ and progress rather than ‘write it up’ – which is probably why I was in Operations and not Projects!

I expect updates will follow the same pattern as the practical restoration. I will try and post things in chunks as I go along, but please bear with me.