1. 3

I guess what I am asking is for a book that covers the history of infrastructure in general but mostly focused on the disruption VMs and now containers meant for the infrastructure industry.


  2. 3

    I’m going to guess that the lack of responses means ‘no’. If you really want to dig into this, the place to start is VM/370. OS/360 was a massive change to the industry, as the first implementation-independent ISA. Before then, every new computer had come with a new instruction set and required all software to be ported. This was a cost sink, but not a huge problem because computers were really expensive and programmers were much less expensive. IBM offered a range of different-sized computers that could run the same programs in 1960.

    VM/370 separated out the computer from the rest of the org chart. With VM/370, you could buy a single large mainframe and partition it into smaller VMs, rather than buying one minicomputer per group and having a load of spare capacity in small slices across the company. Honestly, modern cloud computing is basically the same idea as this, 50 years later. IBM even had a model of renting the mainframe rather than buying it and paying only for the cores / disk that you were using.

    1. 1

      Thanks, David! Would you know about any book that covers these days at IBM? Or I you suggesting I read the docs!?

      1. 3

        I don’t know of a book. When I was doing research for my Xen book, I started with Popek and Goldberg and read a load of technical manuals for VM/370. A lot of the material related to them is on the computer history museum web site and the press material from IBM in the ’60s and ’70s is fantastic. They had a target market that knew almost nothing about computers and so had to explain things in simple terms.

        A couple of years ago, I had the enormous pleasure of attending an alumni dinner at the college where I was a Bye Fellow and sitting next to an alumna who was recruited by IBM in the ‘60s. She’d taken their aptitude test (back then, because there were no relevant degrees for most of what they were doing, IBM had aptitude tests for all recruitment paths) for administrative jobs because her boyfriend had applied as an engineer. He didn’t get the job, she was asked to take the technical version of the aptitude test and ended up working on the system design for the 370.

        1. 1

          What a great story! Thanks for sharing. I wonder if those press materials are digitalized.

          1. 2

            They certainly were (on computerhistory.org), but their search engine doesn’t find them now - I guess it doesn’t search their archives? I only spent 30 seconds looking though, so you’ll probably have better luck if you poke around a bit.

            1. 1

              I’ll ask the folks at IBM if they know anything.