You are currently on IBM Systems Media’s archival website. Click here to view our new website.

Bookmark and Share

Recent Posts

Mainframe Computers Created a Sense of Wonder in Me—Part 3

October 20, 2013

In my previous posts, I wrote about the importance of Assembler Language and system organization to my understanding of the mainframe computers. I also wrote about two other topics—the key role of access methods and interrupts as a mechanism to implement multitasking. In this post, I want to discuss two more attributes of mainframes that really got me thinking and motivated to learn more—timing facilities and virtualization.

Useful Timing Facilities

The Principles of Operations book explains timing systems including the time-of-day (TOD) clock, interval timer, CPU timer and clock comparator. Some of these timing-related systems were a bit abstract to me but I really found the TOD clock handy. With the STCK instruction, I could store a value from the system that I could later translate into a precise point of time including hour, minute, second, month, day, year. I could use this value to capture the moment in time when something happened or compare two values returned by STCK that would make it possible to compute the elapsed time between two events. You can easily see the usefulness of this instruction, and the fun and discovery surrounding capturing the clock value, translating it, and comparing and doing math. Also, there were a lot of utilities already built that let you immediately start to work with the eight-byte TOD value.

Did (or do you) have such an interest in the system clock or the other timing facilities?

Growth of Virtualization

When I first worked with mainframes, virtualization was already in place. I somehow knew that my COBOL programs ran in an address space and that address spaces were logical arrangements with some of the program in real memory and other parts needed to be made real when referenced, and that I did not have to worry about it. I could also see that there were legacy access methods that were upgraded to use virtual storage and new ones like VSAM and VTAM that were “wired” from the start for virtual storage.

When later on I became a system programmer, I became aware of LPARs, which made my late-night, weekend testing possible as I installed maintenance and implemented new facilities on our company mainframe. Later, as a developer for IBM, I came into contact with VM and used it extensively for our testing and development of the CICS, IMS and OPC features of System Automation for OS/390. Over my career, I experienced the importance of virtualization from many viewpoints—COBOL programmer, system programmer and product developer.

From then (1979) until today, I still have a sense of wonder about how virtualization works. It’s useful, ubiquitous and a modern miracle of computing. Did you have a similar experience with virtualization? Please comment on your own experiences.

Posted October 20, 2013| Permalink