Mentors and Mainframe Cores

“Is that core?” I asked.

Posted on by

In my recent blog post, Software Tools and Punchcards, I shared a few stories about one of my early mentors, Bob Stearns. I can’t stress how much I learned from him and how much I strive to share that knowledge with others.

In a time when software was almost 100% written in assembler, COBOL, or FORTRAN, Bob introduced me to Pascal and C. He knew the IBM 3270 terminals inside and out. They ran on Intel 8080s and Bob showed me how to manipulate them to run our own software. He showed me how to create a simple 8080 assembler out of the IBM 370 macros too. I once visited Bob’s home and noticed an unusual doorstop between his kitchen and dining room. It was about two-and-a-half feet long and with clipped off wires along the sides. At first, I thought it was hollow. Then I thought it had a screen. Then I realized what it was.

“Is that core?” I asked.

“Yep,” Bob said. “It’s from the IBM mainframe I bought.”

I could see that there were small 3/4 inch frames all wired together. Inside was a matrix of small ferrite doughnuts, cores, that made the memory of the computer. This was back when you could actually see a bit!

I paused and then asked, “May I have a plane?” He simply handed me a pair of wire cutters. I carefully snipped the wires from the side of the topmost core plane, freeing it so I could take it home. It’s currently in a shadowbox, framed for all to see in our offices.

As I was clipping it loose, I realized that the plane I was clipping was damaged. Turns out Bob had stuck his finger down the small gap in the middle of the wires that hold the core. Instead of a square, there was a Bob-finger shaped hole. For a moment I considered asking for a core on the other end, but I had already started clipping. All these years later, I value the damaged core so much more than I would a pristine one because it was Bob who did the damaging.

I kept clipping, and Bob told about how the computer center had replaced this particular mainframe, an IBM 7094, with the then new System/360. They put the old machine up for bid and Bob won, taking it home for $360.

People were curious about what he was going to do with an old mainframe that required more power to run in an hour than his house used in an entire year. He said nothing, walked past them, removed the silver coated copper cables that hooked most of the machine together and went to the salvage yard, selling them for around $700.

The mainframe needed a very clean power source, so it used a device called a motor generator. It was a three-phase A/C motor hooked to a 400 cycle generator. Drawing from the power grid to turn a shaft, the motor generated power that was cleaner and more stable, than anything from Georgia Power.

The side panels of the 7094 were 1/4" stainless steel. He used them to bridge the gap between the truck he rented and the loading dock of the computer center when he collected the motor generator.

This scrap metal alone paid for the machine many times over, but Bob had more tricks up his sleeve, thanks to a friend in the chemistry department.

The machine had what seemed like thousands of small printed circuit boards, called Standard Modular System, each with a half dozen transistors. But what interested Bob and his chemistry friend were the edge connectors. Those were gold. Bob and his buddy came up with a way to “wash” off the gold from these connectors and purify the metal. All and all they extracted more than a pound of gold from the machine.

Not at all bad for a $360 investment.

Bob went on to follow the development of the new IBM PC, and later the Windows OS. I’d moved on to Unix in a big way, and left the University of Georgia, where Bob worked, and went to New Jersey and eventually California.

Decades later I moved back to Georgia and was delighted to run into Bob again. He’d since retired from UGA, but would meet all the old cronies at a local pizza place for an all-you-can-eat buffet. I began to make sure I was there every Thursday.

From our weekly lunches, I learned that his slightly slurred speech was due to an earlier stroke. His paralysis on one side of his body didn’t keep him from driving through.

I asked him if it was difficult to get renew his license with his paralysis, expecting to hear all the information one could possibly learn about state laws regarding strokes and driving, as was Bob’s way.

To my surprise, he didn’t know anything about those rules. When it was time to renew, he just walked into the DoT, and slapped down his old driver’s license without saying a word.

They just gave him a new one.

I miss Bob. I think of him often. These days I’m juggling NVMe, NVDIMMs, Ethernet storage protocols, accounting issues, business processes, CPU issues, programming in C, scripts, writing, planning, marketing, encouraging, teaching, and always refining the vision for Coraid. I’ve come a long way since my salad days of mainframes and minicomputers and Bob.

But it all rests on a foundation that was created by the generosity and sharing of a jovial systems programmer born in Elizabeth, NJ.

About the Author

Brantley CoileInventor, coder, and entrepreneur, Brantley Coile invented Stateful packet inspection, network address translation, and Web load balancing used in the Cisco LocalDirector. He went on to create the Coraid line of storage appliances, a product he continues to improve today.

Cheaper than the Cloud

Coraid Etherdrive SRX™ software s also easier than the cloud. It’s faster too. It builds out like the cloud, yet provides the security of on-premise data. Let me explain both how and why.

Learn How

←Previous | Blog Archive | Next →