5MG Harddrive

[Bill Bridges]

Loading a 5MG harddrive in 1956.

[Ray Tackett]

Sure looks like a drum memory to me, including what I believe are two of the three "sticks" of heads visible in the photo. I'll take a step farther out on the limb by guessing it was 0.5 meg, not 5 meg.

In 1968, I had my hands on an IBM 0.5M hard disk drive which also needed a forklift, but was roughtly 2.5 feet square by 4.5 feet long. The head positioners were hydraulic.

Ward, what say you?

[Ward Miller]

Ray, that is/was an IBM 305 RAMAC 50 disks. Data stored alphmerically, not binary. First IBM machine with random access disks, 1405. Maximum seek time, inner ring on top disk to inner ring on bottom disk 800ms How's that for speed?

I'll comment further tomorrow.

[Russell Holton]

I'm wondering about the "packing". It seems like the guts are exposed, and even if bolted to the pallet, that's not much protection. I wouldn't think you've be shipping any drive you intended to use that way. OTOH, I'm not sure why they'd be hauling "used' equipment on a airplane like that. That doesn't strike me as a era when "hauling your trash out" was a thing.

[Ward Miller]

Loading the harddisk is not a lengthy job if you are using a forklift, as in that picture. Unloading it can be an entirely different picture -- pun intended.

My IBM 305 RAMAC developed a bad bearing shortly after it was put in service. All of those 50 disks sit on the motor at the bottom. It was direct drive. They had not done this before (it was one of the first machines), so the engineers said I would have to backup all the data currently on it. The 305 only had punched card input and output, so it took about 8 (EIGHT!) hours to punch the data into cards. Something like 5 CASES of cards!

Then they built a tripod over the whole machine and lifted the disks enough to remove and replace the motor. Before going from the punched cards back to the disks, I had them test it The disks were in perfect registration, so we could immediately begin to use it and did not need those many cases of punched cards!

[Ward Miller]

Another IBM 305 RAMAC story from the late '50s.

We had had machine rooms for our punched card machines for decades. Some "experts" insisted the floor finish contain grit so the operators would not slip and fall. But as computers began replacing the card machines, we found the grit was creating particles unfriendly to the new computers. So the "experts" laid vinyl floor tiles in my machine room. Most of the machines could be moved a few inches so the tiles could be installed, but the 305 weight in at over a ton, so they just laid the tiles around the machine's castors, figuring they would come back later and have the moving crew move it.

Now these tiles were about 1/8" thick and the steel castors designed for rolling the 305 around were maybe 4" in diameter. When they tried to roll the machine up that 1/8" and onto the tiles, they found it was too heavy. Eventually, they struggled with a bunch of crowbars and lifter the machine up that 1/8".