One of the mainframe guys I used to work with in the late 90's had a photo of one of the platters from the hard disk embedded in the server room wall, this was one of those early IBM beasts with 24 inch platters that had a catastrophic failure.
One of the mainframe guys I used to work with in the late 90's had a photo of one of the platters from the hard disk embedded in the server room wall, this was one of those early IBM beasts with 24 inch platters that had a catastrophic failure.
I suspect that was a very old photograph, because I heard a similar story in the 1960s. The drive was the IBM 1301, a very large disk drive. According to the story, if a head crashed in the outer track, it provided enough torque to snap the arm holding the head and hurl it out of the drive with enough force to beak the class door and embed itself in the wall. We ran out two IBM 1301s close to the wall and facing it, so if something like that happened only the wall would be damaged.
I heard a story many years ago of a tech who went to investigate a cascading failure in a row of washing machine hard drives. A single platter had failed, which damaged the head. They thought maybe it was just the pack which was faulty, so they put in a new pack, and destroyed that as well. So they thought maybe it was the drive which was failing, so they put the first failed pack in another drive, and destroyed the head on that as well. Their attempts to figure out what was wrong ended up killing three or four disk packs and as many drives.
When the tech turned up, he ran his finger around the inside of the mouth of the drive, and his finger came out covered in red dust. "What's that?" the customer asked. "Data," he replied.
I heard a simpler version of that story, involving the IBM 2311. An operator installed an IBM 1316 disk pack on a drive, but the operating sytem did not recognize it. He moved the pack to another drive but it still was not recognized. After trying all available drives, unsuccessfully, he called for IBM Field Service, who determined that the defective disk pack had damaged all of the drives.
Considering that moving the disk pack to another drive was the only thing the operator could do, this may have happe
One of the mainframe guys I used to work with in the late 90's had a photo of one of the platters from the hard disk embedded in the server room wall, this was one of those early IBM beasts with 24 inch platters that had a catastrophic failure.
I started my career on an IBM 1410 equipped with one of these. You put up with the risk and the expense ($115,000 a month in 1963 dollars) because it held en entire 25 megabytes of data. Nothing else came close.
In the late 60's, our research lab bought a huge six-platter disk drive, with the platters mounted vertically, like wheels. It took 2 3-phase motors to get these things going, after which one would drop out. We calculated that if the case and building were to dissolve suddenly, the disks had enough momentum to roll over the Bay Area coastal range and into the ocean. Ultimately, the device was heavily damaged when the super-careful and super-snobby technician left a tool in it and scratched half the surfaces
In the late 60's, our research lab bought a huge six-platter disk drive, with the platters mounted vertically, like wheels. It took 2 3-phase motors to get these things going, after which one would drop out. We calculated that if the case and building were to dissolve suddenly, the disks had enough momentum to roll over the Bay Area coastal range and into the ocean. Ultimately, the device was heavily damaged when the super-careful and super-snobby technician left a tool in it and scratched half the surfaces. We used the rest for a couple years more. I think the lawsuit is still making its way through the courts.
This sounds like the Librascope disk at the Stanford AI Project. I left at the end of 1968 but still heard about the calculation. I had not heard about the tool damage--thank you for that.
I wrote a diagnosic program for that disk. It was my second experience with unreliable disks, the first being with the swapping disk on the PDP-1. Les Earnest claimed that the Librascope disk would lose data when the temperature changed by two degrees.
I judge a religion as being good or bad based on whether its adherents
become better people as a result of practicing it.
- Joe Mullally, computer salesman
hard drive (Score:5, Interesting)
Re:hard drive (Score:4, Interesting)
One of the mainframe guys I used to work with in the late 90's had a photo of one of the platters from the hard disk embedded in the server room wall, this was one of those early IBM beasts with 24 inch platters that had a catastrophic failure.
I suspect that was a very old photograph, because I heard a similar story in the 1960s. The drive was the IBM 1301, a very large disk drive. According to the story, if a head crashed in the outer track, it provided enough torque to snap the arm holding the head and hurl it out of the drive with enough force to beak the class door and embed itself in the wall. We ran out two IBM 1301s close to the wall and facing it, so if something like that happened only the wall would be damaged.
Re: (Score:2)
Re:hard drive (Score:5, Funny)
I heard a story many years ago of a tech who went to investigate a cascading failure in a row of washing machine hard drives. A single platter had failed, which damaged the head. They thought maybe it was just the pack which was faulty, so they put in a new pack, and destroyed that as well. So they thought maybe it was the drive which was failing, so they put the first failed pack in another drive, and destroyed the head on that as well. Their attempts to figure out what was wrong ended up killing three or four disk packs and as many drives.
When the tech turned up, he ran his finger around the inside of the mouth of the drive, and his finger came out covered in red dust.
"What's that?" the customer asked.
"Data," he replied.
Re: (Score:2)
I heard a simpler version of that story, involving the IBM 2311. An operator installed an IBM 1316 disk pack on a drive, but the operating sytem did not recognize it. He moved the pack to another drive but it still was not recognized. After trying all available drives, unsuccessfully, he called for IBM Field Service, who determined that the defective disk pack had damaged all of the drives.
Considering that moving the disk pack to another drive was the only thing the operator could do, this may have happe
Re: (Score:2)
One of the mainframe guys I used to work with in the late 90's had a photo of one of the platters from the hard disk embedded in the server room wall, this was one of those early IBM beasts with 24 inch platters that had a catastrophic failure.
I started my career on an IBM 1410 equipped with one of these. You put up with the risk and the expense ($115,000 a month in 1963 dollars) because it held en entire 25 megabytes of data. Nothing else came close.
Re: (Score:1)
Re: (Score:3)
In the late 60's, our research lab bought a huge six-platter disk drive, with the platters mounted vertically, like wheels. It took 2 3-phase motors to get these things going, after which one would drop out. We calculated that if the case and building were to dissolve suddenly, the disks had enough momentum to roll over the Bay Area coastal range and into the ocean. Ultimately, the device was heavily damaged when the super-careful and super-snobby technician left a tool in it and scratched half the surfaces. We used the rest for a couple years more. I think the lawsuit is still making its way through the courts.
This sounds like the Librascope disk at the Stanford AI Project. I left at the end of 1968 but still heard about the calculation. I had not heard about the tool damage--thank you for that.
I wrote a diagnosic program for that disk. It was my second experience with unreliable disks, the first being with the swapping disk on the PDP-1. Les Earnest claimed that the Librascope disk would lose data when the temperature changed by two degrees.