News for hackers. Stuff that matters.
Re:ECC matters by slumdev on Sunday January 03, 2021 @02:15PM (#25624223)
Error-correcting code (the "ECC" in ECC) is just a quorum at the bit level.
Re:ECC matters by eevilspock on Sunday January 03, 2021 @02:46PM (#25624529)
I'm surprised that the other replies don't grasp this. This is the proper level to do the quorum.

Doing quorum at the computer level would require synchronizing parallel computers, and unless that synchronization were to happen for each low level instruction, then it would have to be written into the software to take a vote at critical points. This is going to be greatly detrimental both to throughput and software complexity.

I guess you could implement the quorum at the CPU level... e.g. have redundant cores each with their own memory. But unless there was a need to protect against CPU cores themselves being unreliable, I don't see this making sense either.

At the end of the day, at some level, it will always come down to probabilities. "Software engineering principles" will never eliminate that.

Re:ECC matters by slumdev on Sunday January 03, 2021 @03:02PM (#25624706)
I would highly recommend a graduate-level course in computer architecture for anyone who thinks ECC is a 1980s solution to a modern problem.

There are a lot of seemingly high-level problems that are solved (ingeniously) in hardware with very simple, very low-level solutions.

Re:ECC matters by bollu on Sunday January 03, 2021 @04:10PM (#25625273)
Could you please link me to such a course that displays the hardware level solutions? I'm super interested!
Re:ECC matters by slumdev on Sunday January 03, 2021 @04:18PM (#25625339)
Re:ECC matters by andrewaylett on Sunday January 03, 2021 @04:16PM (#25625324)
https://en.wikipedia.org/wiki/NonStop_(server_computers)

My first employer out of Uni had an option for their primary product to use a NonStop for storage -- I think HP funded development, and I'm not sure we ever sold any licenses for it.

Re:ECC matters by sobriquet9 on Sunday January 03, 2021 @03:08PM (#25624767)
Modern error correction codes can do much better than that.