Skip to the content ForoHackers
Log inJoin

Hi, I am here to talk about bits that change on their own

Computer architectureIntroductionsReliability engineering
#985 · bitflip_betty · · Report

Hello. I work on memory reliability: ECC, error rates, the whole "what happens when a bit decides to flip" field. Yes, cosmic rays are real. No, not every crash is a cosmic ray, and when someone says "must have been a bit flip" I ask for the logs.

At some point I will write up how a SECDED code works, because it is lovely and nobody explains it well. Happy to answer questions in the meantime.

Fun fact to start: at ground level, most memory errors in a data centre come from faulty hardware (a weak cell, a bad connector) rather than from particles. The universe is less to blame than your DIMM.

#1334 · null_terminated_hopes · · Report

Excellent. I have a collection of undefined behaviour I blame on cosmic rays. Good to know most of it is simply me.

#1335 · bitflip_betty · · Report

Statistically, yes, it is usually you. But I respect the optimism of hoping for a particle.

#1336 · r3l1c_0f_bbs · · Report

Parity bits on old RAM, a single "parity error" message and the machine froze. I did not know that was the polite ancestor of what you work on. Looking forward to the write-up.

Log in or join to reply.