3 ms·
ECC ram will either correct the error or can raise a non maskable interrupt if it can't be corrected.
by sn 10y ago
ECC ram will either correct the error or can raise a non maskable interrupt if it can't be corrected.
- Retric 10y agoECC vastly lowers, but does not remove this problem. You can for example get bit flip errors while doing the checksum. As a personal user it's a non issue but scale things to ~70k devices and you get ~1.7 million device hours per day.
- sn 10y agoFair enough. Your original comment did not make this distinction. I dug and found https://www.fiala.me/pubs/papers/sc12-redmpi.pdf https://www.fiala.me/pubs/papers/sc12-redmpi.pdf the title of which is "Detection and Correction of Silent Data Corruption for Large-Scale High-Performance Computing." They state that for a cray there was a double bit flip about 1x/day for 75k modules. To (probably incorrectly) extrapolate, if you have a server with 16 modules that would be equivalent to a single double failure about once every 13 years.