4 Jun
2021
4 Jun
'21
2:28 p.m.
On Thu, 4 Mar 2021 12:56:59 +1300, I wrote:
Seems like Facebook’s computation facilities have become so massive that it is starting to see intermittent errors in places you wouldn’t expect <https://www.nextplatform.com/2021/03/01/facebook-architects-around-silent-data-corruption/>:
Not just Facebook, but Google too <https://www.theregister.com/2021/06/04/google_chip_flaws/>. In their paper, they suggest that these intermittent errors are becoming more frequent as ever-more-complex chip designs push closer to the limits of what is physically possible. It seems like some particular chips are more prone to this than others, but it needs more extensive testing to identify them than is normally done by the manufacturers.