[Cryptography] Vulnerability of RSA vs. DLP to single-bit faults
leichter at lrw.com
Sat Nov 1 22:21:45 EDT 2014
On Nov 1, 2014, at 9:08 PM, Dave Horsfall <dave at horsfall.org> wrote:
>> FWIW: The alpha particle cause mostly does not apply. There was a
>> case at IBM where contamination of packages did cause Alpha particle bit
>> flips [...]
> I'm surprised. Everything I was taught about alpha particles (i.e. helium
> nuclei) was that just a sheet of paper will stop them. Or is this alpha
> decay from within, as you seem to imply?
The chips used ceramic packages - made ultimately from clay. The clay used by IBM and other American makers had a much higher concentration of alpha emitters than that used by the Japanese chip makers. ("Much higher" is relative - the clay was far from radioactive in any normal sense and no one had previously considered the significance of naturally occurring radioactive materials in clay. They are actually used now to date pottery: The emitted particles cause detectable damage to fired clay which builds up at a known rate.)
The net effect was that Japanese memory chips were much more reliable than American ones. The effect depends on memory cell size, hence capacity. At some point (I no longer remember where), chips became dense enough for this to be a big issue. There was all kinds of teeth-gnashing and complaints about American workers and interest in Japanese approaches to reliability before the real cause was worked out. But the effect on the industry was dramatic: American makers, who had invented memory chips and completely dominated the market for chips at size S, pretty much completely lost it at size 4S (and, of course, never got it back).
More information about the cryptography