[c-nsp] Parity Errors and Cosmic Rays

[Andre Beck]

On Thu, May 05, 2005 at 12:14:21PM -0400, Bill Wichers wrote:
> 
> I remember reading somewhere once that a neutrino hitting a DRAM cell will
> release just enough energy to flip the state of one bit...

A *neutrino*? So if the starter of this thread really managed to get
*multiple* hits of *neutrinos* in his DRAM *chips*, he should probably
try whether his beeing a unique statistical fluctuation gets him
anything in a lottery or Vegas ;)

If at all cosmic rays (which smells mostly like a BOFH excuse if not
founded by some reasoning like high altitude or bad space wheather in
progress) I'd rather expect alpha to do the damage, or the shower of
particles created by a high energy particle like a myon.

In practice, slowly decaying hardware is often due to way more primitive
reasons, mostly bad electrolyte capacitors (which tend to be used in
power supply components and thus cause supply voltage to be unstable or
out of the expected range). They age progressively at increased
temperatures - something very common in insufficiently air conditioned
IT racks. MTBF reduces drastically with increasing temperature. But
even under ideal conditions, with the practice of selling incorrectly
declared capacitors that came about in the last years, almost anyone
can have this problem.

-- 
                  The _S_anta _C_laus _O_peration
  or "how to turn a complete illusion into a neverending money source"

-> Andre Beck    +++ ABP-RIPE +++    IBH Prof. Dr. Horn GmbH, Dresden <-