marc@oahu.cs.ucla.edu (Marc Tremblay) (10/17/89)
We often hear that most of the errors caused by hardware faults in computer systems are the result of transients faults in memories. Does anybody actually have a reference on this "fact"? Thanks, Marc Tremblay marc@CS.UCLA.EDU
jewett@hpl-opus.HP.COM (Bob Jewett) (10/20/89)
> Does anybody actually have a reference on this "fact"?
None at hand, but there are lots of references. Try a search on "soft
error rate" and "alpha particle induced errors" in the various IEEE
circuit-related journals (ED, EDL, JSSC, ...).
The actual experience in this department, is that there is about one "soft
error" in each 400 megabyte-months of DRAM use. Since we have about 400
megabytes of DRAM installed in various workstations, there is roughly one
parity error per month. Some of those are in ECC RAM, so it doesn't result
in a system crash.
Bob Jewetttoms@omews44.intel.com (Tom Shott) (10/24/89)
Check some of the work published by R.K.Iyer of U.Illinois Urbana Champaign. At one point he was doing system level work on faults. -- ----------------------------------------------------------------------------- Tom Shott INTeL, 2111 NE 25th Ave., Hillsboro, OR 97123, (503) 696-4520 toms@omews44.intel.com OR toms%omews44.intel.com@csnet.relay.com INTeL.. Designers of the 960 Superscalar uP and other uP's