Huge uptimes & cosmic rays

Pavel Machek (
Tue, 8 Jul 1997 20:15:10 +0200


I hear here voices about huge uptimes... Well.

There was some research project, which included _long_
computation. SOmething like year or so. Long enough to worry about
machine failures. Two machines were set up, checking each other. It
was predicted that during that year, two failures will occur. They
occured - computers gave two bad results in one year.

Huge amount of memory & huge uptime is calling for trouble: remember
that once or twice a year, random bit in your machine is selected &
toggled. If it hits kernel or libc or long-lived daemon...

I'm really 	   Pavel
Look at ;-).