1 Nov
2001
1 Nov
'01
8:14 a.m.
Where I used
to work, there was an IBM PS/2 model 80 that was installed
in a closet to do coordination of manufacturing equipment status and
utilization reports when it was new. It ran DOS. It was deinstalled in
May 2001. It had been running continuously without a crash, except for
losses of power, and that only happened twice in the same year, in 1992.
i wouldn't doubt it, i used to run DOS on my 386 many moons ago, and it gave me any trouble. it's great. it's almost
entirely useless for doing
more
than one thing at a time. when you get to be that
simple, stability is just
about given to you. but once you step up to large multiuser systems you
have
all sorts of contention for the same resources that
you don't have in a
single
tasking environment. so yeah, i believe you are
right. i'm not, however
very
impressed. :)
why don't you start throwing mainframe data at us? mainframes run a lot
like
the old vaxen, uptimes in the double digit years
range.
Heh... depends on the mainframe.
The CDC-6600, the world's first supercomputer, was for many years
rated as having a mine-hour MTBF... turned out that was because a
counter was oveflowing after 9 hours of ticking away, and only
under one particular operating system (SCOPE). But even under the
more-stable KRONOS operating system, the field engineers typically
took it down for an hour each morning. Periodically, they would
polish the platters on one of the disk drive units (in the early
90s, we used to kid about using Lemon Pledge to cure stiction, but
they drives *really* did get polished), while smoking a cigarette.
Ok, I'm drifting away from reliability, so I'd better cut & run.
-dq