System A consists of a single ring with 300 stations, one per repeater.
consists of three 100-station rings linked by a bridge. If the probability
failure is P1, a repeater failure is Pr, and a bridge failure is Pb, derive
for parts (a) through (d):
a. Probability of failure of system A
b. Probability of complete failure of system B
c. Probability that a particular station will find the network unavailable,
A and B
d. Probability that any two stations, selected at random, will be unable
communicate, for systems A and B
e. Compute values for parts (a) through (d) for P1 = Pb = Pr = 10-2
He's actually correct.. the MTTF -- mean time to failure for systems running Microsoft operating systems is substantially worse than for *NIX systems.
Did you know that insurance premiums for shops running MS systems is higher than for places with UNIX-like o/ses ?
It also depends on your definition of failure.. do you base failure on the ability of the system to remain running uninterupted ? In which case there is no competition.. Windows boxen need to be rebooted after you install almost any patch..
This is the reason I always laugh at MS loving admins who brag about their Win2k servers uptime of 3 billion years.. the first question I ask them is when was the last time they applied a patch ? Suddenly they keep quiet and retreat to playing solitaire..
UNIX-like systems were designed from the ground up for stability.. Ever noticed how hard it is to actually take down Linux ? Its so compartmentalized.. so I freeze up one TTY.. I switch to another.. out of my choice of around 10... okay, so X borks on me.. I press CTRL+ALT+BACKSPACE and take back control.... even if I lose all of these, I will still be able to manage the box via SSH or similar...
All processes get nicely isolated, if something misbehaves, it usually does not interfere with anything else, unlike Windows where most of the time when something starts dying on you, everything goes in sympathy.
You get so many lives its like playing Pac-Man with cheatcodes !