Messages in this thread | | | Date | Fri, 26 Feb 1999 10:59:17 +0000 | From | Neil Conway <> | Subject | Re: [offtopic] Re: 2.2.2: 2 thumbs up from lm |
| |
Albert D. Cahalan wrote: > > Neil writes: > > > You seem to think "realtime" computing has a wishy-washy definition. > > Not so. > > > > One of the key points of a realtime system is that any latencies be > > bounded. That's something you'd have one hell of job claiming for a lot > > of OS/app combinations. Don't mistake "usually adequately fast" for > > "realtime". > > There is no "bounded". You can make a histogram that shows how often > you will have failure. Maybe "one failure every 6 years" is what you > are willing to pay for. How about "1/1000 chance of failure in my life"? > You can not ensure that failure will never happen. > > On the practical level, normal hardware tends to break. Disasters happen. > On the theoretical level, I believe quantum theory says that any computer > can produce absurd results.
Agreed, but: IMHO there's a real and qualitative difference between "might fail once a year" and "failure probability 10^-6 during the lifetime of our sun".
The bloke in the street sees a real and **qualitative** difference between the risks of not being able to drive to work in his car because (a) it's been stolen and (b) it's exploded due to all of the molecules in the petrol tank ganging up on the tank at the same time. They can both happen though.
Whether or not you agree to recognise this quantitative difference as also being a qualitative difference is the point on which we seem to differ.
The probability of the car spontaneously exploding is non-zero, but it's negligible. That word negligible is the key here. That's why I said bounded latency - I mean that the tail of the distribution is negligible in one's particular set of circumstances. How small negligible actually is will vary of course, but let's not split hairs.
Neil
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |