real-time II

[Geoff Kuenning]

In article <5876 at brl-tgr.ARPA> Barry Shein writes:

>	...and besides, if you don't consider handling disks
>	and high performance network devices a real-time
>	problem then you are kidding yourself. Its really
>	all quite similar.

My definition of a real time constraint is one that will have catastrophic
consequences for the system if it is violated.  What are the catastrophic
consequences of delaying a disk interrupt, or completely missing a newtork
interrupt?  Compare this to the consequences of a missed cutoff of the
space shuttle main engine.

Many people seem to think that 'real-time' is the same as 'high-performance'.
Although the two often go hand in hand, they are really quite different.
-- 

	Geoff Kuenning
	First Systems Corporation
	...!ihnp4!trwrb!desint!geoff