🚀 go-pugleaf

In article <12142@watnot.UUCP> ccplumb@watnot.UUCP (Colin Plumb) writes:
>In article <1903@mmintl.UUCP> franka@mmintl.UUCP (Frank Adams) writes:
>>>The metrics I'm interested in measure speed.  (Basically, I'm hooked
>>>on fast machines.)  Other constraints are less interesting because:
>>
>>I must disagree.  Reliability is at least as important as speed.
>
>I must disagree.  The the idea is to get as much effective speed out of the
>machine as possible.  A machine that is down 50% of the time delivers 1/2
>of its operational speed to the user as throughput. Turnaround time (which is
>what most people are interested in) will suffer more, under most circumstances.
..whatever that paragraph means.

There are a number of things wrong with this attitude. It is completely bogus 
to assume that "50% downtime" means "up from noon to midnite only". 

A lack of reliability implies that you cannot predict your downtime. 50% 
downtime could easily mean that your "big VAX", which takes 20 minutes
to boot, is ready to accept logins for exactly 1 minute before crashing 
and remaining down for 21 minutes. Machine crashes are a stochastic process, 
and this is no good at all if your probability of failure is high.

Then again, this whole argument is silly. Reliability is in some sense 
orthogonal to the other performance statistics. If your machine crashes
all the time, speed simply doesn't enter into the calculation because
the break it puts in the users' work habits is unacceptable. The probability 
of losing valuable work, computed results, etc. is also too high.
I suppose one could trade off reliability for speed, but most manufacturers
realize that unreliable machines are extremely costly in service and annoyance 
time, and therefore manufacturers try to maximize the reliability.
Unreliable machines are hard to sell.

>           -Colin Plumb (ccplumb@watnot.UUCP)


Chris Shaw    cdshaw@alberta
University of Alberta
CatchPhrase: Bogus as HELL !