Re: Record Length - Trivia

[email@xxxxxxxxxxxxxxxxxxx (James W Kilgore)]

Sector size, multiple of, or evenly divisible into was the rule I
learned when the Earth's crust was still cooling.

Since then, it's been the amount of data a disk arm can read without
introducing a rotational delay or evenly divisible into.  The "evenly
divisible into" is because any subsequent reads don't actually perform
any disk IO.

Like you, I stopped worrying about it long ago.  AFAIK, different
drives/controllers are optimized for either a high number of short
bursts or a low number of large bursts.

Oh, oh, I've got the answer: It depends! <VBG>

Pat Barber wrote:
> 
> John P Carr wrote:
> >
> > >>John P Carr wrote:
> > >> TRIVIA QUESTION;
> > >> If there were such a thing as an optimal record length what would it be ?
> 
> >
> > Something like that.  On the CISC machines,  John Sears once said it would
> > be 31 bytes because of something like you said.
> 
> John:
> 
> I can't remember that "cute" number(nobody cares anymore) but I "think"
> it was 256 for the CISC and it jumped to 512 for the risc... but I been
> wrong before. Back before we had machines with a gazillion disk bytes,
> I paid attention to that sort of thing..
+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---