Re: data queues

Simon

At 08:12 PM 5/25/1998 +1000, you wrote:
>Hello Vernon,
>
>Indexed data queues are simply sorted on the key value.  On CISC systems
the queue is sorted when entries are 
>added; RISC system do the sort when entries are retrieved.  The sort is a
not at all efficient On2 (how do I do 
>a superscript in this mail program???? -- that is meant to mean
Order-n-squared).  That is why IBM recommend 
>keeping the number of entries in a data queue to a minimum.  The more
there are the longer the sort will take 
>-- can be hours for hundreds of entries (guess how I found out!!).
>
>User queues are more efficient but I'm not sure of the key implementation.
 User indices expose the system 
>independent index support and do use the internal binary radix tree

O-n**2? Ouch! Does that suggest they're using a bubble sort!  ;-[

Jon Paris presented a "free stuff" at the Connections Conference, where he
mentioned using 'qsort' from RPG IV. It would appear that, with the pointer
to the start of the queue, the number of entries, the length, the overall
size, we might be able to use qsort over it and get much better
performance. But does it need to be fixed-length?

Just thinking—

Vernon Hamberg
Systems Software Programmer
Old Republic National Title Insurance Company
400 Second Avenue South
Minneapolis, MN  55401-2499
(612) 371-1111 x480


+---
| 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
+---