comp.lang.ada
 help / color / mirror / Atom feed
From: rentmei@asterix.uni-muenster.de (Jahn Rentmeister)
Subject: Re: WE NEED A GOOD Ada SORT PACKAGE!
Date: 3 Nov 1994 17:42:13 GMT
Date: 1994-11-03T17:42:13+00:00	[thread overview]
Message-ID: <39b7dl$1hit@obelix.uni-muenster.de> (raw)
In-Reply-To: 2EB8042E@SMTPGATE2.STRATCOM.AF.MIL

Bennett, Chip (KTR) ~U (BennettC@J64.STRATCOM.AF.MIL) wrote:

: I am looking at a copy of "Software Components with Ada", the book that
: accompanies the Booch components.  Under the topic of "external sorting", it
: shows the specs for two utilities: natural merge sort and polyphase sort.  I
: believe the book is available separately.  Any help?

Maybe I don't get it right, but aren't these algorithm's designed for tape
drives? On a Disk using Direct_IO, one could for example use QuickSort.
(Because there is direct access.) There should be even faster possibilities
than a direct translation of quicksort, because one could use available RAM
to presort parts of the data. 

At least polyphase sort will work with sequential access and is bound to it. 
When using direct_io, much faster methods should be available. 
(OK, for variable-length data sets it won't work, i.e. text-files etc.)

Though, I don't know of any readily available packages that do it. But it
does not sound like a big thing to me. (I probably missed the point, eh?)

--
This .sig deliberately left empty.



  reply	other threads:[~1994-11-03 17:42 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
1994-11-02 21:19 WE NEED A GOOD Ada SORT PACKAGE! Bennett, Chip (KTR) ~U
1994-11-03 17:42 ` Jahn Rentmeister [this message]
1994-11-05  5:43   ` Robert Dewar
1994-11-06 15:28     ` Larry Kahn
     [not found] ` <39bvj7$543@goanna.cs.rmit.oz.au>
1994-11-09  7:30   ` Richard A. O'Keefe
  -- strict thread matches above, loose matches on Subject: below --
1994-11-01 22:10 ferguson
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox