From liraola@opera.dia.fi.upm.es  Fri Jan  7 10:44:05 2000
Received: from goofy.fi.upm.es (goofy.fi.upm.es [138.100.8.23])
	by swi.psy.uva.nl (8.9.3/8.9.3) with ESMTP id KAA29628
	for <prolog@swi.psy.uva.nl>; Fri, 7 Jan 2000 10:44:05 +0100 (MET)
Received: from conversion.relay.fi.upm.es by relay.fi.upm.es
 (PMDF V5.2-32 #39034) id <01JKF53I0PWG0002KL@relay.fi.upm.es> for
 prolog@swi.psy.uva.nl; Fri, 7 Jan 2000 10:43:08 MET
Received: from opera.dia.fi.upm.es (opera.dia.fi.upm.es [138.100.11.43])
 by relay.fi.upm.es (PMDF V5.2-32 #39034)
 with ESMTP id <01JKF53HPVZK0002LQ@relay.fi.upm.es> for prolog@swi.psy.uva.nl;
 Fri, 07 Jan 2000 10:43:07 +0100 (MET)
Received: from localhost by opera.dia.fi.upm.es (8.8.4/FI-4.1) Fri,
 7 Jan 2000 08:27:33 GMT
Date: Fri, 07 Jan 2000 08:27:32 +0000 (WET DST)
From: Luis Iraola Moreno <liraola@opera.dia.fi.upm.es>
Subject: Performance degradation revisited
In-reply-to: <200001031009.LAA06251@swi.psy.uva.nl>
X-Sender: liraola@opera
To: prolog@swi.psy.uva.nl
Reply-to: Luis Iraola Moreno <liraola@opera.dia.fi.upm.es>
Message-id: <Pine.SOL.3.95.1000107080022.1669A-100000@opera>
MIME-version: 1.0
Content-type: TEXT/PLAIN; charset=US-ASCII


Regarding the performance degradation hinted by Douglas:

> > I have a very large prolog data file that I consult into a program...
> > 
> > mypl.pl compiled, 41.72 sec, 46,816,660 bytes.
> > 
> > however that was on pl-3.2.8
> > 
> > now under pl-3.3.0 beta 6
> > its taking 2 times longer 

> Nothing serious as far as I can tell.  There are some small
> performance degradations due to the implementation of the `logical
> update-view' semantics, but that should not be more than a few percent
> max.  Since that version there are also changes to read/1 to make it
> reentrant for the multi-threaded version.  That might have slightly larger
> consequences, but twice as long!?

I am afraid I have more evidence supporting a serious performance
degradation in 3.3.0 (beta). I am working with the Prolog distribution of
WordNet 1.6 and there are major differences between versions 3.2.7 (or
3.2.8) and 3.3.0 (last beta), all of them running under Windows NT.

Loading the file "wn_s.pl" (about 174.000 facts) with 3.2.7:

19.4 secs. and 19.22 MB.

The same file with 3.3.0:

186 secs and 21.6 MB.

I have repeated the compilation a few times and the results are stable.
They suggest that the current beta version of the 3.3.0 compiler is
approximately nine times slower than the previous 3.2.7.

Best regards,

Luis Iraola.

P.S.
Indexing the facts by extra arguments besides the first one (using
index/1) does not increment the compilation time. On the contrary, it
seems to speed things up a little bit.



