Date:         Tue, 10 May 2005 20:02:05 +0000
Reply-To:     toby dunn <tobydunn@HOTMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         toby dunn <tobydunn@HOTMAIL.COM>
Subject:      Re: help on macro
Comments: To: nospam@HOWLES.COM, radevenz@ix.netcom.com
In-Reply-To:  <MC8-F179NA0GufdMmMU00021f86@MC8-F17.hotmail.com>
Content-Type: text/plain; format=flowed

Howard and Richard,


It seems (as Howard mentioned) that at a million records there is a
difference:

where upcase(alpha) contains "%upcase(&MatchString)" ;

Real Time 41.43
CPU Time 41.06

where upcase(alpha) contains upcase("&MatchString") ;

Real Time 55.92
CPU Time 50.95


I don't know about ya'll but ten or so seconds difference whoopieeeee, guess
the SAS optimization is better than I thought.  I am not sure about the
scaleability but if say it is constant then unless I have to process say a
billion records I don't think it really makes a big deal.  Well I learned
something today in any event it is off to do something that i do better than
annoy Ian over macro issues that being cooking gourmet food.

Toby Dunn




From: "Howard Schreier <hs AT dc-sug DOT org>" <nospam@HOWLES.COM>
To: Toby Dunn <tobydunn@HOTMAIL.COM>
Subject: Re: help on macro
Date: Tue, 10 May 2005 15:03:57 -0400
Received: from malibu.cc.uga.edu ([128.192.1.103]) by MC8-F17.hotmail.com
with Microsoft SMTPSVC(6.0.3790.211); Tue, 10 May 2005 12:04:03 -0700
Received: from listserv.cc.uga.edu (128.192.1.75) by malibu.cc.uga.edu
(LSMTP for Windows NT v1.1b) with SMTP id <2.00F5799D@malibu.cc.uga.edu>;
Tue, 10 May 2005 15:03:57 -0400
X-Message-Info: JGTYoYF78jFCgn0FGSUA7e3uGIcHyzXKAHNjC1bsRoE=
Return-Path: nospam@HOWLES.COM
X-OriginalArrivalTime: 10 May 2005 19:04:04.0273 (UTC)
FILETIME=[08991210:01C55593]

You need a bigger test file (like a million, or several).

On Tue, 10 May 2005 18:16:02 +0000, toby dunn <tobydunn@HOTMAIL.COM> wrote:

 >Richard,
 >
 >I ran a test with 6000 obs. here are the numbers:
 >
 >
 >where upcase(alpha) contains "%upcase(&MatchString)" ;
 >
 >took 0.70 real time
 >       0.30 CPU time
 >
 >
 >where upcase(alpha) contains upcase("&MatchString") ;
 >
 >took 0.29 real time
 >       0.29 CPU time
 >
 >Isn't that weird I figured the second way would take slightly more time
than
 >the first. But after triple checking my numbers and making sure noi
cacheing
 >was taking place its what I get on my machine.
 >
 >
 >Toby Dunn
 >
 >
 >
 >
 >From: "Richard A. DeVenezia" <radevenz@IX.NETCOM.COM>
 >Reply-To: "Richard A. DeVenezia" <radevenz@IX.NETCOM.COM>
 >To: SAS-L@LISTSERV.UGA.EDU
 >Subject: Re: help on macro
 >Date: Tue, 10 May 2005 13:00:39 -0400
 >
 >toby dunn wrote:
 > > Hawk_eye,
 > >
 > > %macro TestLike(DsnOut = , DsnIn = , MatchString = ) ;
 > >
 > >   data &DsnOut ;
 > >    set &DsnIn ;
 > >    where upcase(alpha) contains upcase("&MatchString") ;
 > >   run ;
 > >
 > > %mend ;
 >
 >Not knowing the what/when/wherefores of the SAS compiler optimizations, I
 >would not give SAS the chance to huff and puff on the second upcase on
every
 >row.
 >
 >     where upcase(alpha) contains "%upcase(&MatchString)" ;
 >
 >--
 >Richard A. DeVenezia