Date:         Mon, 10 Apr 2006 14:45:54 -0400
Reply-To:     "Fehd, Ronald J" <rjf2@CDC.GOV>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "Fehd, Ronald J" <rjf2@CDC.GOV>
Subject:      Re: Efficiency of Proc sort with out =
Content-Type: text/plain; charset="us-ascii"

> From: Sarah Gollin
> Just trying to settle an argument.  My colleague believes
> that it is more

> efficient

that depends on your definition of 'efficient'

> to use an out = statement on a proc sort as opposed to
> overwriting the original dataset.  I have googled this but
> have been unable
> to come up with a definite answer.  Can you help?

generally it is Highly Recommended
to -always- use an out= for proc Sort,
most especially when you are reading
from a two-level data set name

for a one-level name: i.e. work.DataX
then it is not definite.

Using out= leaves your data= data set
in its original order
and, of course, then you have a/your copy
in the order of the by variables.

you might want to subset as you sort:
PROC Sort data = Lib.X(keep  =  ...
                       where = (...))
          out  =     Y;
          by     ...;

which would be another reason
that this topic would discussed under efficiency.

Ron Fehd  the optimization
           or macro maven  CDC Atlanta GA USA RJF2 at cdc dot gov