LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (June 1997, week 4)Back to main SAS-L pageJoin or leave SAS-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Wed, 25 Jun 1997 21:07:19 -0700
Reply-To:     "Karsten M. Self" <kmself@IX.NETCOM.COM>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         "Karsten M. Self" <kmself@IX.NETCOM.COM>
Subject:      Re: Max number of numeric elements in an array
Comments: To: Jan Winchell <WINCHELL@KENTVM.KENT.EDU>
Content-Type: text/plain; charset="us-ascii"

You are typically limited by memory and/or a majik number depending on the type of array you are defining. I've generated test datasets with temporary arrays consisting of over 50,000 elements on our HPUX 9000/K200, running 128 MB RAM at the time. This ate into swap somewhat considerably, but SAS handled it. When defining *permanent* arrays, SAS starts getting a bit unhappy at 2^15 elements (32,768) I get 13,456 elements in your 116 x 116 matrix, so this probably isn't the problem.

Try looking at your MEMSIZE option, which specifies the maximum amount of memory available to your process. Big Iron has other aspects I'm not so familiar with, look to a Tim Berryhill response for specifics on that.

Karsten M. Self (kmself@ix.netcom.com)

What part of "gestalt" don't you understand?

-----Original Message----- From: Jan Winchell [SMTP:WINCHELL@KENTVM.KENT.EDU] Sent: Wednesday, June 25, 1997 11:09 AM To: Multiple recipients of list SAS-L Subject: Max number of numeric elements in an array

I am having memory problems with a SAS program in which I attempt to build a large array. The preliminary version 50 x 50 works in 16M. However, when I attempt to run the entire set of 116 cases which should generate an 116 x 116 matrix, it aborts.

We are using VM/CMS and Rel 6.08 of SAS.

Any help or information about limits on arrays would be appreciated.

Jan Winchell WINCHELL@KENT.EDU Systems Analyst WINCHELL@KENTVM.KENT.EDU Academic Computing and Technology Kent State University


Back to: Top of message | Previous page | Main SAS-L page