I am using SAS Release 9.4 TS1M3 on Red Hat EL 7.2. I am new to SAS; in fact, I help manage the applications on our research clusters, and am not a SAS user. I've a researcher who runs SAS on our clusters, and she is receiving the "ERROR: The SAS System stopped processing this step because of insufficient memory" message in one of her compute jobs. We're trying to figure out if it's an issue in her code, or an issue in our SAS configuration. This is one of the procedures that causes the error: Proc mixed data=temp1 noclprint covtest plots(maxpoints=none);
title "Model 7";
class year families;
model income_constant =families year2 year2*year2 families*year2 families*year2*year2/s ;
random intercept families / subject=year ;
repeated /type=sp(pow)(year3) sub=intercept;
weight weightvar;
format families families. ;
run; The output in the SAS log is: WARNING: The R matrix depends on observation order within subjects. Omitting observations from the analysis because of missing
values can affect this matrix. Consider using a classification effect in the REPEATED statement to determine ordering in
the R matrix.
NOTE: 440948 observations are not included because of missing values.
NOTE: 38 observations are not included because of negative or zero weights.
ERROR: The SAS System stopped processing this step because of insufficient memory.
NOTE: The PROCEDURE MIXED printed page 2.
NOTE: PROCEDURE MIXED used (Total process time):
real time 44.52 seconds
user cpu time 17.14 seconds
system cpu time 20.56 seconds
memory 16456514.29k
OS Memory 16467496.00k
Timestamp 10/16/2018 09:42:32 AM
Step Count 8 Switch Count 0
Page Faults 2
Page Reclaims 10509
Page Swaps 0
Voluntary Context Switches 209
Involuntary Context Switches 41
Block Input Operations 1816
Block Output Operations 2650776 I've tried setting the various memory settings to either max or 0; I've run this on compute nodes with various RAM sizes: 128GB; 384GB; I even tried it on a node with 4TB of RAM. In any case SAS always claims that the "OS Memory" is 16.4GB. Any help would be most appreciated.
... View more