Differences

This shows you the differences between two versions of the page.

--- cluster:sas [2018/10/01 17:57] – created mcloughlin
+++ cluster:sas [2024/11/11 21:03] (current) – removed mcloughlin
@@ Line 1: / Line 1: @@
-====== SAS ======
-The cluster has SAS v. 9.2.  To run SAS, log in to clustersas.econ.umd.edu.  Your same password applies.  You can run SAS with the command:
-<code>
-sas myfile.sas &
-</code>
-Always include the ampersand (&) character or you will have to wait util your job finishes before issuing any other commands.
-To run SAS interactively, you have to have X Forwarding enabled.  If you are running Linux on your personal computer, you probably already have the X windowing environment installed.  From the terminal of your Linux machine SSH to the cluster with the following command:
-<code>
-ssh -X username@clustersas.econ.umd.edu
-</code>
-The default allocation of memory for a SAS job is fairly low.  If you are estimating a model with many fixed effects or many observations, it is likely that SAS will run out of memory when it builds the design matrix.  You can adjust the amount of memory allocated to your process at the command as:
-<code>
-sas -memsize 500M myfile.sas &
-</code>
-where "500M" allocates 500 megabytes to your process.  If you are just doing basic data processing and cleaning with SAS, this step should not be necessary.