Differences

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

--- cluster:162 [2017/09/05 15:01] – created hmeij07
+++ cluster:162 [2017/09/06 18:36] (current) – [Test Queue] hmeij07
@@ Line 1: / Line 1: @@
 \\
 **[[cluster:0|Back]]**
+  * Jobs can be submitted from any node.
+  * ''cottontail'' is primary scheduler login node.
+  * You can login in to any **tail** node directly (via ssh).
+  * All nodes are CentOS 6.x with some exceptions noted.
 ==== Test Queue ====
+  * Wall time (CPULIMIT) has been removed (was 8 hrs/job)
+  * Job slots raised to 48, is comprised of
+    * ''greentail'' our /sancratch provider and redhat 5.5
+    * ''cottontail2'' physically identical to ''greentail'' for fail over
+      * standby scheduler master for ''cottontail''
+    * ''swallowtail'' secondary login node
+    * ''petaltail'', sandbox and Warewolf provisioning node
+    * ''n29'' and ''n31'', our OpenHPC test cluster compute nodes, CentOS 7.3
+      * joined to our Openlava environment
+You can target individual nodes with '#BSUB -m node-name' in your submit script.
 ==== DR ====
-  * ''sharptail2dr'' is our disaster recovery node (to back up ''sharptail'')
+  * ''sharptail2dr'' is our disaster recovery node (to back up ''sharptail'''s /home)
-  * Every couple of days users' /home will refreshed on this DR node
+  * Every couple of days users' /home will be refreshed on this DR node
   * 17TB of usable storage
-  * Users selected for first or refresh backup are obtained from job submission lists last 30 days
+  * Users selected for first or refresh backup are obtained from job submission logs last 30 days
   * No logins are permitted
   * Node will eventually be moved to Freeman DR location
-  * If users become inactive backup home directory will not be removed (while space is available)
+  * If users become inactive, backup home directory will not be removed (while space is available)
+  * vlan52 bonded interfaces (2G connection)