With IBM zOS systems it is valuable to have monitoring active looking for issues that may impact performance of your Sysplex/LPAR/CPCs workloads.  This monitoring can alert you of issues that need to be addressed before workloads are impacted.   

I have an example that happened recently to share.   I noticed first when looking at my important LPARs CPU use over last 7 hours that one LPAR was using more CPU constantly.  You can see RSB1 in the 40%-55% CPU range for last 7 hours.  Also the MVS Overhead is very high for this LPAR. This is not normal for this LPAR.

The details show Preemptable Home SRB time used of 126,236 Seconds by AXR03 - System Rexx address space.

Now I also could of (or even should of) seen this in the Situation Status Tree if I looked there first.  It shows an alert for AXR03 on RSB1 for high CPU usage.

Drill down there takes me to the same AXR03 SRB CPU use information:

So next we figured out two different AXR03 starts happened that ended up driving this improper CPU usage.   Getting the LPAR setup to only one proper AXR03 address space took the CPU back to normal.

So pro-active monitoring of my LPARs showed an exceptional CPU use issue.  Diving down we found the root cause and corrected it to save our shop unneeded CPU use.