RZ20 for CPU_Virtualization_Power
- the CPU_Virtualization_Power subtree is in responsibility of the agent
- all other CPU_Virtualization_* nodes are in responsibility of saposcol
This means:
1) agent creates CPU_Virtualization_Power and is responsible for the content
2) fortunately - the data presented in the subnode are gained from saposcol and are reachable in other CPU_Virtualization_* subtrees
3) agent does some kind of renaming in the nodes, e.g. "Entitlement" could be found as "Guaranteed Capacity" in CPU_Virtualization_Virtual_System subtree
Summary - the values present in CPU_Virtualization_Power are collected by agent from other CPU_Virtualization_* subtrees, which are maintained by saposcol. It's kind of duplicity. The saposcol is always the originator of the data. There are no additional data in the _Power node which are not present in other virtualization subtrees.
This means that you can say that in the old EWA reports the OS CPU operating statistics are based on 4x60 is 240 measurements each hour.
After this SAP decided to create new monitor structures for the OS virtualization but the monitoring data for the CPU_Virtualization_Power
is not stored in the 24 hour SAPOSCOL buffer anymore. Actually the data is still collected by SAPOSCOL but only as a snapshot and we do need
the CCMS monitoring segment to store the data for a longer period. This can be done by using an ABAP report which reads the SAPOSCOL
snapshot data and stores the data in the CCMS monitor segment but this report only can be started each 5 minutes or this can be done by using a
SAPCCM4X agent and the agent can read the SAPOSCOL snapshot data every 60 seconds and sometimes even a little bit faster than this.