We wanted to provide “cpu service-load” (vsmd load) and “memory system-load” (overall system memory usage) on the Director UI monitoring based on feedback from customers globally. Whereas the cpu oid provides “cpu system-load” and memory oid provides “memory service-load” (vsmd memory), hence the discrepancy which you cited – this is our implementation currently (also, there is no oid for cpu service-load or memory overall system-load)
,
More details for clarity
- CPU
Oid – the oid 1.3.6.1.4.1.42359.2.2.1.1.1.1.2 provides “overall” cpu usage of the “system” (or what we call “system load”)
Monitor – shows the cpu usage of vsmd service (or what we call “service load”) which essentially encompasses all the load involved in data-plane processing (worker/poller) and versa-vsmd.
- Memory
Oid – the oid 1.3.6.1.4.1.42359.2.2.1.1.1.1.3 provides the memory usage of VSMD (service-load)
Monitor – shows the memory usage of the system (derived from calculating free-memory/used-memory values present in “show system details” output), as the overall memory of the system is the true measure of the memory consumption
Side note: Is the below a typo from your side? You seem to have inter-changed cpu and memory oids
