Overview
A few CPU-related values are reported in a Nexus Repo Manager support's ./info/jmx.json file:
- AvailableProcessors
- SystemLoadAverage
- SystemCpuLoad
This article will explain the meanings of each, and how to use this information to assess the CPU load at the time the support zip was captured,
AvailableProcessors
Before any discussion of load, we'll need to recognize the significance of AvailableProcessors. This value corresponds to the maximum processing resources, and give color and context to the load values. Two identical load values could correspond to drastically different in situ behavior.
SystemLoadAverage
This value represents the 1-min average load on the host (e.g. cat /proc/loadavg | awk '{print $1}'). The value can range from 0 (no load), through the AvailableProcessors value (100% utilization of all processors) and beyond. If the value is greater than AvailableProcessors, then threads are queued and the CPU resources are overwhelmed.
SystemCpuLoad
This value represents the CPU load when the support zip was captured. It should be close to the SystemLoadAverage, but may be lower due to the timing of the load events wrt. the support zip capture.
For more information about load averages, you can check out: https://www.digitalocean.com/community/tutorials/load-average-in-linux