Data collection mechanism of High CPU Utilization for Linux server

[Vinay kumar]

I want to know the mechanism of data collection of High CPU utilization.
How does the item key "system.cpu.util" is collecting the data in Linux ?
What is the source of the data collection of High CPU utilization in Linux ?
Based on what metrics the CPU calculation is done ?

[Atsushi]

The Linux kernel constantly manages the state of the OS, and that information can be referenced by accessing virtual files. For example, information about system.cpu.util would have been obtained by referencing the file /proc/stat. In this file, how much time was spent on what kind of processing is recorded. Based on that information, it is possible to periodically obtain what percentage of the total was used for each process and obtain it as the value of the item.

[Vinay kumar]

Hi Atsushi,
Thank you for the response.

May I know, If there is any data differences of monitoring from AWS CloudWatch & Zabbix of and AWS instance. if yes then what is the reason ?

[Atsushi]

Please refer to the information described in the URL below for the metrics that can be obtained as EC2 instance metrics using CloudWatch.

https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/viewing_metrics_with_cloudwatch.html

I don't think it was obtained directly from the OS on the instance, so the same value cannot be obtained, and the granularity that can be obtained is also different.