Process metrics in Infrastructure dashboard

Elastic Observability Metrics
1

I have ingested process metrics from a Windows Server with 50 GB of RAM and 12 CPU cores. However, I notice that the values shown in the ELK Infrastructure Dashboard are different from those shown in Windows Task Manager. How to calculate the correct value

I have attached the Prcess metric screenshot of Chrome.exe in comment as elastic didn't allow me to attach two images in single description

image
image509×203 7.47 KB

2

image (1)2

3

Hello @srini29 ,
Welcome to the community.

As per the screenshot shared from Windows Machine , for chrome.exe is showing 4.8% CPU & 11.7% Memory & you are expecting the same values to be shown on Kibana Dashboard.

On the Kibana Dashboard screenshot shared for chrome.exe it is showing 31.5% CPU & 0.7% Memory.

Your query is why the difference between the two?

  1. Verify the data collection interval and aggregation settings in Kibana.
  2. Check the configuration of the data source to ensure it accurately reflects the metrics you want to monitor.
  3. Compare the data over a longer period to identify any patterns or spikes that might explain the differences.

Thanks!!

4

hi @Tortoise

I have configured it to ship process metrics for every 1s

5

Is this a multi-session computer with more than user at given time?

6

Just as a test set the collection interval back to default 10s ... 1s BTW is usually much to granular than is needed.