Dynatrace custom license consumption

[y_buccellato]

Hello friends of the community,

I do have a question on Custom License Consumption but first lets get out of the way the official documentation:

https://www.dynatrace.com/support/help/reference/monitoring-consumption-calculation/

I Got 66 Application Server in my Environment.
If I want some precious jmx metric I use the good old JMX/PMI plugin editor.

I activate the plugin to catch the new custom metric "Hogging Thread".

Then I decide that I would also love to know the pending user request count per each application server.

So my guess is that I will consume 66 x 2 (pending user & hogging thread) =132
But I'm now consuming quite more than 198 plus custom metrics.

I'm sure I'm missing something on consumption calculation, can you support me ina having a clear picture on subject?

Thank you

[Radoslaw_Szulgo · Accepted solution]

Custom metrics are consumed per component and per collected dimensional value.

If you have 2 metrics, at 66 hosts with 3 applications (processes) each you will consume:

2 * 66 * 3 = 396 custom metrics

Can you please share more details on how many hosts and processes you turned it on?

[y_buccellato]

Oh Hello, sorry for the delay!

I have 66 Application Server (Processes) and 2 metrics (Hogging thread and Pending user request) so the calculus will be 2*66 and this will be correct.

As I casually discovered there is a way to see the actual consumpion per activated plugin as mentioned by @Julius L. : Settings > Integration > Custom metrics overview:

although I'm a little bit disappointed about jmx being a custom metric since it was a common one in my previous monitoring experience.

Have a great day

[Julius_Loman · Accepted solution]

There is an overview within the tenant available at:

Settings > Integration > Custom metrics overview

This will give you more insight into consumption. Metrics are consumed as @Radoslaw S. mentions. If you have multiple dimensions or more processes, you might easily find yourself consuming much more metrics than initially expected.