Peak utilisation is one of those numbers that looks useful until you try to make a decision from it.
A circuit can hit 96% once and look terrible in a report. Another circuit can sit above 80% repeatedly during the same business window and create a much stronger capacity signal.
If both are only shown as peak values, they can look far too similar.
That was the gap I wanted to close with this dashboard.
The aim was not to replace the normal interface charts or traffic reports. Those are still useful.
The aim was to add the missing context around the peak:
- when did it happen?
- how long did it stay above threshold?
- did it repeat at the same clock hour?
- was there enough evidence to justify a review?
- what action does the evidence actually support?
For this version, I kept the scope deliberately simple: monitored Internet circuits, reviewed over the previous full calendar month.
That makes the dashboard more useful for service review conversations, because it is not trying to behave like a live alert page. It is answering a different question:
What did last month’s utilisation evidence tell us?
The main view shows the usual capacity information, but adds a few extra pieces of context:
- monitored interfaces
- circuits needing review
- circuits to watch
- data gaps
- peak utilisation
- threshold exposure
- recurring clock-hour patterns
- action basis
The “Action Basis” part became the most important section for me.
I did not want the dashboard to jump straight from:
to:
That is too blunt.
A better capacity view should show why a review is being suggested.
For example:
Peak utilisation: 96%
Hours above 80%: 18
Hours above 90%: 5
Recurring window: 09:00
Action basis: same clock-hour exceeded 80% on multiple observed days
That gives an engineer, service owner, or customer a much better starting point.
It also helps avoid overreacting to short spikes. Some interfaces look bad in a top utilisation report, but once you add duration and recurrence, the story can change.
The useful bit here is not the visual design on its own. It is the shift in how the data is shaped.
Peak utilisation still matters.
But for capacity conversations, peak should not be the whole story.
The more useful question is:
Was the utilisation sustained, repeated, and strong enough to justify action?
That is what this dashboard is trying to answer.
✌️
PS. I’ll keep sharing these SolarWinds dashboard and observability solution patterns under #Observability Solutions.
