Metrics cost
Reduce time series that nobody queries.
Metrics Engineering.
Well-named metrics, with just the right dimensions and cardinality under control.
The problem we solve.
One extra dimension on a metric can multiply its time series and its cost without adding anything. And metrics with different names for the same thing make it impossible to compare services.
What’s included.
- Metric catalogue and naming conventions
- Cardinality analysis
- Custom metrics from applications
- Aggregation and dropping of dimensions
- Business metrics
How we work.
Five phases, always in the same order. Select each one to see what happens in it. In full projects they map onto the stages of our method.
We review what's monitored today, with which tools, which incidents went unnoticed and what it costs.
We design the collection layer with OpenTelemetry: agents, gateways, egress paths, common attributes and sampling.
We apply naming conventions, aggregation rules and custom metrics.
We compare cardinality before and after and confirm that dashboards and detectors still work.
We tune cardinality, sampling and detectors based on real usage to contain cost and noise.
Technical capabilities.
- Metric time series (MTS)
- Dimensions and properties
- Metrics Pipeline Management
- Histograms
- OpenTelemetry Metrics
- Usage analytics
Use cases.
Business metrics
Measure orders, sign-ups or payments alongside the technical ones.
Standardise
Get every team measuring the same thing the same way.
Benefits for your organisation.
- Metrics cost under control
- Reliable comparisons between services
- More precise detectors
- A shared catalogue
Deliverables.
- Metric catalogue
- Cardinality report
- Applied aggregation rules
- Conventions guide
Frequently asked questions.
What is cardinality?
The number of distinct time series a metric generates based on its dimensions. It's one of the main cost drivers.
Will we lose detail by aggregating?
We only drop the detail nobody uses; we check that against real usage first.
Can metrics be created from traces?
Yes, Splunk APM generates metrics from traces that can be used in dashboards and detectors.
Shall we talk about Metrics Engineering?
Tell us about your situation. If this service is not what you need, we will tell you; if it is, we will propose a concrete first step.
Request this service