Cloud Metrics: The Cost-Value Equation (All Things Open 2025)

Are you collecting just about every metric under the sun and the kitchen sink too? Understanding the cost of collecting metrics and the usefulness of those metrics is the only way to scale in a cloud native world. You can’t get away with just collecting everything as you grow. Your observability teams need to make decisions about what to collect, what to drop, what to aggregate, and still be able to alert, triage, remediate, and do their root cause analysis on a daily basis. Gain immediate insights into high cost data (DPPS), when to drop time series data, and how to determine when the value of that data is at its lowest. Session includes a recorded demo video of it in action.

1. Cloud Metrics: The Cost-Value Equation Eric D. Schabell Director Community & Developer Relations @ericschabell.org | @ericschabell{@fosstodon.org} Oct 2025 All Things Open 2025

2. How much could one metric cost anyways? $30,000!!!

3. I’m not instrumenting anything ever…. Dev Team

4. Cloud Native O11y Payment Cloud Native O11y Payment Cloud Native O11y Payment Cloud Native O11y Payment Cloud Native O11y Payment Cloud Native O11y Payment Cloud Native O11y Payment Cloud Native O11y Payment

5. Why is the bill so high? What can you do to bring it down to $XX amount?

6. What factors influence the cost of metrics? # of containers and infra components Granularity Retention Window Cardinality

7. # of apps and infra components

8. Granularity - Downsampling CPU sample: 5s CPU sample: 30s

9. Retention Window

10. Cardinality - Math 1 0,000 timeseries

11. Cardinality - Aggregation 100 timeseries 99% reduction

12. Cardinality - Adding single dimension 30,000 timeseries

13. Cardinality - top 10 only 1,100 timeseries 99.96% reduction

14. Are we getting enough value from this observability data to justify the cost ? ?

15. A valuable metric is one that’s used used in queries & alerts for on-call (issues) charted on used and maintained dashboards directly queried by engineer(s) is referenced in a shaping rule

16. Puts explanation of value for every metric in your system; where each metric is used, how much, and by whom

17. Cost defined in terms of persisted writes per second

18. Value defined in terms of persisted writes per second

19. This metric is valuable But not in all dimensions

20. Customer Impact - Control Plane optimization of data volumes optimization of data volumes metrics reduction via aggregation 98 % 80 % 41 % query latency improvement 8 x

21. How much could one metric cost anyways?

22. Questions?