How Does CostGraph Work? - CostGraph Documentation

This guide explains, at a high level, how CostGraph arrives at its Kubernetes recommendations.

How recommendations are made

CostGraph recommends requests and limits for the containers in your Kubernetes cluster. Recommendations are based on how those containers actually used CPU and memory over a recent period.

Recommendations
Requests and limits
Resource quotas

CostGraph works by:

Collecting CPU and memory usage for every container over a recent period.
Grouping that usage by container, node, and namespace.
Filtering out short-lived outliers so recommendations follow the workload’s real, stable behaviour.
Recommending requests and limits that fit that behaviour, with headroom for normal peaks.
Serving the recommendations through the CostGraph dashboard and API.

Cost Usage Reporting (CUR) Pricing

All pricing information on CostGraph is provided through our pricing API. CostGraph currently all major cloud providers, with our list expanding fast. Once the CPU and Memory values are computed, the values are multiplied by the per-hour CPU and Memory pricing to obtain the pricing results.