Recommendations for Kubernetes

CloudZero analyzes your Kubernetes environment and generates recommendations that identify workloads where you can reduce costs by rightsizing resource requests. Each recommendation includes the affected workloads, the estimated savings, and guidance on how to address it.

For details on how to work with recommendations in the CloudZero UI (search, filter, group, take action), see Recommendations.

What you need

All Kubernetes recommendations require version 1.2 or later of the CloudZero Agent for Kubernetes.

Overview of Kubernetes Recommendations

Recommendations

Rightsize Kubernetes CPU Over-Provisioned Workload

CloudZero has identified Kubernetes workloads where CPU resources are over-provisioned by 30% or more based on actual utilization patterns over the last 30 days. These workloads are requesting significantly more CPU than they actually use, resulting in wasted cluster capacity and higher infrastructure costs.

Over-provisioned workloads tie up cluster resources that could be used by other applications or allow you to reduce your overall cluster size. By rightsizing CPU requests to match actual usage patterns, you can reduce costs while maintaining application performance.

This recommendation requires at least 50 data points over 30 days to ensure reliable analysis.

How to address this

• Review CPU requests: For the identified workloads, compare actual utilization patterns to current requests
• Reduce CPU requests: Better match actual usage, leaving an appropriate safety margin for traffic spikes
• Test application performance: After rightsizing, ensure requirements are still met
• Monitor workload performance: After changes, verify CPU resources remain adequate
• Consider implementing Vertical Pod Autoscaler (VPA): For automatic rightsizing recommendations
• Update deployment manifests: Update Helm charts with optimized CPU request values

---

Rightsize Kubernetes Memory Over-Provisioned Workload

This recommendation identifies Kubernetes workloads where memory is over-provisioned by 30% or more based on actual memory utilization patterns over the last 30 days.

What it identifies

• Analyzes memory utilization vs. memory resource requests for Kubernetes pods
• Identifies workloads using less than 70% of their requested memory (30%+ over-provisioned)
• Calculates potential cost savings from rightsizing memory requests
• Uses 30-day historical utilization data with data quality filters to ensure reliable recommendations

How savings are calculated

Savings formula: workload_cost × (% memory over-provisioned - 30%)

How to address this

• Review memory resource requests: For identified workloads, compare to actual utilization patterns
• Reduce memory requests: Match actual utilization patterns with an appropriate safety margin
• Test application performance: After rightsizing, ensure requirements are met
• Monitor workload performance: Track memory utilization after changes
• Consider implementing Vertical Pod Autoscaler (VPA): For automatic rightsizing, or Horizontal Pod Autoscaling (HPA) to reduce replica count
• Update deployment manifests: Update Helm charts with optimized memory requests