Skip to content

Right-sizing

Sizing a micro-service requires running performance testing and constant optimization. Komodor analyzes the actual consumption against the allocated resources and provides an extensive overview of suggested modifications and savings.

Our right-sizing dashboard includes: - High-level picture of your current cost recommended cost and the potential saving
- Visualization of our recommendations against your requests during the last 30 days, for both Memory & CPU
- Detailed right sizing recommendations per service
- The table will only include services in which there is a saving potential

When to use Right Sizing

If Then
You’re looking to cut down significantly increased Kubernetes cloud costs without compromising performance Choose the perfect optimization strategy for your needs to confidently right size
You (or your teams) are unsure how to configure CPU & memory requests for your containers, or duplicate requests from existing containers Komodor cannot recommend initial requests but can offer adjusted right-sizing recommendations based on actual usage within a week

Understanding the right-sizing page

  • Our right sizing recommendations are based on the last 30 days
    • Therefore, expect to see different results vs. your cost allocation page, in which you can set different timelines
  • Your overview data can be scoped to one or more clusters (default: all)
    • However, you can apply additional filters for the detailed report at the bottom of the page by clicking on ‘Filters’ right above it
  • The right sizing report looks quite similar to the cost allocation report, but there are a few key differences:
    • This report offers a more granular default view, showing recommendations per service
    • The CPU (core) and memory (MiB) columns will include both your current status (similar to the Cost Allocation report) and the recommended new requests

Komodor’s Optimization Strategies

When making right-sizing recommendations, it’s important to understand your preferences and needs. Choose a specific optimization strategy to ensure our recommendations are based on the right balance between your costs and your resource availability.

Strategy Name Description
Conservative Prirotizies availability over cost optimization
Moderate - Recommended, default strategy A mixed, balanced strategy
Aggressive Prioritizes cost optimization over availability