Overview

 

Komodor’s advanced cost capabilities are offered as part of Komodor’s paid plans, including 14-day trials.
Learn more about our pricing plans, or contact us to request trial access

Komodor analyzes your cluster resources and consumption and provides 2 main capabilities:
1. Complete visibility into your Kubernetes cost allocation
2. Actionable optimization via right-sizing recommendations

When clicking on ‘Cost Optimization’, you’ll notice tabs for two different pages: 1. Cost Allocation - full visibility into your costs structure from your overall environment to specific clusters, namespaces, or services
2. Right Sizing - Detailed recommendations based on actual consumption vs. configured requests

Prerequisites

  • An agent update is required per cluster
    • Agent version 0.2.42 and above
    • Metrics enabled
    • Update command:
      helm repo update; helm upgrade --install k8s-watcher komodorio/k8s-watcher --set metrics.enabled=true --reuse-values
    • A relevant notification will appear if the agent isn’t up-to-date
  • To generate valuable, significant analysis & recommendations, we recommend waiting at least 7 days after the agent update to review your data.

Please Note: Our calculations are based on default values for CPU & memory monthly prices. If needed, you can modify these values to fit your exact prices.

Komodor Attributes

  • Optimization Score: How optimized your environment is, using a 0%-100% range. Calculated based on current costs, the savings recommended by Komodor, and the savings already applied.
    • For example, if your current costs are $10k, and Komodor identifies a saving opportunity of 30% ($3k), your optimization score will be 70%.
  • Share of Allocated Cost: The % out of the total allocated (requested) costs associated with each row, in the selected scope and timeframe.

Cost Optimization Score Overview

For both right-sizing and allocation, the cost optimization score is classified as follows:

  • <80: Could be improved.
  • >80: Good.
  • >90: Great.

Was this article helpful?

0 out of 0 found this helpful

Comments

0 comments

Article is closed for comments.