StormForge analyzes your existing Kubernetes observability metrics using machine learning to generate continuous, actionable rightsizing recommendations for CPU and memory resource settings. It integrates with your existing Prometheus-compatible metrics stack — no new observability tooling required.
Who it’s for
Platform engineering and DevOps teams managing Kubernetes workloads at scale. If your organization is overpaying for cloud resources, dealing with OOMKilled pods, or spending engineering time manually tuning resource requests and limits, StormForge automates that work.
How it works
- Install the Agent — a lightweight component that collects metrics from your clusters and reports workloads. Installation takes under two minutes.
- Get recommendations — StormForge’s ML engine analyzes historical usage and generates rightsizing recommendations per container, continuously.
- Apply on your terms — review and apply recommendations manually from the UI or CLI, or install the optional Applier to deploy them automatically on a schedule.
You can tune the optimization goal (cost savings vs. reliability), set per-workload schedules, and configure thresholds to control how aggressively recommendations are applied.
Try the sandbox
Not ready to install? The StormForge sandbox lets you explore StormForge with view-only access — no account or installation required. It runs a simulated e-commerce application across two clusters: one with recommendations applied automatically, one without. A good way to see what rightsizing looks like in practice.
For a guided walkthrough, see the sandbox guide in the documentation.
Start your free trial
No credit card required.