Built burn because every K8s cost tool I tried showed estimates but none checked if those numbers matched the real bill. Pulls actual AWS CUR or Azure Cost Management data exact dollar diff per node, per disk, per load balancer. SP, RI, Spot detected from billing. Orphaned volumes and LBs flagged. Works on cloud and on-prem. Reconciliation production-tested on AWS and Azure. Featured on WebProNews. Open source. Single binary. No agent.
No reviews yetBe the first to leave a review for burn
Framer AI AgentsDesign and publish professional sites with AI
Promoted
Maker
📌
Hey, I'm Ozlem.
I run production Kubernetes clusters on AWS and Azure. A few months ago I pulled up our AWS bill and the numbers didn't match what our cost tool was reporting. Looked into it, found orphaned disks nobody remembered creating, a load balancer charging us with nothing behind it, and nodes on full on-demand pricing even though we were paying for Savings Plans.
The cost tool showed everything was fine. The bill told a different story.
So I built burn. One command , pulls your real AWS CUR or Azure Cost Management data, matches it against what your cluster thinks it's spending, and shows you where every dollar actually goes. Per node, per disk, per load balancer.
Started as a side project to scratch my own itch. Now it runs on our production clusters daily.
Got picked up by WebProNews. Speaking about it at SREday London and Amsterdam this year.
Still just me building this. If you've ever looked at your K8s bill and thought "that doesn't look right", that's exactly why burn exists.
github.com/tanrikuluozlem/burn