Velero donated to CNCF
Broadcom donated Velero—an open‑source Kubernetes backup and recovery project—to the Cloud Native Computing Foundation, positioning it as a community standard for K8s data protection. (x.com) The move could reshape enterprise thinking about Kubernetes backup/recovery options within cloud‑native stacks. (x.com)
Most Kubernetes outages do not start with code. They start when a cluster disappears, a storage volume corrupts, or a team realizes the app came back but the data did not. Velero exists for that exact moment: it backs up Kubernetes resources and persistent volumes so they can be restored after loss or moved to another cluster. (velero.io, velero.io) Velero does not just copy files. Its documentation says each backup and restore is stored as a Kubernetes custom resource, which means the backup job is described inside the cluster in the same way Kubernetes describes apps, policies, and storage. (velero.io) For the actual data, Velero uses two layers. It saves cluster metadata to object storage, and it can also protect attached volumes through snapshots or file-system backup using tools such as Kopia and Restic. (velero.io, velero.io) That makes Velero one of the few projects aimed at the least glamorous part of cloud-native computing: getting stateful applications back after something breaks. In Kubernetes, stateless services can be recreated quickly, but databases and attached storage are the part that turn a restart into a recovery project. (velero.io, cncf.io) Broadcom has now handed Velero to the Cloud Native Computing Foundation as a Sandbox project, and both Broadcom and the foundation describe that move as a shift to vendor-neutral, community-driven governance. The formal announcement was published during KubeCon + CloudNativeCon Europe in late March 2026. (broadcom.com, blogs.vmware.com) The Sandbox label matters because it is the Cloud Native Computing Foundation’s entry point for early-stage community projects. It does not mean “finished”; it means the project now lives inside the foundation’s public pipeline instead of sitting under one company’s umbrella. (cncf.io, broadcom.com) That changes the sales conversation around backup software. A bank or retailer deciding how to protect Kubernetes data is usually wary of betting on a tool controlled by a single vendor, and a foundation home makes Velero look more like shared infrastructure than a product feature. (cncf.io, blogs.vmware.com) It also fits a broader pattern inside cloud-native software. The most widely adopted layers of the stack, including Kubernetes, Prometheus, and Envoy, became defaults partly because they were governed in public and could be extended by many vendors at once. (cncf.io, cncf.io) Broadcom is not walking away from the code. Its March 2026 statement paired the donation with continued enterprise Kubernetes support, which suggests the company wants the core project to become common infrastructure while it sells support and integrated platforms around it. (broadcom.com, blogs.vmware.com) If that works, Velero stops being known mainly as a tool that came out of the VMware orbit and starts being judged the way foundation projects usually are: by contributor depth, plugin ecosystem, and how often operators trust it during the worst day in production. (velero.io, cncf.io)
