Open-source Rafay alternatives

Rancher is an open-source container management platform for organizations that run containers in production. It aims to make Kubernetes easy to run in many places, to meet IT requirements and to support DevOps teams. It is written in Go, released under the Apache-2.0 license, and distributed as a container image that you can start with a single docker run command and then open in a browser.

The main repository is a meta-repository used for packaging that contains most of the Rancher codebase, with other projects and modules listed in go.mod. Stable releases are published as versioned images and a stable tag, and release announcements are posted on the SUSE forums. The README points to the installation guides, a support matrix of operating systems, hardware and software requirements, product documentation, and community help through forums and Slack.

KubeVela is a modern application delivery platform that aims to make deploying and operating applications across hybrid and multi-cloud environments easier and more reliable. It follows a render, orchestrate, deploy approach built on the Open Application Model, and is a CNCF project written in Go.

Deployment plans are declared as workflows that can be run by any CI/CD or GitOps system and extended with CUE, so there is no need for ad hoc scripts. It natively supports multi-cluster and hybrid-cloud scenarios, including progressive rollout across test, staging and production, automatic canary and blue-green releases, continuous verification, and placement strategies across clusters and clouds. Observability dashboards, LDAP integrations, multi-tenancy and fine-grained RBAC are built in.

The architecture is lightweight, with a control plane that can run in a single pod, and extensible through reusable modules and community addons. KubeVela is licensed under Apache-2.0. It suits platform engineering teams that want to give developers a simpler way to ship applications onto Kubernetes.

Claudie is an open-source platform for managing multi-cloud, hybrid-cloud and on-premises Kubernetes clusters. It lets nodepools mix several cloud providers and your own bare-metal or co-located servers, all defined in one YAML InputManifest, with the aim of a Kubernetes engine you never need to migrate away from.

Typical reasons to use it, according to the project, include cost savings, data locality and compliance such as GDPR, managed Kubernetes for providers that do not offer it, cloud bursting, service interconnect, integrating existing servers as nodepools and adding GPU nodes for AI workloads. You supply cloud credentials, describe the cluster declaratively, and Claudie provisions the infrastructure. Scaling up or down means editing a few manifest lines, and a managed load-balancing solution is included for ingresses and the Kubernetes API server.

Supported providers in its topics include AWS, Azure, Cloudflare, GCP, Hetzner and OCI. Claudie is written in Go, licensed under Apache-2.0, and documented at docs.claudie.io with an introduction video.

KubeStellar is a Cloud Native Computing Foundation Sandbox project that simplifies deploying and configuring applications across multiple Kubernetes clusters. It aims to give an experience similar to working with a single cluster and integrates with tools developers already use, without requiring existing resources to be modified.

Instead of configuring each cluster individually, developers define a binding policy between clusters and Kubernetes objects, and KubeStellar uses regular single-cluster tooling to deploy and configure each cluster accordingly. The README lists reasons for using multiple clusters, such as separating development, testing and staging, isolating teams, meeting security or data governance rules, improving resilience across clouds, accessing heterogeneous resources and running at the edge, including in disconnected environments.

The topics mention placement, scheduling, synchronization, rollout, status summarization and GitOps. KubeStellar is written in Go, licensed under Apache-2.0 and aimed at teams moving from one cluster to many or seeking a more streamlined multi-cluster developer experience.