Learn Cloud Native & CNCF - The Graduated Projects Catalog

#Introduction

After understanding the project lifecycle in episode 4, it's time to see who has reached the top. Graduated projects are the most mature technologies in the CNCF ecosystem — those that have passed third-party security audits, have documented governance, and are widely used in production.

The number of graduated projects keeps growing, reaching around 35 projects as of this episode's writing. There's no need to memorize them all; far more useful is understanding the role of each group and its position in the cloud native landscape you studied in episode 2.

By the end of this episode you will know the most influential graduated projects, understand their role groups, and be able to read the official catalog at cncf.io/projects.

#What a Graduated Project Is

#The Highest Maturity Standard

A graduated project has proven itself across every dimension discussed in episode 4: audited security, mature governance, sustainable community, and broad production adoption. This status isn't granted once and forever — projects remain under oversight and are expected to maintain their quality.

Consequently, graduated projects can be considered the ecosystem's "first class." Organizations choosing technologies usually start with this group, and vendors build products on top of them because the demand is already certain.

Note that graduated status doesn't protect a project from changing times. A project relevant today can be displaced tomorrow by a better technology — that's the natural nature of the technology world. What this status guarantees is quality and governance at the present moment, not eternity.

#Kubernetes: The First Graduated Project

History records Kubernetes as the first graduated project, in March 2018 — about two years after CNCF was founded. This momentum simultaneously marked container orchestration officially becoming the main foundation of the cloud native ecosystem.

After Kubernetes, the door to graduation opened wide. Prometheus and CoreDNS followed the same year, and since then the graduated catalog has kept expanding across domains.

What's interesting is that promotions never stopped. Almost every year a new project reaches the top, and the gaps between graduations are often filled by incubating projects fighting to meet the requirements — a story we'll continue in episode 6.

#The Graduated Projects Catalog by Role

#Orchestration, Runtime, and Data Storage

At the heart of the ecosystem is Kubernetes for orchestration, containerd and CRI-O as container runtimes, and etcd as the key-value store that is the brain of cluster configuration. For storage and virtualization, there are Rook, Longhorn, and KubeVirt.

This group answers basic needs: running containers, managing them, and giving them a place to store data and the ability to run non-container workloads.

#Observability: Metrics, Logs, and Tracing

The observability field is one of the densest areas in the graduated catalog. Prometheus is the standard for metrics, Jaeger for distributed tracing, Fluentd for logs, and OpenTelemetry — officially graduated on May 21, 2026 — for the telemetry standard that unifies them all.

OpenTelemetry marked a major milestone: for the first time the observability world has a truly open and neutral foundation, with more than 12,000 contributors from thousands of companies.

#Service Proxy, DNS, and Service Mesh

In the network layer there are Envoy as a high-performance service proxy, CoreDNS as DNS for the cluster, and Linkerd and Istio as service meshes. These four projects ensure traffic between services runs fast, securely, and observably.

The term service mesh may sound complicated, but the essence is simple: it adds a control layer for communication between microservices without changing application code. You'll see this layer again in the trends discussion in episode 20.

The fact that two major service meshes are both graduated — Linkerd and Istio — reveals a distinctive CNCF pattern: healthy competition is nurtured side by side. Instead of picking a single winner, CNCF lets both develop and lets users decide.

#Delivery, Packaging, and Autoscaling

For delivering applications, CNCF has Helm as the package manager, Argo for continuous delivery, Flux and Flagger for GitOps and progressive delivery, and KEDA for event-based autoscaling. Harbor completes this group as an enterprise container registry.

GitOps — the practice of making Git the source of truth for infrastructure — will be covered specifically in episode 19. For now, just note that its main players already hold graduated status.

#Security and Other Services

In security there are OPA as the policy engine, Kyverno for Kubernetes policy, and Falco for runtime threat detection. Vitess serves MySQL database sharding, CloudEvents standardizes the event format, and Fluentd and Helm were already mentioned.

This list is certainly not complete and will keep growing. For the latest list, open cncf.io/projects — that source is always more accurate than anyone's memory.

#Seeing Graduated Projects in Real Life

#The Official Catalog and Real Clusters

The easiest way to verify the graduated list is to open cncf.io/projects. But there's another, livelier way: observing graduated projects already running in a Kubernetes cluster. The commands below show the services typically running along with their system components:

kubectl get pods -A
helm list -A
kubectl get svc -n kube-system

The kubectl get pods -A command shows all Pods across all namespaces, while helm list -A shows installed Helm releases. By comparing the service names that appear with the graduated catalog, you'll realize these technologies are already all around you every day.

Although this series doesn't require you to have a cluster, trying these commands in an environment like minikube can be an eye-opening experience. You'll see directly how graduated projects work together in a single system — a real demonstration of the landscape concept from episode 2.

#Why the Graduated Catalog Matters to You

#A Trust Signal When Choosing Technologies

For those who are learning, the graduated catalog serves as a list of technologies worth studying deeper. If you later want to specialize in a field, graduated projects are the safest starting point: big communities, mature documentation, and abundant learning material.

This catalog also helps you see career maps. Want to focus on observability? Learn Prometheus and OpenTelemetry. Interested in cloud native security? OPA, Kyverno, and Falco are the entry points. Every graduated project is a door to a specific area of expertise.

Additionally, understanding graduated projects helps you speak the same language as practitioners. When a job posting mentions CKA or experience with Prometheus, you know exactly which technologies are meant and at which layer they sit in the cloud native architecture.

#Closing

Episode 5 introduced the CNCF graduated projects catalog: from Kubernetes, the first to graduate in March 2018, to OpenTelemetry, which followed in May 2026. You also learned how to group dozens of projects by role — orchestration, observability, networking, delivery, and security.

Key takeaways:

• Graduated projects are the peak of maturity: audited, governed, and widely used.
• Kubernetes was the first graduated project in March 2018.
• OpenTelemetry graduated on May 21, 2026 as the de facto observability standard.
• The graduated catalog covers orchestration, runtime, observability, networking, delivery, and security.
• Helm, Argo, Flux, KEDA, OPA, and Falco are key names in their groups.
• cncf.io/projects is the official source for the latest list.

In episode 6 we'll go one level down: incubating and sandbox projects — from Cilium, Dapr, Crossplane, and cert-manager that are on the rise, to early experimental projects in sandbox that could become future stars.