This page describes a high-level view of the main priorities for the project, agreed as a result of collaboration between Cilium’s committers and the broader community. You’ll also find here some pointers on how you can influence the roadmap.

Major Feature Status

eBPF Networking

Stable (Roadmap Details)

Kubernetes CNI


Load Balancing


Network Policy


Kube-proxy Replacement


Egress Gateway


Multi-Cluster (ClusterMesh)

Stable (Roadmap Details)

Hubble Observability

Stable (Roadmap Details)

Service Mesh

Stable (Roadmap Details)

Tetragon Security


“Stable” means that the feature is in use in production (though advanced features may still be in beta or in development).

Release Cadence

We aim to make 2-3 point releases per year of Cilium and its core components (Hubble, Cilium CLI, Tetragon, etc). We also make patch releases available as necessary for security or urgent fixes.

Focus Areas

For a finer-granularity view, and insight into detailed enhancements and fixes, please refer to issues on GitHub.

Welcoming New Contributors

As a CNCF project we want to make it easier for new contributors to get involved with Cilium. This includes both code and non-code contributions such as documentation, blog posts, example configurations, presentations, training courses, testing and more. Check the Development documentation to understand how to get involved with code contributions, and the Get Involved guide for guidance on contributing blog posts, training and other resources.

CNCF Graduation

Cilium has applied for CNCF Graduation, please add your support on the PR!

Cilium Service Mesh

Our eBPF-accelerated Service Mesh is the main focus for major enhancement, and it’s a natural evolution of Cilium’s networking capabilities. We released a beta at the end of 2021 and had very valuable feedback from our user community. The next steps we’d like to take for Cilium Service Mesh (in no particular order) are:

  • Graduating Prometheus metrics and OpenTelemetry collector to stable

  • Using Kubernetes as service mesh control plane

    • Simple to use sidecar-free service mesh configured using Kubernetes Services and Ingress with support for additional annotations

  • Graduating EnvoyConfig CRD to stable

  • Extended sample Grafana dashboards for L7 visibility

  • SMI integration

  • SPIFFE integration

  • Gateway API Integration

  • Next-generation mutual authentication datapath framework

    • Support for integrated runtime identity

    • SSL-based mutual authentication

    • Support for any network protocol

  • Performance benchmarking


Core ClusterMesh is stable and widely adopted. Future extensions include:

Advanced Networking Features

There are several advanced networking features currently in beta, several of these are already in production use with a set of adopters. We expect the following features to graduate to stable:

Hubble Observability

Hubble provides visibility into network flows through the Hubble CLI (stable) and UI (beta), with support for Prometheus and OpenTelemetry metrics. Areas of focus currently include:

CI Test Improvements

We have a comprehensive set of tests running in CI, but several contributors are currently working on CI improvements to make these more reliable and easier to maintain. This is a good area to get involved if you are interested in learning more about Cilium internals and development.

Tetragon Security

Tetragon provides security observability and runtime enforcement through the JSON events and the Tetragon CLI for things like process execution, file access, network observability, and privileged execution.

Codebase modularization

As the project is growing in complexity it is becoming increasingly important to be able to divide it into more manageable chunks. To achieve this, we’re working on modularizing the codebase and going from a tightly coupled design (one large initialization and configuration) to a more loosely coupled design of mostly self-contained modules. This will make Cilium internals easier to comprehend, test and extend.

Contributions in this area are very welcome. To get started, take a look at the Guide to the Hive documentation and the issues referenced from modularization meta issue. If you have any questions or ideas please join us on the #sig-modularization channel on Cilium Slack.

Influencing the Roadmap

You are welcome to raise feature requests by creating them as GitHub issues. Please search the existing issues to avoid raising duplicates; if you find that someone else is making the same or similar request we encourage the use of GitHub emojis to express your support for an idea!

The most active way to influence the capabilities in Cilium is to get involved in development. We label issues with good-first-issue to help new potential contributors find issues and feature requests that are relatively self-contained and could be a good place to start. Please also read the Development for details of our pull request process and expectations, along with instructions for setting up your development environment.

We encourage you to discuss your ideas for significant enhancements and feature requests on the #development channel on Cilium Slack, bring them to the Weekly Community Meeting, and/or create a CFP design doc.

This roadmap does not give date commitments since the work is dependent on the community. If you’re looking for commitments to apply engineering resources to work on particular features, one option is to discuss this with the companies who offer commercial distributions of Cilium and may be able to help.

Changes to this Roadmap Page

This roadmap page will be updated through documentation pull requests in the usual way, but the Cilium committers should be consulted beforehand about anything other than trivial fixes.