Merge pull request cncf#842 from ppatierno/strimzi-incubation-proposal

Strimzi proposal for CNCF Incubation

proposals/incubation/strimzi.md

@@ -0,0 +1,136 @@
+# Strimzi Incubation Stage Proposal
+
+## Introduction
+
+Strimzi is a project focused on running [Apache Kafka][ak] on [Kubernetes][kube].
+Apache Kafka is the _de facto_ standard platform for building real-time data pipelines.
+It provides support for high-throughput / low-latency messaging, as well as sophisticated development options that cover all the stages of a distributed data streaming pipeline, from ingestion, through storage, and on to processing.
+But running Kafka on Kubernetes can be complex and tedious.
+Strimzi makes it easy by using the Kubernetes operator pattern and declarative deployments.
+It provides operators not just for deploying Kafka clusters, but also for managing topics, users and connectors to make sure running Apache Kafka feels Kube-native.
+
+Strimzi also integrates with many other CNCF projects such as:
+
+* Kubernetes
+* [Prometheus][prom] for metrics and monitoring
+* [Open Policy Agent][opa] for authorization
+* [KEDA][keda] for autoscaling of Kafka clients
+* [Jaeger][jaeger] and [OpenTelemetry][otel] for tracing
+
+In addition, Strimzi is used by [Knative][knative] to provide its Kafka source.
+
+Strimzi can be installed using [Helm][helm] charts or using Operator Lifecycle Manager, which is part of the [Operator Framework][operatorframework] project.
+Both the Helm charts and OLM bundles are available on [Artifact Hub][artifacthub].
+
+Strimzi joined CNCF Sandbox in September 2019
+
+## License
+
+The Strimzi project is licensed under [Apache License 2.0](https://github.com/strimzi/strimzi-kafka-operator/blob/main/LICENSE).
+
+## Statement on alignment with the CNCF mission
+
+> The Foundation's mission is to make cloud native computing ubiquitous
+
+While Kafka is _capable_ of horizontal scalability, achieving such things as zero-downtime upgrades, and scaling clusters within Kubernetes without an operator is too high a barrier for many organizations.
+This prevents ubiquitous use of Kafka anywhere with cloud native platforms such as Kubernetes.
+
+Strimzi provides declarative APIs for Apache Kafka resources, including broker clusters, Kafka Connect clusters, connectors, and Kafka MirrorMaker instances (for inter-cluster replication).
+
+Thanks to the declarative APIs, Strimzi simplifies the way to deploy a Kafka cluster on Kubernetes as well as helps to automate the day-2 operations like upgrading brokers configuration, Kafka version, and security related issues.
+
+By automating the operational burdens of running Kafka, Strimzi brings this important technology into the hands of anyone using Kubernetes.
+
+## Community
+
+In the Strimzi community, users and developers can interact in several places:
+
+* A public #strimzi channel, on the CNCF Slack workspace, with almost 1700 users.
+* Mailing lists.
+* Github issues and discussions.
+* A public "community call" every other week. The full list with all recordings is [here](https://www.youtube.com/watch?v=wK4k3cAstbQ&list=PLpI4X8PMthYfONZopcRd4X_stq1C14Rtn).
+* Forums and events. Strimzi has participated in numerous KubeCon "Office Hours" events.
+
+Strimzi encourages contributions from the community, and has a good record of merging such contributions.
+
+
+## Incubation Stage Requirements
+
+### Production usage
+
+Strimzi is used in production by a number of organizations (see [ADOPTERS.md](https://github.com/strimzi/strimzi-kafka-operator/blob/main/ADOPTERS.md) for some public references).
+Some of the public adopters are presented on the homepage of the official Strimzi [website](https://strimzi.io/).
+In addition, Strimzi maintainers are aware of further non-public users of both Strimzi and commercially supported distributions of Strimzi. 
+
+### Committers
+
+Strimzi has currently 7 maintainers from three companies (Red Hat, IBM, and CN Group).
+The full maintainer list can be found in our [GitHub repository](https://github.com/strimzi/governance/blob/master/MAINTAINERS).
+
+In addition to the maintainers, 8 other contributors have _Write_ permissions to individual repositories and act as owners for the sub-projects within these repositories.
+The full list can be found in our [GitHub repository](https://github.com/strimzi/governance/blob/main/COMPONENT-OWNERS) as well.
+
+### Commits and merged contributions
+
+Current status of the Strimzi project:
+
+* 2133 commits in the last 12 months
+* 411 contributors in the last 12 months
+* more than 4k stars and ~1.2 forks on github.
+
+See our [devstats](https://all.devstats.cncf.io/d/54/project-health-table?orgId=1&var-repogroup_name=Strimzi) for more details.
+
+### Versioning scheme
+
+Strimzi uses a 1–2 month release schedule for the "core" operators component.
+All Strimzi component releases use [semantic versioning](https://semver.org/).
+The project's release version in described in [RELEASES.md](https://github.com/strimzi/governance/blob/main/RELEASES.md).
+
+### Security processes
+
+Security is taken seriously by the Strimzi team. Any vulnerabilities should be reported to Strimzi maintainers privately, to minimize potential attacks against current users of Strimzi before they are fixed.
+Reported vulnerabilities will be investigated and patched on the next patch (or minor) release as soon as possible.
+The project's security processes are described in [SECURITY.md](https://github.com/strimzi/.github/blob/main/SECURITY.md).
+
+## Future Plans
+
+### Functionality
+
+There are several major features in development or under consideration that will further increase Strimzi capabilities.
+
+Kafka itself is migrating from ZooKeeper for metadata management to its own Raft-based quorum, known as [KRaft][kip-500].
+For Strimzi the challenge is ensure that users can access this new functionality and provide a migration path from existing ZooKeeper-based clusters.
+
+To support more sophisticated use cases, we want to improve the manageability of Kafka clusters by integrating [Cruise Control][cc] more deeply within Strimzi.
+We also intend to investigate enabling Kafka clusters to be spread over multiple Kubernetes clusters.
+
+Finally, we see that there is a lot of value in leveraging the community's operational experience by providing common tooling, standard operating procedures, and documentation.
+
+The Strimzi roadmap of new future features is available [here](https://github.com/orgs/strimzi/projects/1).
+
+### Usage and community
+
+In the short to medium term we see two main drivers of adoption. 
+
+Strimzi is already seeing significant production usage across many companies, but this is at odds with its current 0.x version number.
+Announcing Strimzi 1.0 will help signal the software's maturity, especially to those who have been held back by an apparent lack of production-readiness. 
+
+A second driver is the increased visibility of the project if Strimzi is accepted into CNCF incubation.
+Taking Knative as an example, we would like to more actively promote Strimzi as a building block for other cloud-native technologies with data streaming requirements.
+
+We hope that expanding the number and diversity of organizations using Strimzi will have a corresponding increase on the number of committers and contributions being made to the project.
+
+[ak]: https://kafka.apache.org/
+[kube]: https://kubernetes.io/
+[prom]: https://prometheus.io/
+[opa]: https://www.openpolicyagent.org/
+[keda]: https://keda.sh/
+[jaeger]: https://www.jaegertracing.io/
+[otel]: https://opentelemetry.io/
+[knative]: https://knative.dev/
+[helm]: https://helm.sh/
+[operatorhub]: https://operatorhub.io/
+[operatorframework]: https://operatorframework.io/
+[kip-500]: https://cwiki.apache.org/confluence/display/KAFKA/KIP-500%3A+Replace+ZooKeeper+with+a+Self-Managed+Metadata+Quorum
+[cc]: https://github.com/linkedin/cruise-control
+[artifacthub]: https://artifacthub.io/