CNCF Graduated Project Update Focuses on Landscape Diversity and Maturity at KubeCon EU
MMS • Olimpiu Pop
Article originally posted on InfoQ. Visit InfoQ
The CNCF’s charter defines its mission: “is to make cloud-native computing ubiquitous”. Their supported technologies aim to allow organizations to build public, private, or on-premise clouds based on loosely coupled systems that are resilient, manageable, and observable. During KubeCon EU, the CNCF provided a status update of the graduated projects.
As described in their graduation criteria, “CNCF projects have a maturity level of sandbox, incubating, or graduated, which corresponds to the Innovators, Early Adopters, and Early Majority tiers of the Crossing the Chasm diagram. The maturity level is a signal by CNCF as to what sorts of enterprises should be adopting different projects. Projects increase their maturity by demonstrating their sustainability to CNCF’s Technical Oversight Committee: that they have adopted, a healthy rate of changes, and committers from multiple organizations; have adopted the CNCF Code of Conduct; and have achieved and maintained the Core Infrastructure Initiative Best Practices Badge. Full details are in the Graduation Criteria“.
The CNCF updates were presented by the KubeCon program committee, consisting of Jasmine James, Emily Fox, and Ricardo Rocha:
Envoy: Originally developed at Lyft, it is an edge and service proxy promising high performance with a small footprint. Among its initiatives, it just introduced Envoy Gateway, an Envoy-based ingress gateway that builds on the work undertaken within the Contour and Emissary-ingress projects. This will help reduce redundant work going on in the community but also make Envoy a much stronger out-of-the-box option for development and platform teams. The GitHub project has almost 20K stars and it is supported by almost 900 individuals.
Jaeger: Originally developed at Uber as a monitoring and troubleshooting tool for microservices-based distributed systems. Its latest features include distributed context propagation and transaction monitoring, service dependency analysis, root cause analysis, and performance/ latency optimization.
Among the current initiatives: Adaptive sampling fully automated dynamic control, going all-in on OpenTelemetry by adopting OpenTelemetry SDKs, and by retiring native Jaeger SDKs. Another ongoing effort is focused around adopting Cassandra and Elasticsearch, an initiative where the help of the community would be highly appreciated. The project has almost 16K stars and is powered by the contributions of 100 individuals.
Prometheus: originally built at SoundCloud, this is a systems monitoring and alerting toolkit that collects metrics from configured targets at given intervals. It joined CNCF in 2016 as the second hosted project after Kubernetes.
Its initiatives include the Prometheus conformance program targeted at ensuring interoperability between different projects and vendors in the Prometheus monitoring space. Another initiative is the Prometheus agent mode, a feature targeted to scenarios where no local storage is needed or possible (like edge nodes). In these situations writing to remote storage is available (like Thanos or cortex). Prometheus’ GitHub project counts almost 43K stars and has 701 contributors.
Linkerd: It presents itself as being the smallest, simplest service mesh in the world. To provide extra assurance that it is ready for production, it passed an external security audit. Among its most notable adopters, Xbox Cloud Gaming can be found. Its current initiatives count automatic cross-cluster failover capabilities that promise automatic traffic redirection from a failing service to different ones. The project gathered 8.5K stars so far and encapsulates the effort of 255 developers.
TiKV: A cloud-native distributed key-value storage promising single-digit millisecond latency. Among the ongoing initiatives is raft engine availability, which should reduce bandwidth by 30%, hence making it more cost-effective on the cloud. Pessimistic transactions are 20% faster in the tool. The project has a bit more than 11K stars and encapsulates the work of 360 contributors.
TUF: a framework for secure content delivery of updates. It protects from various types of supply chain attacks and promises resilience to compromise. One of the most important initiatives on the project side was the python refactor that resulted in a much smaller and robust codebase to maintain, but also in a more ergonomic API. The efforts of the 67 contributors got awarded 1.4K stars on the GitHub repository.
ROOK: Cloud-native storage operator for Kubernetes. Release 1.9 promises to deliver support for Ceph Quincy, the most recent release of Quincy. The GitHub project has almost 10K stars and 385 contributors.
HELM: the widely adopted package manager for finding, and sharing apps on Kubernetes. One of the most requested features from the HELM community was Open Container Initiative support, so starting with version 3.8.0 of the helm you can store charts in container registries. Its GitHub project is powered by 562 contributors and has almost 22K stars.
Seven years after its inception, CNFC’s graduate landscape includes sixteen projects. More than that, the number of incubated projects reached above 30, making the CNFC landscape more and more diverse. More than just software, the foundation promotes cloud-native learning through certification programs as well, the latest of which focused on Prometheus.