BT

Facilitating the Spread of Knowledge and Innovation in Professional Software Development

Write for InfoQ

Topics

Choose your language

InfoQ Homepage News KubeCon and CloudNativeCon 2018 Summary: Kubernetes 1.13, Envoy Update, and New Hosted Projects

KubeCon and CloudNativeCon 2018 Summary: Kubernetes 1.13, Envoy Update, and New Hosted Projects

At the recent KubeCon and CloudNativeCon North America events, held in Seattle, USA, a series of updates were provided about the Cloud Native Computing Foundation (CNCF) hosted projects. Highlights included the release of Kubernetes 1.13, an overview of the progress of the Envoy Proxy project (and a discussion of its increasing ubiquity in the cloud native stack), and the inclusion of the Rook storage orchestration and Harbor image registry projects.

The event co-chairs, Liz Rice, technology evangelist at Aqua, and Janet Kuo, software engineer at Google, opened the event with a keynote summarising the latest developments within the cloud native ecosystem. According to the CNCF, cloud native technologies empower organisations to build and run "scalable applications in modern, dynamic environments such as public, private and hybrid clouds".

The CNCF aims to provide "open source components of a full stack cloud native environment", and it does this by hosting a range of projects at various "maturity levels" -- sandbox, incubating and graduated -- that are inspired by Geoffrey Moore's "Crossing the Chasm" (which the InfoQ team also use for curating topics). The CNCF provides facilities such as a neutral home for projects, funding for improving documentation, knowledge share and conference facilitation, marketing, legal services and access to expertise from industry luminaries and end user communities.

CNCF project services and maturity levels.
Project services and maturity levels (image taken from CNCF website)

Building on the successful launch of Kubernetes 1.12, version 1.13 was released in tandem with the event. New functionality included: simplified Kubernetes cluster management with kubeadm in GA; the Container Storage Interface (CSI) is now GA; CoreDNS is now replacing kube-dns as the default DNS server for Kubernetes; and several other small enhancements.

The introduction of kubeadm in GA will make operational work with Kubernetes easier. kubeadm handles the bootstrapping of production clusters on existing hardware and can configure the core Kubernetes components in a "best-practice-manner" to provide "a secure yet easy joining flow for new nodes and supporting easy upgrades". Further improving cluster management, "kubectl diff" has also graduated to beta, which allows users to run a kubectl command to view the difference between a locally declared object configuration and the current state of a live object.

The CSI specification has reached a 1.0 status, and is now GA after being introduced as alpha in v1.9 and beta in v1.10. With CSI, the Kubernetes volume layer becomes more extensible which provides an opportunity for third party storage providers to write plugins that interoperate with Kubernetes without having to modify the core code. Topology Aware Volume Scheduling is now stable, which makes the scheduler aware of a Pod's volume's topology constraints, such as zone or node. Additionally, raw block device using persistent volume source is graduating to beta, which makes non-networked raw block devices available for consumption via a Persistent Volume Source.

After being announced as a GA release in Kubernetes 1.11, CoreDNS is now replacing kube-dns as the default DNS server for Kubernetes. CoreDNS is a general-purpose DNS server that provides an extensive (but backwards-compatible) integration with Kubernetes. The release blog post states that CoreDNS has "fewer moving parts than the previous DNS server, since it's a single executable and a single process, and supports flexible use cases by creating custom DNS entries".

Kubernetes 1.13 is available for download on GitHub and can also be installed now using kubeadm. For readers new to Kubernetes, there are a series of interactive tutorials on the Kubernetes website, and the KataCoda tutorials are also a popular resource.

Turning attention to other CNCF hosted projects, Rice discussed that the Prometheus monitoring and metric collection platform had become a "graduated" project this year. As reported previously on InfoQ, there is also a lot of associated activity within this space, with Grafana adding log data correlation to time series metrics. Fluentd, the open source data collector for unified logging, also saw additional support for Splunk and Amazon Kinesis added. In regard to distributed tracing, Open Tracing saw Lua support added, and the Jaegar platform saw the addition of a "Jaeger Operator" to reduce the operational overhead of running this on Kubernetes.

The Envoy Proxy project also graduated, and InfoQ has provided coverage of the inaugural EnvoyCon that ran on the Monday before the main event. It was evident from the amount of related conference presentations and vendor products (including AWS App Mesh), that Envoy has joined Kubernetes in becoming part of the cloud native data plane, often referred to as the cloud native "operating system" fabric. The Linkerd service mesh team also released v2 (based on the previously released Conduit project), which allows the execution of Linkerd in a "service sidecar" model.

In regard to cloud native storage, the Rook project became hosted at the incubating level. Rook is an open source cloud-native storage orchestrator for Kubernetes, providing the "platform, framework, and support for a diverse set of storage solutions to natively integrate with cloud-native environments". Vitess, a database clustering system for horizontal scaling of MySQL through the use of generalized sharding outside of application logic, saw the release of v3, and included functionality such as VReplication, Promethues monitoring integration, and a series of performance enhancements.

In miscellaneous news, the NATS project, a "simple, high performance open source messaging system" for cloud native applications, saw the inclusion of secure multi-tenancy and network topology optimisations over the past year. The Harbor project, "an open source trusted cloud native registry project that stores, signs, and scans content" was also introduced as a incubating project.

Community and commercial interest within the cloud native ecosystems and the CNCF hosted project has clearly grown over the past two years. Approximately 8000 attendees were present at this latest installment of the conference (and related training and workshop days); a figure that has grown significantly since the event was held in Seattle in 2016, which attracted just over 1000 attendees. There are also currently 160+ CNCF meetups being hosted globally, and this year $300,000 in diversity scholarships was raised to enable attendees from diverse and minority background to make the journey to Seattle for the event.

The full schedule and many session slides can be found on the KubeCon / CloudNativeCon Sched page, and videos of presentations can be found on the CNCF's YouTube channel.

Rate this Article

Adoption
Style

BT