Automatically provision and consolidate the Nodes in your cluster to adapt to demand and optimize cost.

Production-Grade Container Orchestration

FEATURE STATE: Kubernetes v1.27 [stable] An increasing number of systems leverage a combination of CPUs and hardware accelerators to support latency-critical execution and high-throughput parallel computation. These include workloads in fields such as telecommunications, scientific computing, machine learning, financial services and data analytics. Such hybrid systems comprise a high performance environment. In order to extract the best performance, optimizations related to CPU isolation, memory and device locality are required. However, in Kubernetes, these optimizations are handled by a disjoint set of components.

FEATURE STATE: Kubernetes v1.29 [stable] Controlling the behavior of the Kubernetes API server in an overload situation is a key task for cluster administrators. The kube-apiserver has some controls available (i.e. the --max-requests-inflight and --max-mutating-requests-inflight command-line flags) to limit the amount of outstanding work that will be accepted, preventing a flood of inbound requests from overloading and potentially crashing the API server, but these flags are not enough to ensure that the most important requests get through in a period of high traffic.

This page explains how to configure the kubelet's cgroup driver to match the container runtime cgroup driver for kubeadm clusters. Before you begin You should be familiar with the Kubernetes container runtime requirements. Configuring the container runtime cgroup driver The Container runtimes page explains that the systemd driver is recommended for kubeadm based setups instead of the kubelet's default cgroupfs driver, because kubeadm manages the kubelet as a systemd service.

Device plugins let you configure your cluster with support for devices or resources that require vendor-specific setup, such as GPUs, NICs, FPGAs, or non-volatile main memory.

Kubernetes core components such as the API server, scheduler, and controller-manager run on a control plane node. However, add-ons must run on a regular cluster node. Some of these add-ons are critical to a fully functional cluster, such as metrics-server, DNS, and UI. A cluster may stop working properly if a critical add-on is evicted (either manually or as a side effect of another operation like upgrade) and becomes pending (for example when the cluster is highly utilized and either there are other pending pods that schedule into the space vacated by the evicted critical add-on pod or the amount of resources available on the node changed for some other reason).

FEATURE STATE: Kubernetes v1.26 [alpha] (enabled by default: false) This page shows how to migrate nodes to use event based updates for container status. The event-based implementation reduces node resource consumption by the kubelet, compared to the legacy approach that relies on polling. You may know this feature as evented Pod lifecycle event generator (PLEG). That's the name used internally within the Kubernetes project for a key implementation detail.

In a Kubernetes cluster, a node can be shut down in a planned graceful way or unexpectedly because of reasons such as a power outage or something else external. A node shutdown could lead to workload failure if the node is not drained before the shutdown. A node shutdown can be either graceful or non-graceful. Graceful node shutdown The kubelet attempts to detect node system shutdown and terminates pods running on the node.

This website contains documentation for the current version of Kubernetes and the four previous versions of Kubernetes. The availability of documentation for a Kubernetes version is separate from whether that release is currently supported. Read Support period to learn about which versions of Kubernetes are officially supported, and for how long.