| Automated bin packing |
Increases resource utilization and cost savings using a mix of critical and best-effort workloads. |
| Batch execution |
Manages batch and continuous integration workloads and automatically replaces failed containers, if configured. |
| Cloud Controller Manager |
A Kubernetes control plane component that embeds cloud-specific control logic, allowing integration with cloud providers' APIs. |
| Cluster |
A set of worker machines (nodes) that run containerized applications. Every cluster has at least one worker node. |
| Container Orchestration |
The process of automating the container lifecycle of containerized applications. |
| Container Runtime |
The software responsible for running containers. |
| Control Loop |
A non-terminating loop that regulates the state of a system. |
| Control plane |
The container orchestration layer that exposes APIs to define, deploy, and manage containers. |
| Controller |
Control loops in Kubernetes that watch the cluster state and make changes to move it closer to the desired state. |
| Data (Worker) Plane |
The layer providing capacity (CPU, memory, network, storage) for containers to run and connect to a network. |
| DaemonSet |
Ensures a copy of a Pod is running across a set of nodes in a cluster. |
| Declarative Management |
Expressing the desired state of a resource, allowing Kubernetes to actively maintain the observed state. |
| Deployment |
An object that provides updates for Pods and ReplicaSets, suitable for stateless applications. |
| Designed for extensibility |
Ability to add features to a cluster without modifying source code. |
| Docker Swarm |
A tool for automating the deployment of containerized applications, designed to work with Docker tools. |
| Ecosystem |
A composition of services, support, and tools available within a given environment or platform. |
| etcd |
A highly available key-value store containing cluster data, serving as the source of truth for cluster state. |
| Eviction |
The process of terminating one or more Pods on Nodes. |
| Imperative commands |
Directly creating, updating, and deleting live objects. |
| Imperative Management |
Defining actions and steps to reach a desired state. |
| Ingress |
An API object managing external access to services in a cluster, typically for HTTP. |
| IPv4/IPv6 dual stack |
Assigning both IPv4 and IPv6 addresses to Pods and Services. |
| Job |
A finite or batch task that runs to completion. |
| Kubectl |
The command-line tool for communicating with a Kubernetes cluster's control plane via the Kubernetes API. |
| Kubelet |
The primary "node agent" running on each node, responsible for managing containers described in PodSpecs. |
| Kubernetes |
An open-source platform for container orchestration, developed by Google and maintained by CNCF. |
| Kubernetes API |
An application serving Kubernetes functionality through a RESTful interface and storing the cluster state. |
| Kubernetes API Server |
Validates and configures data for Kubernetes API objects (pods, services, etc.) and serves REST operations. |
| Kubernetes Controller Manager |
Manages control processes monitoring cluster state and ensuring it aligns with desired state. |
| Kubernetes Cloud Controller Manager |
Embeds cloud-specific control logic, enabling cloud provider API integration. |
| Kubernetes Proxy |
A network proxy running on each node to maintain network rules allowing communication to Pods in the cluster. |
| kube-scheduler |
A control plane component selecting nodes for pending Pods to run on. |
| Label Selector |
Allows filtering resources based on labels. |
| Labels |
Tags objects with identifying attributes. |
| Load balancing |
Balances traffic across Pods for performance and high availability. |
| Marathon |
An Apache Mesos framework for scaling container infrastructure. |
| Namespace |
An abstraction for resource isolation within a single cluster. |
| Node |
The worker machine in a Kubernetes cluster running user applications. |
| Nomad |
A cluster management and scheduling tool supporting Docker and other applications on various platforms. |
| Object |
An entity in the Kubernetes system, representing the state of your cluster. |
| Persistence |
Ensures an object exists in the system until modified or removed. |
| Preemption |
Logic in Kubernetes helping pending Pods find suitable Nodes by evicting lower-priority Pods. |
| Self-healing |
Automatically restarting, replacing, rescheduling, or terminating failing or unresponsive containers. |
| Service |
An abstraction for exposing an application running on a set of Pods as a network service. |
| Service Discovery |
The process of discovering Pods using IP addresses or a single DNS name. |
| StatefulSet |
Manages the deployment and scaling of a set of Pods with guarantees about ordering and uniqueness. |
| Storage |
A data store supporting persistent and temporary storage for Pods. |
| Storage Orchestration |
Automatically mounting chosen storage systems for containers. |
| Pod |
The smallest Kubernetes object, representing a single instance of an application running in a cluster. |
| Proxy |
In computing, a server acting as an intermediary for a remote service. |
| ReplicaSet |
A controller aiming to maintain a set of replica Pods running at all times. |
| Workload |
An application running on Kubernetes. |