A set of physical and/or virtual machines called nodes. Each cluster contains at least one master node and zero or more worker nodes that work together to run containerized applications.
A lightweight and executable package that holds the software to run an application. Containers are portable, stand-alone and independent from their environment. Containers are created from container images.
An unchangeable, static file that holds the executable code and other software dependencies to run an application.
A software component installed on each node that focuses on running and managing containers on a host operating system.
A component that manages container orchestration in a Kubernetes cluster, including the etcd storage, API server, scheduler and controller manager.
An open-source platform that develops and packages applications into containers. Docker then deploys and runs the containers. Docker can be considered a container runtime. For further information about Docker and other container runtimes, check out this Kubernetes Meetup presentation here.
The hardware and software setting where developers run, deploy, program and/or build applications.
An open-source, key-value system that stores the cluster’s metadata and configuration information. Pronounced et-see-dee.
A collection of YAML files installed in a cluster that describe a set of Kubernetes resources for the purposes of deploying an application. A Helm chart allows templating and parameterizing application deployment.
Identity and Access Management (IAM) is an AWS security measure in which each user’s role has a profile that grants them access to certain services, based on their role in the enterprise. The term IAM can also be used in a wider sense to refer to any identity and access management system, not necessarily specific to AWS.
Ingress or Ingress Rule
An API object that routes external traffic to a cluster’s services, ultimately giving an end user specified access to the cluster, typically via HTTP.
A detailed specification that describes a particular variation of an object. Sometimes, the term “instance” is also used to describe a virtual or physical server; this usage originates from AWS EC2 service, which calls virtual machines “EC2 Instances.”.
An external client that communicates with a cluster’s control plane to create new deployment objects and manage the cluster’s applications.
A service provided by a third-party company that manages, deploys and maintains an enterprise’s Kubernetes operations.
A file in JSON or YAML format that describes and specifies the desired state of a Kubernetes API object.
Smaller, independent applications that are broken down into fewer lines of code and cohesively contribute to run one application.
Used to separate and organize objects into groups in a cluster by user, team, project or other category label.
A physical or virtual worker machine, separated into two categories: master and worker nodes. Master nodes run system components in a cluster, while worker nodes run application containers.
The representation of a group of containers that contains shared storage and resources instructing how to run the containers.
Role-Based Access Control is a framework that restricts a user’s access to the Kubernetes API based on their role within an enterprise.
An abstraction that exposes an application, then connects the application’s set of pods to a name and IP address, and directs external traffic across the pods.
A tool that securely manages service-to-service communication over a network while maintaining scalability.
A data serialization language that stores information about key-value pairs. YAML is a superset of JSON that is more lightweight and emphasizes easy formatting and readability.