Concepts

Basic terms

Cluster: A cluster is a collection of cloud resources that are required to run containers, including KEC instances, SLB instances, and VPCs.

Dedicated cluster: A dedicated cluster is a cluster on which you have full management permissions. Master and etcd nodes in a dedicated cluster are deployed in your own VPC.

Managed cluster: A managed cluster is a cluster in which master and etcd nodes are managed by KCE. You only need to create worker nodes. This frees you from cluster management.

Serverless cluster: A serverless cluster is built on KCI and fully compatible with the Kubernetes ecosystem. You can directly deploy containerized applications and manage them in a Kubernetes-native way without the need to purchase or manage underlying nodes.

Node: A node is a single server registered to a container cluster, for example, a KEC, EPC, or GEC instance.

Container: A container is a running instance that is created from a Docker image. Multiple containers can run on one node.

Service: A service defines a group of containers with the same configurations and the rules to access the containers. A service is the deployable unit in the microservice architecture and the minimum running unit for your business applications.

Instance: An instance is composed of one or more containers, and corresponds to a Kubernetes pod. These containers share the same storage and network spaces.

Configuration item: A configuration item is a collection of configurations and allows you to manage different environments and different applications conveniently.

Image: An image is a package of the operating system, operating environment, and applications that are required to deploy a service. You can use an image to deploy a replica of a service. The definition of an image is the same as that in Kubernetes. Docker images are stored in image repositories.

Image repository: An image repository is used to store Docker images.

Kubernetes terms

Namespace: A namespace is a separate virtual space. Resources in different namespaces are isolated. You can create multiple namespaces in one Kubernetes cluster. The cluster can manage resources by namespace.

Pod: A pod is the minimum unit for deploying Kubernetes applications or services. A pod is composed of one or more containers that share the same storage and network spaces.

Deployment: A Deployment provides declarative definitions for pods and ReplicaSets. You can use a Deployment to deploy stateless applications. Based on your business requirements, you can define the number of replicas, scheduling policies, and update policies for pods running in a Deployment.

StatefulSet: A StatefulSet is the workload object used to manage stateful applications. Like a Deployment, a StatefulSet manages pods that are based on identical container specifications. Unlike a Deployment, a StatefulSet maintains a persistent identifier for each of the pods. The identifier of a pod is retained after the pod is rescheduled to another node or restarted. If an application does not require persistent identifiers or ordered deployment, deletion, or scaling, we recommend that you use a Deployment.

DaemonSet: A DaemonSet ensures that all or some nodes run replicas of a pod. When nodes are added to the cluster, the pod is added to them. When nodes are removed from the cluster, the pods on these nodes are reclaimed. If you delete a DaemonSet, the pods that it created are deleted. Typically, you can use a DaemonSet to run a cluster storage daemon or a log collection daemon on every node.

Job: A job creates one or more pods and ensures that a specified number of them successfully terminate.

CronJob: A CronJob creates jobs on a repeating schedule. It runs a job periodically on a given schedule.

Service: A Service manages Layer 4 network access in a cluster. Based on the Layer 4 network, a Service exposes container capabilities in a cluster.

Ingress: An Ingress is a collection of rules for authorizing access to cluster services. An Ingress can be configured to provide externally-reachable URLs, load balancing services, SSL services, and name-based virtual hosts. You can send a POST request for Ingress resources to the API server to request an Ingress. An Ingress controller is responsible for implementing an Ingress, usually with a load balancer, though it may also configure your edge router or additional frontends to help handle the traffic in HA mode.

ConfigMap: A ConfigMap is used to store non-confidential data in key-value pairs. Pods can consume ConfigMaps as environment variables, command-line arguments, or configuration files in a volume. A ConfigMap allows you to decouple environment-specific configuration from your container images, so that your applications are easily portable.

Secret: A Secret is a key-value pair for storing sensitive information, such as passwords, OAuth tokens, and SSH keys.

Volume: A Kubernetes volume is similar to a Docker volume. However, a Docker volume is specific to a container, but a Kubernetes volume is specific to a pod. Volumes declared in a pod are shared by all containers in the pod.

PV: A persistent volume (PV) is a piece of storage in a cluster that is provisioned by an administrator or dynamically provisioned by using a StorageClass.

PVC: A persistent volume claim (PVC) is a request for storage by a user. It is similar to a pod. Pods consume node resources and PVCs consume PV resources. Pods can request specific levels of resources, including CPUs and memory. PVCs can request specific sizes and access modes.

StorageClass: A StorageClass describes a storage type. You can define different storage types for a cluster as the cluster administrator.