Last updated:

Workloads

This section explains how to manage workloads.

In KubeSphere, applications run in pods. To ensure business continuity in case of pod failures, it is common to create multiple replicas of a pod for an application. Workloads are responsible for managing replicas of pods and provide the following key functionalities:

Automatically scale the number of pod replicas based on user-defined conditions.
Automatically update pods based on user-specified update strategies.
Roll back pods to previous versions.

KubeSphere supports the following types of workloads:

Deployments: Used to manage stateless applications. Multiple replicas managed by a Deployment share a common volume. When a pod replica is recreated, the new and old replicas have different names.
StatefulSets: Used to manage stateful applications. Multiple replicas managed by a StatefulSet can use independent volumes. Each pod replica has a fixed index in its name, and when a pod replica is recreated, the new and old replicas have the same name.
DaemonSets: Used to manage daemon processes. DaemonSet ensures that one pod replica runs on all or specific nodes. When a new node is added to the cluster, if the new node meets the scheduling rules, DaemonSet will automatically create a pod replica on the new node.