Alauda Container Platform

Workload Types

In addition to creating cloud-native applications via the Applications module, workloads can also be directly created in Container Platform > Workloads:

  • Deployment: The most commonly used workload controller for deploying stateless applications. It ensures a specified number of Pod replicas are running, supporting rolling updates and rollbacks, ideal for stateless services like web servers and APIs.

  • DaemonSet: Ensures a Pod runs on every node (or specific nodes) in the cluster. Pods are automatically created when nodes join and removed when nodes leave. Ideal for node-level tasks such as logging agents and monitoring daemons.

  • StatefulSet: A workload controller for managing stateful applications. It provides stable network identities (hostname) and persistent storage for each Pod, ensuring data consistency even during rescheduling. Suitable for databases, distributed caches, and other stateful services.

  • CronJob: Manages time-based Jobs using cron expressions. The system automatically creates Jobs at scheduled intervals, ideal for periodic tasks like backups, report generation, and cleanup jobs.

  • Job: A workload for running finite tasks. It creates one or more Pods and ensures a specified number of successful completions before terminating. Suitable for batch processing, data migrations, and other one-time operations.

In addition to creating workloads via the web console, Kubernetes also supports direct management of lower-level resources via CLI tools::

  • Pod: The smallest deployable unit in Kubernetes. A Pod can contain one or more tightly coupled containers sharing storage, network, and lifecycle. Pods are typically managed by higher-level controllers (e.g., Deployments).
  • Container: A standardized unit encapsulating application code and dependencies, ensuring consistent execution across environments. Containers run within Pods and share the Pod's resources.