Deploying Db2 using a custom resource

Once the Db2 Operator is installed, you can deploy Db2 by using a custom resource (CR). Db2 provides two custom resource options for deployment: Db2uInstance and Db2uCluster, each optimized for different environments and performance needs.

Note: Db2uCluster Custom Resource is deprecated and will be discontinued in a future release. As a resolution, migrate from Db2uCluster to Db2uInstance. To migrate to Db2uInstance, perform a Db2 native backup on your Db2uCluster instance and then restore to a new Db2uInstance. For more information on this procedure, see Backing up a Db2 or Db2 Warehouse database offline and Restoring Db2 or Db2 Warehouse from an offline backup.

Db2uInstance Custom Resource for Db2: Db2uInstance provides dedicated volumes for each MLN, eliminating data sharing between nodes. This separation enhances scalability and performance, especially in cloud platforms such as Amazon Web Services (AWS) and Microsoft Azure.
The new Db2uInstance custom resource has dedicated multiple logical node (MLN) volume support that aligns with the shared-nothing architecture of Db2 for cloud-native environments.

Db2uCluster Custom Resource for Db2: The Db2uCluster custom resource was the initial offering for deploying Db2. It supports environments where multiple logical nodes (MLNs) share the same data volume within each pod. This configuration works efficiently on software-defined storage types, such as OpenShift Data Foundation (ODF) and Portworx, providing strong performance and simplified management in on-premise or hybrid cloud environments.

Whether you are deploying to on-premise with software-defined storage types or to on-cloud platforms with cloud-native storage, you can choose to use either custom resource. Db2uCluster is well-suited for on-premise and software-defined storage environments, offering reliable performance with centralized data management. Db2uInstance is designed for cloud-native deployments, where distributed storage and independent node operation provide significant performance advantages for analytics workloads.