Oracle beefed up its Kubernetes automation provision through its AI database suite and a new Red Hat partnership.
The updates are led by Oracle AI Database Operator for Kubernetes v2.2.0, which aims to standardize database operations with Kubernetes-centric patterns.
Oracle's new tool expands life cycle automation and reconciliation for Oracle Autonomous AI Database, across infrastructure components, backup, restore, and multitenant workloads, with finalizer-based cleanup support for improved resource life cycle management. Broader automation is also possible for single-instance databases, and Oracle's Real Application Clusters (RAC), Globally Distributed AI Database, Restart, and Data Guard Broker topologies.
Oracle RAC specifically touts improved Oracle Automatic Storage Management (ASM) life cycle management, incorporating disk and persistent volume chain (PVC) provisioning and validation. Oracle Restart introduces phased reconciliation across validation, storage, workload deployment, and finalization, and supports both static and dynamic persistent volume (PV) and PVC handling.
In another development, Oracle True Cache connects the cache provision to databases in the same Kubernetes cluster, a separate cluster, or externally. With it, the operator can generate the required configuration blobs and ConfigMaps while utilizing service registration, external-access options, validation, life cycle controls, transmission control protocol (TCP), and TCP with secure socket layer/transactions per second (SSL/TPS) connectivity.
Oracle AI Database Operator for Kubernetes v2.2.0: under the hood #
Further details on Oracle's updated Database Operator revealed improved multitenant flexibility via controller-based internal communication and configuration. This comes from the addition of pluggable database (PDB) application-user creation through Kubernetes Secrets, reconciliation of PDB initialization parameters, simpler reset-bitmask status handling, and support for transparent network substrate (TNS) topology.
A new PrivateAI controller adds phased dependency and workload reconciliation, update-lock status, and rollout tracking. It also supports TLS-secret life cycle management, virtual large language model (vLLM), and graphics processing unit (GPU)-based deployments, giving private AI services more Kubernetes-native operational control.
Oracle's REST Data Services (ORDS) now support HTTP-only edge deployments, forward and persist HTTP access logs, and bootstrap through the Instance API. Teams can also configure resource limits, metadata and Java virtual machine (JVM) options through the Kubernetes operating model.
The solution also includes a new networking preview as part of improved security and observability in the package, including secure HTTPS diagnostics, a bolstered manager security context, expanded role-based access control (RBAC) and webhooks, alongside broader test coverage.
Finally, a preview of Oracle Connection Manager Controller supports Oracle Connection Manager (CMAN) endpoints and generated routing rules.
Under the Hat #
Red Hat OpenShift on Oracle Cloud Infrastructure (OCI) was also expanded with autoscaling support to help adjust worker capacity automatically as application demand changes. The updates are available for OpenShift 4.22 and later versions.
"Autoscaling for OpenShift on OCI drives the standard OpenShift autoscaling machinery rather than replacing it, and operates alongside the autoscaling controls you already use," Oracle Senior Software Engineer YuTing Peng explained. "When pods cannot be scheduled, the cluster autoscaler increases the replica count on the relevant machine deployment. Cluster API creates the corresponding machine and machine set resources, and the Cluster API Provider for OCI provisions the OCI compute instances behind them. You set the boundaries once, on an OCIClusterAutoscaler custom resource that carries minNodes and maxNodes, and the existing controllers do the rest."
The OpenShift update comes alongside Red Hat compatibility with the aforementioned AI Database Operator Kubernetes updates, which also support Oracle Linux Cloud Native Environment, Google Kubernetes Engine (GKE), Azure Kubernetes Service (AKS), and Amazon Elastic Kubernetes Service (EKS).