EDB is thrilled to announce that our brand new product, EDB Postgres Distributed for Kubernetes (PGD4K), is now officially GA! Last month, EDB announced the preview of PGD4K, which provides a Kubernetes operator that simplifies creation and management of EDB Postgres Distributed (PGD) clusters inside Kubernetes. Thanks to the efforts of our preview participants in getting our product to the finish line, we are pleased to announce that PGD4K is now generally available, providing organizations with a solution for an active-active architecture across multiple K8s clusters.
By extending the functionality of the popular open source CloudNativePG Kubernetes Operator from EDB, the PGD4K Operator brings the best-in-class high availability delivered by EDB Postgres Distributed to Kubernetes clusters by providing multi-master replication and geo-sharding of data. As a result, you can leverage robust high availability without the need for specialized domain knowledge by enabling the PGD4K Operator to automate the deployment and other operational aspects of the PGD cluster.
The PGD4K Operator provides the ability to create PGD clusters across multiple regions, enabling customers to geographically distribute their data. This can aid in meeting data sovereignty requirements and/or improving application performance by housing data close to where you need it.
Until now, users running EDB Postgres for Kubernetes could not leverage the advanced high availability capabilities of PGD, while at the same time taking advantage of the highly available, multi-cluster architecture that Kubernetes provides. The PGD4K operator closes this gap by providing the following features:
- Deployment of EDB Postgres Distributed clusters: Let PGD4K easily handle deployment of PGD within your Kubernetes cluster (versions 5 and later).
- Additional self-healing capabilities: PGD4K encapsulates best practices and operational knowledge of PGD to provide self-healing capabilities by automating recovery and rejoin of failed PGD nodes If a node in a PGD cluster fails, it is automatically restarted, removing the operational overhead associated with having to manually restart the node.
- Support for Local Persistent Volumes with PVC templates: With local persistent volumes, storage is physically attached directly to the nodes in your Kubernetes clusters. This type of storage is particularly useful for applications that require direct access to the data layer, as it offers a significant performance improvement when compared to network-attached storage.
- Continuous object store store backups: Postgres continuous backups provide a robust and flexible backup strategy. With support for storing backups in an object store, customers can now take advantage of robustness and reduced cost benefits.
- Support for pausing and resuming a PGD cluster: When a cluster is “paused,” the underlying data remains on the persistent volume storage, but the pods that make up the cluster are removed. This allows the user to maintain the database on the underlying persistent storage without having to pay for the compute resources needed to keep the pods up and running.
Summary
From automated deployment across regions, to enhanced fault tolerance and automatic failover mechanisms, to streamlined disaster recovery and seamless scalability of standby nodes, the PGD4K Operator provides unparalleled reliability to developers using Kubernetes. As a result, embracing EDB Postgres Distributed for Kubernetes isn't just a best practice—it's essential for organizations looking to thrive in today's hypercompetitive digital landscape.
For more information on EDB Postgres Distributed for Kubernetes operation, check out our documentation and the EDB Postgres for Kubernetes webpage.
Additionally, you can get started today on EDB Postgres Distributed and experience continuous high availability Postgres—sign up for an EDB account in less than 60 seconds