Authors: This blog was co-written by Simon Notley, Senior Product Manager, and Lizzy Nguyen, Principal Product Marketing Manager.
Automation and single pane of glass management and observability across hybrid data estates.
Last month, EnterpriseDB (EDB) announced the EDB Postgres AI Software Deployment, a single software installation enabling you to deploy, manage, and scale EDB Postgres AI in any self-managed, hybrid environment or cloud.
EDB Postgres AI consolidates structured and unstructured data in a single multi-model data platform to accelerate transactional, analytical, and AI workloads. It also unlocks a number of new capabilities, including the new Hybrid Control Plane, which enables a hybrid database-as-a-service (DBaaS) experience with automation and advanced observability across 200+ metrics to provide a cloud-like experience – even in your private data center.
Read on for a deeper dive into some of the key features of the Hybrid Control Plane, or watch it in action in this demo by Senior Product Manager, Simon Notley.
Cloud Scale in Hybrid Environments
Today, DBAs spend 30-50% of their time on administrative tasks that could be automated. However, that automation is often only possible in public cloud environments that limit deployment flexibility and can risk data sovereignty and full control of your most precious asset – your data. Modern enterprises manage data across multi-cloud and on-prem deployments. Without a unified solution, using multiple tools for different deployments leads to data silos and adds administrative burden.
EDB’s new Hybrid Control Plane enables cloud agility in self-hosted environments by providing a single place to deploy, manage, and scale Postgres for transactional, analytics, and AI workloads.
The Hybrid Control Plane is a Kubernetes-driven, centralized management and automation solution that boosts productivity up to 30%. It automates time-consuming and expensive administrative functions like backups, provisioning, and point-in-time recovery. Users can monitor, observe, and respond to issues in real-time with visibility into 200+ metrics, keeping databases secure and enabling up to 99.999% availability. With query diagnostics, you can identify problems and bottlenecks up to 5x faster and accelerate application performance up to 8x. Plus, a suite of built-in migration tools allows you to seamlessly migrate from Oracle or Postgres in days instead of weeks or months – without significant application rewrites.
This orchestration solution runs in a Kubernetes cluster (currently available in preview on Red Hat OpenShift and Amazon EKS), making it portable to run wherever you like – meaning organizations that run on-premises or leverage hybrid environments for cost savings, performance, or data sovereignty requirements can now achieve a cloud-like experience without infrastructure lock-in.
Let’s dive into the user interface to see how you can leverage the Hybrid Control Plane.
Single Pane of Glass: Estate View
The Estate View allows you to see all your Postgres in one place with a quick glance into metadata, top-level metrics, and important alerts. Not only do you have visibility into your EDB Postgres deployed by the Hybrid Control Plane in Kubernetes, but also your existing Postgres running on virtual or physical machines. Support for third party Postgres services such as Amazon RDS for Postgres, is coming in 2025, creating a single pane of glass for Postgres without migrating any of your existing clusters. You also have a view of your current Postgres migrations, which you can read more about in the 'Data Migrations' section below.
From here, you can search, filter, and sort your clusters to find your projects quickly or customize the metrics shown in the Estate View. It provides easy access to click into your different clusters to dive into your workloads.
Creating a Cluster
Creating a new cluster takes only a few moments. From the Estate View, you can select “create new” then “database cluster”. From here, you can create a cluster from a template or a custom build. Templates are ideal for large enterprises that want to empower developers to quickly spin up databases to react to business needs but without breaking any IT requirements or guidelines. This is also great for users who aren’t Postgres experts – they can get started quickly with just a few clicks of a button.
Custom builds allow you to spin up a database with your own configurations like Postgres version, CPU, and memory.
Observability: Monitoring, Alerts, and Health Status
Let’s dive deeper into some observability features. When you click into a cluster, you’ll see alerts straight away, letting you know if anything requires your attention. Alerts for CPU utilization, memory usage, disk iOPS, and more can be easily customized by toggling them on or off, changing the thresholds as seen below, and choosing notification settings. You can expect additional alerts and customizations to be added in 2025. In the meantime, you can also configure your own alerts in the backend.
By clicking into the “Health Status” tab, you can learn more about the overall health of your cluster, which isn’t always visible when looking at node-level metrics. This tab provides visibility into important metrics like 1) RAFT status for EDB Postgres Distributed clusters, meaning that the cluster can make RAFT elections and make decisions to ensure that your data is replicated safely and consistently, 2) component status, ensuring that enough nodes and other cluster components, such as proxies, are up to make this cluster viable for production workloads, and 3) Replication status displayed as matrix of replication lag.
The “Monitoring” tab shows metrics at the cluster level, which can be broken down to the node level for more detailed information. You can also zoom in or out of specific time periods for a more holistic view of your workloads.
As a DBA looking to make the database more efficient, you can go beyond this assumption using built-in Query Diagnostics within the Hybrid Control Plane. Read on for more details about this feature.
Query Diagnostics
Query Diagnostics gives visibility into what the database was doing by reporting on historical query performance. These built-in query diagnostics enable you to identify problems and bottlenecks up to 5x faster and provide insight to help you accelerate application performance up to 8x.
Simply select a period of time that is of interest – this could be last week, or even just an hour when unusual behaviour was reported – and Query Diagnostics will show you the top queries during that time. You can order the list by total runtime, total time on CPU, total time spent waiting, or various other measures. For each query, you can then drill in and see how that query spent its time. For example, a query that is reading a lot of data from disk might spend a lot of time waiting on I/O. A query that is competing for resources with other queries would spend a lot of time waiting for locks.
Query Diagnostics is “always on” by default meaning this information is always available when you need it, even if the incident you are investigating is already over.
We plan to make Query Diagnostics even more powerful over future releases. We’ll soon be adding query latency statistics and histograms and allow users to view execution plans, buffer usage, and other deep query data. We’ll also be adding AI-powered recommendations based on this data, so you don’t have to be a Postgres expert to run a tightly optimized cost-efficient database.
Logs
With cloud solutions, it can be surprisingly difficult to get to your Postgres logs – you often have to stream them to a bucket and read them out of the bucket or send them to a separate log analysis solution, which can be time consuming and costly. With the Hybrid Control Plane, your logs are streamed straight into the UI.
This provides quick and seamless insight into event logs, so you can ensure the health and performance of your database infrastructure without added services or costs. With filters for errors and warnings, you can proactively identify threats and bugs to maintain the integrity of your system.
Data Migrations
Last but certainly not least, the Hybrid Control Plane enables seamless migrations from Oracle and Postgres source databases through the EDB Data Migration Service (DMS) and Data Sync. This enables organizations with strict security compliance and data privacy needs to utilize EDB’s migration capabilities in their own environments.
When Oracle databases are registered with the Hybrid Control Plane, a quick migration assessment is automatically performed, which includes a calculation of migration complexity and level of effort (LOE) for migrating the databases. Migration complexity is determined based on an analysis of the Oracle features that are in use in the database. Migration LOE provides an estimate of how many days it typically takes to perform the database portion of the migration based on the number and type of incompatible Oracle features found in the database.
Thanks for reading!
To learn more about the Hybrid Control Plane, check out a recording of our webinar demonstrating its features here. You can also learn more about enabling your own hybrid DBaaS experience here. Stay tuned as we drop new and exciting announcements about the Hybrid Control Plane in 2025!
If you’re interested in getting started with the Hybrid Control Plane, fill out this interest form to request to join the tech preview.