Oracle Global Leaders Summit 2024 – An Overview of New and Upcoming Developments to Autonomous Database

Last month, I attended the Oracle Global Leaders event in Tirana, Albania. Amongst all the excellent presentations, Çetin Özbütün’s talk on Autonomous Database really piqued my interest. Çetin is Executive Vice President of Oracle DW & Autonomous Technologies and his presentation covered what is new and upcoming in Oracle Autonomous Database. It was pretty exciting and in this blog, I’m going to give an overview of some of these innovations, some of which are already live, and some that are due to be released in the next few months.

Multi-Cloud Deployments

To paraphrase Larry Ellison in his 2023 keynote at Oracle CloudWorld ‘the future is Multi-Cloud’ and Autonomous Database deployment options are indeed coming thick and fast:

• Oracle Cloud Infrastructure
• Dedicated Cloud@Customer
• Oracle database services in Microsoft Azure datacenters. This involves Oracle Database Cloud services running *inside* Azure data centres, purchased with Azure Cloud Commitment credits and fully integrated Azure-native user experience
• Oracle databases in Google Cloud – as above but in GCP –is being rolled out worldwide over the next few months.

The general message is that the user will get the ‘same data ecosystem everywhere’ – i.e. wherever it’s deployed, Autonomous Database will look identical. The benefits to customers are obvious – no more siloed environments, but the chance to unify complex estates through Autonomous Database, regardless of which Cloud platform you are using.

Low-Cost Development Environments

More ‘new’ and ‘coming soon’ news for Low-Code too. Users can now develop in Autonomous Database locally and for free! It presents an identical database experience, the built-in tools you’ll be familiar with like SQL Worksheet, APEX and Data Studio, and offers easy integration into CI/CD pipelines. Download here.

And, coming soon will be the low-cost (approx. $30 pcm) Developer Edition of Autonomous Database with the same features as above but managed by OCI Control Plane which includes patches and updates.  
 

Data Sharing and Collaboration

Autonomous users can now share data across databases using standard open-source protocol (Delta-Sharing, https://delta.io/sharing/) across Autonomous data warehouses and transactional systems. This includes versioned data shares and live data shares.

As well as Data Sharing there will also be a Data Marketplace enabling you to load data into your Autonomous Database, this opens up some really interesting possibilities, it’s not yet clear how extensible will this be, i.e. general marketplace, or tenancy/organisation specific, but it’s another way to reduce silos in organisations. BUT, it will also raise questions around data duplication - watch this space, it will be interesting.

Advanced Machine Learning

In the next few months, Oracle will release Autonomous Database support for GPUs, which will mean users will have access to high-performance data processing and faster performance. This will mean the end of separate provisioning or configuration of GPU resources – and therefore a much more straightforward user experience. Governance will also be integrated, using the same secure environment as existing data. Access will be available via Python, and GPU-enabled Python packages can use third-party packages in conda environment, like Tensorflow, PyTorch, Keras, etc. The pricing structure will also be straightforward and will dramatically increase compute power at a small multiple of CPU compute cost.

Exascale

Oracle are soon to release a new hyper-elastic multi-tenant software architecture. It will involve a loosely coupled scale-out design that removes the need for cluster-based storage management and runs on all existing ROCE-based Exadata hardware. This will make Exadata’s extreme performance available to smaller workloads and businesses than has previously been available. Users will be able to easily and quickly clone databases for development and test while retaining all extreme performance features. What’s more, tenants only pay for the CPU and storage they need and space-efficient snapshots reduce costs for test and dev – making this a compelling option from a number of viewpoints.

Oracle Database 23ai

Naturally, this has been the big news in the Oracle database world recently and is now available for Autonomous as well. It includes over 300 new features with an emphasis on developers and AI. It’s more scalable, available and has improved security and analytics. Automated upgrades (scheduled for whenever the user wants) is in the short-term roadmap and in the longer term, Oracle will be automating upgrading databases to 23ai. The diagram below nicely illustrates what’s now available:

In short, a lot of new information from Çetin and plenty for Autonomous Database users to wrap their heads around. If you’d like to go through any of it in more detail, please do get in touch or find out more about  Oracle Autonomous Database here.