The Empire strikes back with the most advanced
Oracle version to date. The focus of this new version is surprise, surprise, AI , which is so embedded into the new product that Oracle found prudent to change its name from 23c to 23ai to make the point. Let's find out what's on offer.

While AI might be the number one priority, there's also improvements on the areas of developer productivity and the safeguarding of critical systems that store mission critical data.

The most important new feature, AI capabilities wise, is the addition of the AI Vector Search which makes it possible for LLMs to query private business data using a natural language to provide accurate and relevant results. Acting like a Vector store it lets you have your embeddings stored alongside your JSON or relational data under the same roof so that customers can seamlessly integrate AI Vector Search into existing business applications.

This integration with LLMs such as Cohere and Llama that lets you ask questions using natural language, aims to make the wealth of information stored in your enterprise database approachable to anyone, as easy as chatting with ChatGPT. Therefore even those who do not understand SQL can easily and quickly query business data and generate reports.

As far as the aspect of developer productivity goes, the new features include:

JSON Relational Unification
This allows you to leverage the power of both the relational and JSON development approaches all within a single app. By creating simple JSON Duality views on top of relational tables you can build applications that use documents to query and update the underlying data sets.

The notion here is that an application can treat data as relational data and run SQL against it or the same application can treat the exact same data as JSON documents and read and write it using native JSON APIs, as such also benefiting from JSON's convenience.

Operational Property Graph with SQL
Under this option you are not just able to use Oracle as both a relational and document oriented database but use it as a Graph database too; you can treat the same data as if it was a graph, hence you can enjoy all the advantages arising from treating your data as nodes.

JSON Schemas
You can also use the industry-standard JSON Schemas
to ensure only valid data is inserted into a JSON column.

JavaScript stored procedures
Until v21c you could write stored procedures in PL/SQL, Java and more languages through the use of External Procedures. In 21c this capability was extended to cover JavaScript too, thanks to GraalVM. Version 23c comes with enhanced Javascript support by introducing JavaScript modules and inline procedures. JavaScript modules are stored as schema objects and can be created in-line with the module header, based on Character Large Objects (CLOBs), or BFILEs stored in the file system. Inline JavaScript Functions and Procedures can be used in cases where you just need a JavaScript function instead of a module.

It's free
It can be downloaded straight from oracle. com without a user account or license click-through requirements, in the form of a Container Image, VirtualBox VM and Linux RPM installation file.

Now on to Mission Critical Workloads.

Oracle is mostly run on large enterprises, and with reason. It is one of the most durable and resilient databases out there. The new version is set to make the architecture that underpins the mechanisms which guard the mission critical data more simple and scalable.

One such aspect is sharding. While database sharding has been baked into Oracle since version 12c, 23ai extends it by introducing Raft replication. With Raft replication - a consensus-based protocol that facilitates the automatic configuration of replication across all shards, the globally distributed database in the advent of a shard failure or reconfiguration automatically rebalances the data quicker than before and with zero data loss.

Oracle True Cache
True Cache is a diskless in-memory Oracle Instance running in front of the primary database, acting as a consistent, application transparent, high-performance middle-tier cache. It is helpful because it improves application response times while reducing the load on database servers

And finally, there's a new in-database SQL Firewall which protect databases against unauthorized SQL, including SQL injection attacks.

Note that those features we touched on were the most noteworthy of 23ai but there's been more on of "secondary" importance, such as :

• Boolean datatype
• Priority Transactions
• Real-Time SQL Plan Management
• XML and JSON Search Index Enhancements

of course the usual bug fixes and performance improvements.

In a sense 23ai extends Oracle beyond a mere relational database system by rendering capable of handling documents, graphs, vectors and Json, as well as acting like Redis, and as a firewall.

While it is certainly true that the main selling point for Oracle is the reliability of the product which adheres to industry based standards, that's not the only one. After sales support is the other. Oracle support on call 24/7, priority bug fixing, insurance based responsibility with accountability.

Those are reasons that is preferred by the Enterprise. 23ai does all that, but also steps up its game in offering a much better experience for both developers and, up the line, managerial personnel too.

And while I am a big fan of its open source counterparts, such as Postgres, I've got to say that 23ai really does make a difference.

More Information

Announcing Oracle Database 23ai : General Availability

Related Articles

Python Driver For Oracle Database Goes Asynchronous

To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Twitter, Facebook or Linkedin.

Comments

Make a Comment or View Existing Comments Using Disqus

or email your comment to: comments@i-programmer.info