Oracle’s New Database Can Self-Protect, Self-Heal And Self-Update

Oracle’s New Database Can Self-Protect, Self-Heal And Self-Update
Image: Oracle via Facebook

During the Oracle OpenWorld keynote address, Oracle’s Chairman and CTO Larry Ellison unveiled the latest version of the company’s flagship database. Ellison says the new Oracle Autonomous Database Cloud is the world’s first 100% self-driving autonomous database. It can defend itself against cyber-attacks and automatically upgrade, patch and tune itself while running.

The idea of a database that can automatically manage its own performance is certainly enticing. Oracle says their new database is able to reduce planned and unplanned downtime to less than half an hour a year while cutting costs in half compared to Amazon.

Autonomous systems are the next step in the evolution of the cloud as automation and orchestration become mainstream. The idea that a system can self manage means people can let the system carry out routine tasks so they can focus on higher value activities.

This is, of course, Oracle’s big play. As the shift from on-prem to the cloud continues, Ellison knows his company could be reduced to little more than a bit player as AWS and Microsoft Azure continue to increase their hold on the cloud market. If the Oracle Autonomous Database Cloud can deliver on all the promises he made during the opening keynote then Oracle could offer a significant point of difference from AWS and Azure, giving companies entrenched in Oracle’s database platform a path to the cloud that offers them an easily quantifiable benefit.


  • Anyone who has had experience with using Oracle products at an enterprise scale knows to take anything they say with a pinch of salt.

    They will promise everything under the sun to get the sale, then once you start having implementation and scaling problems, will charge you handsomely to have their consultants come in and explain that the current version has issues at the scale you’re using it, and that you’ll need to pay to upgrade to the newer version (which surprise, surprise, has the same issues).

    One particular project went for 5 years, 3 versions of software and millions upon millions of dollars. In the end, they had to scale back capabilities to have the software run with acceptable performance.

Show more comments

Log in to comment on this story!