2018 – Autonomy vs Automation

At OpenWorld 2017 last year (it still seems strange saying last year), Oracle announced “The world’s first Autonomous Database”. The marketing literature states:

 “Oracle Autonomous Database Cloud eliminates complexity, human error, and manual management, helping to ensure higher reliability, security, and more operational efficiency at the lowest cost.”

When I first heard about the autonomous database, I didn’t quite get it. I’m no database expert but I thought that we had database management well in hand. I know DBAs with decades of experience who can manage databases with their eyes shut, usually using lots of scripts and automation.

So, what’s different about the new Oracle Autonomous Database? The penny dropped for me when I realised the difference between automation and autonomy. A common misunderstanding is that Oracle has just automated the database. That is not the case.

Automation refers to a set of sequential steps that have are executed in order, usually using a script. Think of an unattended installer. You give it the settings and it executes a number of predefined steps to install your piece of software with the settings you define. Another example might be a DBA who has written a script to automate the patching of a server. The script will run through a series of steps, such as, connect to the server, upload the patch, execute the patch, verify the patch, then, restart the server. Both of these are examples of automation, not autonomy.

When Oracle talks about the Autonomous Database, they aren’t saying that they have just written a number of scripts to automate several steps, they are talking about autonomy, i.e. self-management.

What this means is that, as the administrator, you will define the parameters within which the database must operate and the database will take care of that for you. For example, you will define the service level you need, or the information retention policy you must enforce. Then, the database will do the rest, under the covers to meet that requirement. No more setting up RAC or DataGuard to configure HA and DR.

From a security perspective, the Autonomous Database also reduces the risks associated with manually managed databases. Yes, we have some very clever and experienced DBAs with mature scripts, but, in today’s world of increasing cyberattacks and more data breaches than ever before, against larger and more sensitive data, we need to remove as much of the manual processes associated with security as possible. There will, of course, always be a need for some manual intervention, but the security posture in any organisation, and the response to any threat needs to be more rapid than waiting for an overworked DBA or SOC Analyst to get around to dealing with it.

For example, the Autonomous Database will patch itself regularly with the latest patches and always enable encryption, so you don’t inadvertently leave data stored in the clear.

At the moment, industry is losing the cat and mouse game against the cyber criminals. Looking beyond databases, I can see lots of other places where autonomy, underpinned by capabilities such as machine learning, will play a crucial role in the cyber war in the near future.  There is a long way to go, but it’s an exciting time at Oracle, seeing the emergence of technologies such as the Autonomous Database, as well as our newly designed Identity SOC, really looking at how they address this changing threat landscape using the latest and greatest innovations.

2018 is going to be an exciting year.

Leave a Reply

Your email address will not be published. Required fields are marked *