Oracle Autonomous Database debut heralds new cloud hopes

Machine learning meets the DBA

Increasingly, securing databases will require machine learning automation, according to Richard Niemiec, chief innovation officer at Viscosity North America, a Dallas-based cloud and database consultancy.

It needs no human intervention. It will patch itself.

That is important as needs arise to monitor audit logs, logins, locations, questionable IP addresses and questionable user access, he said.

"What's new with this announcement is the security advances that use machine learning and automatic patching," he said. "Fixing security vulnerabilities as they happen with automatic patching and using machine learning log mining for anomalies is the first step of autonomous security in the database."

This fits with an overall trend that sees DBAs handling higher-level problems, he indicated.

"DBAs are moving from being 'database administrators' to 'data administrators' where data is both in the cloud and on premises," he said.  As a result, they need to be skilled with many divergent types of data stores. 

Use of the Oracle Autonomous Database on the cloud could free up the time of data professionals, so they can pay more particular attention to the most critical databases, security issues and the new variety of diverse data sources, he suggested.

Equifax makes patching top of mind

A pivotal event in corporate cloud data was last year's Equifax cyber breach. It laid bare records of over 145 million customers, and led to early retirement for several in its executive suite. The company had failed to fully patch known vulnerabilities in an open source web application framework, exposing data.

Ellison and others on hand at the cloud data warehouse rollout pointed to Equifax as an example of the increasing need for automated patching to dodge such cyber bullets.

"We've all heard about the Equifax breach. We don't want that," said Benjamin Arnulf, director of business intelligence and analytics at Hertz in Estero, Fla. Arnulf's team has had early access to the data warehousing service on the cloud. It has been available to selected Oracle early users such as Hertz since the end of last year.

Arnulf favored moving patching work away from DBAs, so they could focus on helping more general analytical improvements at the auto rental giant, and he saw benefits in this regard for the new Oracle offering.

"If you can remove the pain points, you can remove hours of work from the DBA," Arnulf said. "They can reinvest all these hours into innovation."

The Oracle Autonomous Database is intimately connected with the company's cloud services, reminds Craig Mullins, who is president and principal consultant of Mullins Consulting Inc. in Sugar Land, Texas.

Cloud implementations generally mean that you use the cloud provider's database, and that they provide the most up-to-date patching, he said.

And, handing off patching chores is an ongoing trend, he said.

"If I am a DBA, I don't want to be involved with all these patches. That is grunt work. When that gets taken away, most DBAs are going to be thrilled," Mullins said.