SAN FRANCISCO—A few weeks before Oracle’s annual OpenWorld conference kicked off, Chairman and CTO Larry Ellison promised news about a revolutionary “Autonomous Database.” He did not disappoint.
Ellison spent two hours on the first night of the conference here on Oct 3, touting Oracle 18c, a new version of the company’s database that “manages itself, patches itself, backs itself up, tunes itself, with no human administration, and no human error,” Ellison said. “Amazon can't do this.”
Database autonomy powered by machine learning, is just one way Oracle is seeking to differentiate itself from Amazon Web Services, which Ellison has been attacking for the past three years. There are few pieces of its product line that Oracle does not compare with AWS, especially around cloud infrastructure and its ubiquitous database.
Except for an occasional mention of Microsoft’s Azure Cloud or Workday’s SaaS-based business applications, AWS is the first competitor mentioned when Oracle compares its offerings. Why ignore the others, including SAP's cloud Enterprise Resource Planning apps or Google Cloud Platform?
It’s simple. The cloud is the future, and Amazon is the still the leading cloud infrastructure provider by a big margin. With cloud migrations picking up steam, Oracle’s mission is to retain as many of its 420,000 customers as possible and keep them from switching to AWS or other cloud or SaaS providers.
AWS, for its part, has been busy for the past three years offering new and improved databases in its cloud, along with a range of migration and schema translation tools to move Oracle users into AWS.
The right cloud for customers
The other overriding factor is there’s a finite number of customers. So, behind the scenes, away from the glare of Ellison’s keynotes, other top executives talked about how Oracle is going to carry out its customer-retention mission.
The company is now in the second year of its “Generation 2” data center build-out. Oracle last week announced the opening of its third new data center region built around the Gen 2 architecture, in Frankfurt, Germany. London will be the next in Q1 2018, giving Oracle a total of four regions with 12 “Availability Domains” (discrete data centers, three per region).
This is tiny compared with AWS’s current 16 regions and 44 Availability Zones. Oracle may never catch up in the number of data centers, but it may not need to. The vendor wants to build “the right cloud for our customers,” said Don Johnson, Oracle’s Cloud chief. Oracle asserts that Johnson is one of “hundreds” of former AWS engineers now working on Oracle’s Cloud.
The “right” cloud means whatever works best for the customer, Johnson said. Oracle has options that include a bare metal service for “lift and shift” workloads and virtual machine instances for “born in the cloud” applications, Johnson said.
Oracle also is pushing in two other areas: It wants to keep its lead on database technology—hence the Autonomous Database news—and embed AI and machine learning into as many applications as possible.
Oracle announced new integration of intelligent features across all its SaaS and E-Business Suite of applications. These “Applied Intelligence” services can add advice options for workers running enterprise asset management software, or give sales people the next-best-option on a client sales process.
By now, artificial intelligence and machine learning features are table stakes for any application vendor out there. These functions are not much different from what Salesforce.com offers with its Einstein technology.
But vendors do not seem too worried about being copy cats when it comes to AI. In fact, the companies are likely using the same technology to build and train its models. “These are our own algorithms, written using commonly known techniques,” said Steve Miranda, Executive Vice President of Oracle Applications Product Development.
If Oracle seems to understand one thing better than the other vendors, it’s that it's not making huge promises about AI saving the planet or curing cancer. Oracle understands that the opportunities offered by machine learning are large, but that this is only the beginning. CEO Mark Hurd even said people should not get too excited about AI.
“It’s just a feature,” he said. It’s an important feature, but it can only augment other applications and therefore Oracle is focusing on “integrating AI and getting it as close to the applications as possible.”
Scot Petersen is a technology analyst at Ziff Brothers Investments, a private investment firm. He has an extensive background in the technology field. Prior to joining Ziff Brothers, Scot was the editorial director, Business Applications & Architecture, at TechTarget. Before that, he was the director, Editorial Operations, at Ziff Davis Enterprise. While at Ziff Davis Media, he was a writer and editor at eWEEK. No investment advice is offered in his blog. All duties are disclaimed. Scot works for a private investment firm, which may at any time invest in companies whose products are discussed in this blog, and no disclosure of securities transactions will be made.