When Oracle introduced its next-generation cloud strategy with intelligent applications at its 2016 OpenWorld conference, the company opened the door for further cloud adoption and possibly artificial intelligence in the cloud.

Intelligent cloud applications will help organizations take the next best action by using their existing data in the cloud to aid decision making. The biggest hurdle to an intelligent cloud is getting workloads into the cloud in the first place, according to experts.

Approximately 6% of the world’s workloads are in the cloud, according to R “Ray” Wang, principal analyst and CEO at Constellation Research, citing an Oracle statistic. Even if 10% or 15% of workloads were in the cloud, customers still need entry points, including different deployment options such as pure cloud or hybrid cloud options, he said.

Additionally, customers still need to consolidate their cloud environments. SaaS, PaaS, and IaaS layers should be streamlined to make the cloud deployments more efficient and ready for a more intelligent cloud experience, according to Wang.

SEE: Oracle’s Hurd: Cloud migration driven by business reality; there will be two SaaS suites (ZDNet)

Proactive SaaS coming down the pike

In the meantime, the intelligent cloud means more proactive SaaS applications, according to Todd Scallan, vice president of products and engineering at Axcient. These applications will anticipate the needs of the user or network and respond accordingly, eliminating or greatly reducing manual intervention.

Disaster recovery as a service is one example of a proactive SaaS application. Proactive applications in a disaster recovery application will include automatically detecting changes to systems, applications, and data within primary compute environments; spinning up secondary instances of part of or the entire environment based on rules and usage patterns; migrating changes from secondary back to primary; and spinning secondary instances back down, Scallan said.

Assuming that there’s cloud adoption and consolidation, organizations will eventually be able to benefit from artificial intelligence in their cloud applications, according to Constellation Research’s Wang. “Some of them will be like the Cortana-type examples, but a lot of them are going to be about predicting the next best actions or preventing and mitigating risk. These intelligent clouds will all have a data component and be built on neural networks,” he said.

SEE: SaaS Research 2017: Adoption rates, business benefits, and preferred providers (Tech Pro Research)

Building the next-generation intelligent cloud

For organizations already leveraging the cloud for their workloads, they’ll need to take into account how to develop intelligent cloud applications, and that means adopting a DevOps mentality, according to Kevin Summers, senior director of business development at Mitel. Incremental changes to applications and services will need to be pushed continually to users, he noted.

Because of this, configuration management will become more critical, as will automation for testing and deployment. A/B testing of functionality and interfaces will also be important to ensure that end users are getting value from the application and don’t realize that it’s actually an intelligent application making the decisions, Summers said.

Oracle’s intelligent cloud strategy is just one step toward cloud-based applications that aid in decision making and automate processes. Once organizations add more workloads to the cloud, the demand for these applications may increase and lead to artificial intelligence in the cloud. It’s really just a matter of time.