kran77 - Fotolia
Enterprises will embrace new technical advances in artificial intelligence and container orchestration in 2017 to push the continuous-deployment envelope. These mark uncharted territory for historically conservative enterprise IT departments, but forward-thinking technologists see the shift in attitudes as inevitable and imperative.
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
"IT professionals should resolve to disrupt themselves before they are disrupted by an external entity," said Greg Arnette, CTO of Sonian Inc., an email archival and analytics cloud service provider in Waltham, Mass. "The rapid pace of change in IT -- fueled by cloud advancements, as well as new AI and machine learning frameworks -- means all technologists need to step up, increase innovation and not rest on past success."
"It's not just, 'How do we set up your CI or CD system?' But, 'How do we then use all the information and knowledge that we have at hand, given the millions of builds going through the system, to help make your process better without necessarily requiring your effort?'" said Jim Rose, CEO at CircleCI, a continuous integration and continuous delivery software-as-a-service provider in San Francisco.
Rose said he views DevOps monitoring as analogous to road-mapping systems.
"Five to 10 years ago, we [in IT] had the equivalent of a paper atlas," Rose said. "What we have now is the equivalent of a junky GPS system." The goal in 2017 is to get to the Google Maps or Waze equivalent for IT monitoring, "where before you even end up in the traffic jam, it redirects you."
Users are eager to pursue this as part of a DevOps plan.
Greg ArnetteCTO of Sonian
"We'll be working to increase and extend our cloud security, automated remediation and heuristics-based inspection of performance, security and quality of service," said Alex Witherspoon, vice president of platform engineering at FlightStats Inc., a global data service company in the aviation space now owned by FlightGlobal and based in Portland, Ore.
A nuanced humanlike intelligence can go beyond existing automation to offer recommendations on much more complex observations than threshold-based alerting. This allows IT engineers to spend less time on simple issues, Witherspoon said.
IT security will also need a proactive DevSecOps plan in 2017, Arnette said.
"Sophisticated, AI-driven, malicious bots are poised to disrupt the status quo, and countermeasures need to be implemented," he warned.
Serverless, continuous deployment to emerge in 2017
Another consensus among IT pros is containers and container automation will be ready for prime time in 2017 to form the foundation for more new DevOps plans.
Containers will drive further automation into the code delivery process, so teams can more frequently push code directly to production environments.
"As we see more teams realize that they've got the right tooling in place to actually deploy continuously, especially with the adoption of containers ... [big companies] are actually going to release direct to production," said Robert Stroud, analyst with Forrester Research.
Some organizations worry about the separation of duties for compliance purposes in such a scenario, but Stroud said those fears will dissipate next year.
"You can make decisions based on successful testing and have automation push the change into production, so you are actually separating the developer from production," he said.
Containers already underpin serverless cloud offerings and will begin to enable on-premises serverless setups for large enterprises in 2017, some IT experts said.
"Over time, containers will be seen as a way to support serverless architectures and event-driven programming," said a senior vice president for a major financial services company, who spoke on condition of anonymity. "AWS [Amazon Web Services] uses Lambda heavily in a virtuous cycle that streamlines its internal offerings, and on-premises deployments need a similar construct."
DevOps plan turns up the heat on developers
IT career requirements will obviously also need to shift, and some IT pros are focused on helping junior developers face the onslaught of change.
"In 2017, there will be a shift in requirements for the successful coder, with an emphasis on adaptability, a strong personal brand and flexibility in working environments and development languages," said Chris Riley, DevOps analyst with Fixate IO, a content strategy consulting firm in Livermore, Calif., and a TechTarget contributor.
Riley's company has worked on a project called Sweetcode.io, which allows up-and-coming developers to learn new skills and display online portfolios.
Follow the example of DevOps adherents
Automation demands systems programmers
Build automation -- for your boxes