The 2019 technology forecasts are still a few months away, but some predictions are already a safe bet — especially when it comes to DevOps. New estimates from IDC suggest that the DevOps software market will grow from its 2017 results of $2.9 billion to $6.6 billion in 2022. What will drive that growth? At Pythian, we’re seeing five major trends.
- DevOps will finally become mainstream
To DevOps fans, this first prediction might come as a bit of a surprise. After all, DevOps has been a topic of enthusiastic conversation for the better part of a decade. But in the real world, departments and development tools still remain stubbornly walled off from each other, either for budget reasons or because of entrenched corporate cultures. The goal of eliminating silos remains unfulfilled right now, but our second trend offers reason for hope. - Executives are getting on board
Business leaders are burdened with competing priorities, and saying yes to one urgent thing means saying no to something that might be equally important. For many years, DevOps found itself on the losing side of that zero-sum game. But new research from Gartner points to a meaningful shift in CEOs’ thinking. Growth remains the number one concern of most leaders, but there is an increased openness to technology as a means of achieving that growth. Among CEOs’ top five business priorities, digital investment has risen from 2.1% in the 2012 survey to 13.4% in 2018. Moreover, 61% of respondents plan to increase their IT spending in 2018. - Security is becoming a priority
DevOps is rightly praised for moving functional testing earlier in the development process. “Shifting left” has allowed bugs to be spotted and fixed sooner, which in turn has enabled faster and more frequent production. But early iterations of DevOps delayed considerations around security. This was clearly a miss, as can be seen from the endless reports of data breaches around the world. As DevOps continues its growth, it will likely move closer to DevSecOps in its approaches. In other words, mainstream DevOps will start treating security as code, making it the entire team’s responsibility from the earliest stages of the development lifecycle. - Kubernetes is king
With the growth of DevOps comes a growth in the size and complexity of container production clusters. An orchestration tool is needed to deploy and manage container clusters, and Kubernetes has emerged as the orchestration tool of choice. As a demonstration of Kubernetes’ dominance, Docker is now distributing and supporting the tool while still investing in its own orchestration system, Swarm. If a new tool comes along to challenge Kubernetes, we predict it won’t be anytime soon. - The pipeline is maturing
DevOps is typically described as a pipeline, a linear workflow moving from build to test to release. As DevOps has grown in popularity, though, that metaphorical pipeline has become more of a cycle, one that incorporates added phases of testing, monitoring and continuous improvement. In this way, DevOps has been a victim of its own success, because its growth has outpaced the availability of tools to support the pipeline from start to finish. Fortunately, new tools (most notably from Microsoft) show a vendor commitment to supporting the continued evolution and success of DevOps.
DevOps is about more than technology. It represents a way of thinking that can help enterprise thrive at a time when speed to market means the difference between success and failure. Find out how Pythian can help your organization benefit from the potential of DevOps.
2 Comments. Leave new
Great article Greg!
Shifting left and taking on the tech debit up front is key, and sometimes a struggle culturally within companies. You hit the nail on the head security is definitely a part of this, and far left and developer responsibility as it relates to their code going through the pipeline. Also, great new tools out there to value stream this whole process, identify waste and improve the process. (CI)
No silos! Transparency, empowerment of the engineers to make the tech decisions, not top down, might sound odd but even celebrate and make visible the failures. Only the best have failed, and failed big before. Always cross train, always have time for training if you can’t spend a day a week just training and improving there is larger issue.
Serverless, like the rest has been around a little bit but is cutting operational cost overhead dramatically, and with more and more options emerging covering more use cases. What once was a 100k bill now in some cases can be around 100bucks.
Continuous improvement is The Toyota Way!
So glad i found this article. I am doing a project on this and i found it very useful. Thanks for sharing.