4 DevOps trends to watch in 2020
Whether you’re just beginning with DevOps or you’re pretty far down the road, this way of working is about constant improvement. Check out four big trends
Just the phrase “DevOps 2020” looks and sounds somewhat funny: It’s a milestone reminder that the term – and the culture and practices the term represents – has shed its “new and shiny” status in IT.
That doesn’t make DevOps passé; it means it has become pervasive, and it will mature in various ways in the year ahead.
“In 2020, we will see organizations [continue to] embrace DevOps practices at all levels,” says Tom Iverson, managing principal, cloud and DevOps at RBA.
DevOps will mature in some key ways during the year ahead.
By “levels,” Iverson’s not referring to organizational hierarchies – executive, managerial, and so forth. Rather, he’s noting that every team’s starting point with DevOps was a bit different, including when they got started. Indeed, some organizations are still getting going. More advanced DevOps teams might be innovating and optimizing their continuous delivery processes in the year ahead, for example, while organizations farther back in the adoption curve might still be experimenting in earlier stages of their software pipeline, or grappling with the all-important cultural components of DevOps success.
[ Culture change is the hardest part of digital transformation. Get the digital transformation eBook: Teaching an elephant to dance. ]
Regardless of your particular DevOps phase, Iverson and others anticipate ongoing trends worth watching. While the term might not be as buzzy (thankfully, perhaps) anymore, DevOps is ultimately about ongoing change and improvement. Here are four areas to watch.
1. DevOps maturity will fuel continued interest in cloud-centric tools
As DevOps improves with age, so to speak, it will enable and drive adoption of cloud-centric (or cloud-native, if you prefer the term) technologies. One example in particular stands out:
Expect to see the continued and increased adoption of Kubernetes.
“In 2020 we expect to see the continued and increased adoption of Kubernetes for organizations to deploy services,” says Josh Komoroske, senior DevOps engineer at StackRox. “The combination of its ease of use, depth of features, and integration with major cloud platforms will make it an increasingly appealing choice for developing and deploying software – even more so if your organization already deploys [containerized] software.”
[ Related read: Kubernetes: 3 ways to get started ]
This is a two-way street, too: Just as DevOps culture and practices drive interest in various tools, these technologies will help enable ongoing cultural shifts and process improvements in the year ahead.
“Cloud technologies will continue to grow and a ‘DevOps first’ mindset will be baked into new developments,” Iverson says. “Heavy-handed governance models will turn to more of a custodial approach due to the amount of automation available in the cloud space. For example, IT will look to provide self-service options to support rapid provisioning of temporary resources for proof-of-concept work.”
Moreover, Iverson sees that kind of self-service expansion being brought to teams where they work: “Commodity tools such as office suites, low-code solutions, and ChatOps tools will provide the interface for self-service,” he says.
2. IT leaders will revisit DevOps success metrics
Ongoing measurement has been a key part of DevOps culture and process for a while. That gave way to some common metrics, such as mean time to deploy (MTTD), for quantifying your progress. And this is sensible: If your initial case for DevOps was something like “we’ll be able to deploy faster and more frequently than before,” then bolstering that case requiring showing both claims were actually true.
Expect such “traditional” DevOps metrics to be increasingly joined by the somewhat more subjective measurements that sit under the outsized umbrella of “business value.”
Look for more metrics emphasizing business value and customer satisfaction, as opposed to release cadence.
“We’ve learned that faster is not always better when it comes to software release cycles,” says Justin Donato, VP of IT at Nintex. “As a result, we’re seeing organizations pivoting away from release cadences and emphasizing business value and customer satisfaction.”
You could think of this as an intersection point between DevOps and digital transformation. The point is not to simply operate faster for the sake of it, but to figure out how that increased speed benefits your business. And while “business value” might sound fuzzy, the abstraction is really a placeholder for those specific benefits in your organization. Donato points to accelerated sales cycles or time to revenue as examples of specific ways to measure the business value of DevOps.
“[Companies will increasingly] prioritize those kinds of things ahead of time-to-deploy,” Donato says. “Businesses today aren’t looking so much for fast releases as for quality releases. Competitors can spring up overnight, so it’s key to deliver capabilities that keep your customers satisfied and loyal; you don’t want them looking elsewhere.”