The rise of cloud-native architectures should not be taken lightly. Their impact on everything from adoption of automation to the deployment of app services is not trivial. The codependent nature of cloud-native applications on cloud-native infrastructure shows itself in our findings regarding preferences for application service form factors on-premises. In 2018, just 9% wanted to deploy app services in containers. In 2019? That grew to 14%. And today? That number is an impressive 19%. Just over the percentage of cloud-native apps in an average app portfolio. That preference shows clearly in the deployment rates of related app services like service mesh, ingress control, service discovery, and API gateway.
And it shows in the roles tasked with deploying and operating those app services. We found that as cloud-native apps consume a greater portion of the corporate app portfolio, DevOps practitioners gain greater responsibility for deploying and operating the app services that go along with them.
And as they do, they bring with them the tools they rely on to automate the delivery pipeline. Python (30%), Ansible (25%), and GitHub Enterprise (13%) are the top three tools in use for automation and orchestration of pipelines today.
What may be surprising is that deployment rates of these cloud-native app services is not merely taking place in the public cloud. On-premises and public cloud deployment rates are nearly the same for all cloud-native app services. Proof that cloud-native is not just an app deployed in the public cloud and that the architectural impact will resound in corporate data centers for years to come.
The adoption of automation, too, goes hand in hand with modern architectures and the digital transformation journey. This year we saw greater consistency of automation across all four key components of the deployment pipeline. What's fascinating is that out of the five generational app architectures, both monoliths/mainframes and cloud-native tend to have statistically higher correlation with automation of the deployment pipeline. Traditional architecture and mobile apps? Mostly manual pipelines.
While we expected to see automation associated with modern architectures—it is innate to their operation, after all—we were surprised to also see automation pop up alongside monoliths/mainframes. Our conclusion? Some apps are so important to the business that their operation needs to be modernized to assist in uptime and upgrades on a regular basis.
That importance of and, in many cases, dependence on apps as the basis for business should lead to discomfort in the face of self-reported security skills deficits. The number-one skill deficit reported by respondents is app security. More disconcerting is that 56% of security roles pointed to app security as their biggest skills deficit. This is certainly a challenge that organizations will need to address in the coming years. The criticality of apps is only likely to increase, and security of those apps may be the key—or barrier—to success.