Software and IT Operations professionals recognize the pace at which they’re required to build, release, and manage new software is accelerating, driven by business and departmental needs to digitize their services. While the pace of software is accelerating, so too is the complexity caused by the adoption of cloud, microservices, and container technologies that support modern application architectures.
However, while development is accelerating and the application technology environments are increasingly complex, the need to maintain quality and deliver against user expectations is greater than ever.
This is why the industry is moving towards DevOps, according to Bernd Greifeneder, Founder and CTO of Dynatrace, who spoke during a recent webinar. DevOps methodologies bring together development and operations teams to create agile processes and environments that accelerate software development.
Dynatrace anticipates that the continued growth of microservices and containers will underpin how applications are delivered. This will, in turn, drive a paradigm shift to radically different tooling and solutions that DevOps teams require to be successful.
In just the last two years, there has been an explosion in the complexity of applications and supporting infrastructure. Davis, Dynatrace’s AI engine, for example, runs 32 trillion causations per minute, and that number is going to continue to increase exponentially into the future.
Consequently, the sheer volume of data analysis and interpretation that goes into deployments is growing exponentially too and is already beyond what can be reasonably expected to be analyzed and comprehended through the traditional manual operations process.
According to Greifeneder, we’re effectively hitting the wall of what is possible with traditional operations.
Even with DevOps, where the environments are software-defined and ready in an instant for developers to utilize – infrastructure as code – there is a complex and dynamic relationship between the entire technology stack including the application code, processes, services, hosts, and datacenters. This makes identifying problems extremely difficult, resource-intensive and time-consuming, and knowing precise root-cause virtually impossible.
One solution to this problem, and the one that was the subject of Greifeneder’s recent webinar discussion, is implementing a fully autonomous cloud solution that can self-heal and auto-remediate issues as they arise. Crucially, a human doesn’t need to be in the loop and doesn’t need to manually execute any of the traditional operational tasks to deploy or operate software that’s been built with autonomous cloud capability.
This evolutionary next step from DevOps has given rise to the approach of “NoOps,” a concept that Greifeneder likened to moving from an old-fashioned auto assembly line, in which workmen are standing shoulder to shoulder, each adding their particular part to the chassis, to its modern equivalent, in which automated machines are critical parts of the manufacturing process.
And, like in this analogy, adopting autonomous cloud in software productions and deployments boosts productivity and speed and frees up a workforce to focus on higher-order tasks, like research and development.
One may think that increasing speed and leaving applications to heal by their own devices might lead to a big decrease in quality – more bugs, and more problems that our customers need to deal with. However, as Greifeneder points out, that’s not necessarily the case. As his own company has implemented NoOps, they found that even though they were releasing code ten times more often than they were before they embraced NoOps, they had massive decreases in the number of bugs being reported to them by customers.
As software and application development continue to accelerate and increase in complexity, it’s no longer feasible to rely on manual processes in development and deployment. Developers need NoOps to remain relevant in the market today, Greifeneder argued, and it will be even more vital as we go forward into the future.
Images courtesy of Dynatrace
