-
A majority of organizations are looking to invest in DevOps automation over the next year, a new report from Dynatrace found
-
But many lack a plan for implementation, in part because they are overwhelmed by technology options
-
Dynatrace's Chief Product Officer said orgs should focus instead on tackling human bottlenecks and making sure the right data is available for automation
Speed or security? Which is more important? And which of the myriad tools available do you use to implement each? These are some of the questions enterprises are up against as they work to implement DevOps practices and automation in an increasingly software-driven world, Dynatrace Chief Technology Strategist and Head of Open Source Alois Reitbauer told Silverlinings.
A fresh report just released by the company found that over the next 12 months, organizations are looking to invest in DevOps automation to support security and compliance management (55%), infrastructure provisioning and management (52%) and performance optimization (51%). But only 38% have a clear vision of how to make these plans a reality.
According to Reitbauer, cloud adoption is one of the key drivers for DevOps implementations. However, he noted how we build and ship products (including everything from banking apps to smart lightbulbs) is also changing, as are user expectations around feature improvements. DevOps, of course, refers to the end-to-end processes associated with developing and shipping software in a repeatable way and retiring old systems.
“Where do you invest first to take a limited amount of talent that you have available at a time, to make it more secure or to make it faster?…You’re always living in this situation where you have to be in many places at the same time,” Reitbauer said of the enterprise dilemma.
He added organizations can also be overwhelmed by both the technology options available as well as the rapid pace of change. Some of the tools that have become commonplace for cloud apps didn’t even exist three, five or 10 years ago, he noted. That includes tools like Kubernetes and OpenTelemetry. The result is a sort of “paralysis” that affects organizations.
Rather than focusing on the technology element, Reitbauer said Dynatrace always encourages organizations to look for steps in the development process that require manual human intervention and aim to automate those first. This will not only make the process faster, but it will also free up employees to work on more important tasks, like feature development.
“People eventually always become bottlenecks. That’s one of the reasons you want automation,” he said.
Digging into data
Automation, though, relies on the availability of key business data. So, Reitbauer said another point of focus should be ensuring that the right data is accessible and can be fed into artificial intelligence (AI) and automation systems.
“About 80% of automation really relies on observability data,” he said.
Thus, enterprises looking to utilize AI especially should consider “Do I even have the data available to build automation? …Where do I get this information from? Is it reliable? Do I have it available in real time?”
Speaking of AI, Reitbauer cautioned that enterprises should be aware that it is really more of an assistant rather than a replacement for engineering talent.
“It’s a helpful tool if you know what the outcome would be” when coding or completing another development task, he concluded. “It won’t make you a DevOps expert.”
9/29/2023 4:00 PM ET: This story has been updated to correct Reitbauer's title.
Want to discuss AI workloads, automation and data center physical infrastructure challenges with us? Meet us in Sonoma, Calif., from Dec. 6-7 for our Cloud Executive Summit. You won't be sorry.