Automation is on the rise in IT shops, especially in the infrastructure and operations realm. Some automation discussions make it sound like a boogeyman – automation’s coming to get you – but that misses the point. It’s important to understand – and be able to articulate – the why of automation.
Automation enables teams to unlock the potential of modern technologies, methodologies, and IT’s ongoing evolution from back-end support shop to business driver. It’s not as much a trend as it is a necessity for today’s development and operations work.
“There are several factors that are increasing adoption for automation, [such as] APIs, computing horsepower, cloud technology, and machine learning,” says Jeff Kavanaugh, VP and senior partner in the high-tech and manufacturing practices at Infosys. “However, it is the combination of these several factors that is accelerating overall adoption.”
[ Get wisdom from your peers on leading DevOps teams: Culture, metrics, talent, and more. See our resource DevOps: The IT Leader's Guide. ]
For sure, the confluence of those technology components, from cloud services to containers, plays a big part. But these factors aren’t limited to technology: Putting your people in a position to succeed is another driver. Let’s dig into five factors fueling automation in IT.
1. Cloud, containers, and microservices
Cloud computing and cloud-native technologies lead the list of technical factors driving increasing automation in IT shops, especially along the development and operations continuum.
“The speed of cloud adoption is one area we see pushing automation in enterprise IT,” says Mor Cohen, senior product manager at Turbonomic.
Cloud environments still involve complexity, especially when it comes to optimizing usage and costs. That’s even more true in environments where the infrastructure footprint is likely to grow, sometimes considerably, over time, Cohen says.
That’s where containers and microservices become appealing in many organizations, and automation soon becomes vital.
“You could manually deploy and configure 50 or even 500 servers, but when you hit 5,000 servers that is no longer an option,” says Ned Bellavance, director of cloud solutions at Anexinet. “Automation gives the administrator tools to effectively deploy scalable workloads, without a commensurate increase in staff.”
Indeed, when you consider an automation tool like Ansible or an orchestration platform like Kubernetes in this context, that’s exactly what they’re intended for: They help realize the potential of cloud, containers, and microservices in a scalable manner without constantly throwing new bodies at growing and evolving operational needs.
[ Want Kubernetes advice? See our related article, 5 Kubernetes must-reads: Tips and trends. ]
2. The explosion of APIs
We could have rolled APIs up into “key technical factors” above, but it merits its own star turn. That’s because we’re in the “API-driven everything” age, as Bellavance puts it, and that’s an enormous boon for automation. This wasn’t always the case.
“One of the restrictions on automation was the inability to effectively interact with infrastructure resources in the same way that you would interact with applications. Adding a programmable API to infrastructure enables automation to be a first-class citizen in both the cloud and on-premises,” Bellavance explains. “I now hear [major] hardware vendors talking about API interfaces for all of their hardware and management systems. They recognize that we are in a software-driven world, and their solutions need to support that approach.”
3. The increasing speed of software development
IT must deliver more than ever, faster than ever: The monolithic software development lifecycle of yore can’t keep up.
The underlying technological forces driving automation adoption – cloud, containers, microservices, APIs, etc. – also enable IT to become a strategic catalyst of the business goals instead of a bottleneck. In this sense, automation isn’t just a technical matter, but a business strategy.
“Aspiring to drive market share, be more competitive, and meet customer demands and expectations, the business always has new requests for IT,” says Dan Juengst, principal technology evangelist at OutSystems. “This means that IT must move faster and needs to move from manual, human-driven processes to automation.”
Modern software development cultures and practices such as DevOps and Agile also require reducing technical debt – including kludgey, time-consuming patches – and making way for automation. It’s a key part of shrinking deployment times without sacrificing quality.
“The result of automation is scalability – less effort per person to maintain and grow your IT environment, as Red Hat VP, Global Services John Allessio recently noted. “If adding manpower is the only way to grow your business, then scalability is a pipe dream. Automation reduces your manpower requirements and provides the flexibility required for continued IT evolution.” (See his full article, What DevOps teams really need from a CIO.)
We’re not just talking about increasing speed, but improving quality.
“Teams are realizing that attempting to reach a market quickly is beneficial to the business but are also recognizing that how the speed is applied is more important,” says Chris Ciborowski, CEO of Nebulaworks. “It is this second item of the velocity equation, like reducing error rates or providing an easier path to roll-back on deployments that requires new tools and processes. [These] tools, like containers or continuous deployment platforms, and optimized processes are best leveraged with high degrees of automation.”
4. The human element: Maximizing talent
While there are certainly fundamental technology and business forces underpinning IT automation, it’s also about people.
If your engineers and other talent spend the bulk of their time on painful, manual tasks, are you getting their best work? Probably not, and that’s a killer when you’re constantly challenged to do more with less.
“In the enterprise IT space, one of automation’s biggest use cases is ensuring businesses have the right people performing the right tasks,” says Mark Kirstein, VP, products at BitTitan.
You need your people working on strategic initiatives, not manual maintenance tasks that can be automated away. Moreover, redirecting talent to strategic priorities often engenders a healthier IT culture. It’s high-value work that requires critical thinking and problem-solving; the same can’t always be said of some mundane, albeit necessary, operational tasks.
“Automation – and some strategic benchmarking to go along with it – helps businesses optimize the use of their own talent and match work with experience level,” Kirstein says. “This can be hugely positive for company culture, as employees are now working on tasks that match their experience level, challenge them, and are often more critical to the end business outcome.”
“Furthermore, employees who feel more valued at work and can tackle challenges commensurate with their expertise will be happier, stay longer, and probably work harder,” he says.
5. The different kinds of “costs”
The people factor translates to costs, too. If you have senior engineers (likely earning six-figure salaries) working on routine maintenance tasks, you’re spending too much on routine maintenance, period. And even if you limit such operational needs to junior employees, you’re probably still spending too much.
But Kirstein points out that while “cost” (as in: cost savings) is often cited as a driver of automation projects, too many organizations focus only on cost savings. The smart shops are also looking at opportunity costs, too – as in, what you gain by redirecting your people away from manual operational tasks to more strategic work, how faster time-to-market for new services impacts the bottom line, and much more. (Or, in other terms: What you lose by not doing so.)
“Automation is either saving or making money, and businesses that are focused solely on measuring the cost of implementing automation are doing it incorrectly,” Kirstein says. “The opportunity cost is what smart businesses are capitalizing on and what’s fueling this initiative.”
Automation is key because it has become very difficult to construct a coherent continuous delivery toolchain in that complex and diverse technology arena. Different silos are forming and teams on the same project are drifting further apart, not closer together, using increasingly diverse tools and approaches.