How to Build a Better DevOps ModelBy Samuel Greengard | Posted 2017-09-04 Email Print
WEBINAR: On-demand webcast
Next-Generation Applications Require the Power and Performance of Next-Generation Workstations REGISTER >
Establishing a DevOps strategy and putting it into motion can prove extremely challenging. Success depends on factors such as technology, processes and culture.
The complexities of digital business aren't lost on anyone, but achieving the level of coordination and orchestration required to navigate today's challenges can be overwhelming.
"The ability to introduce new business services, new capabilities and new functions—and push them out to customers and others quickly and effectively—is dependent on the IT and business sides of the enterprise being in total sync," explains David Newberry, principal for application services and platforms at Capgemini Consulting.
The answer for a growing number of organizations? DevOps, which combines elements of "development" and "operations" though a framework that places a premium on collaboration and communication. When DevOps is used effectively, it allows software developers and other IT professionals to work with line-of-business (LoB) groups to automate and improve processes that revolve around infrastructure and software.
"It's a very powerful tool for removing bottlenecks," says Aater Suleman, co-founder and CEO of Flux7, a business consulting firm that specializes in DevOps.
Yet, establishing a DevOps strategy and putting it into motion can prove extraordinarily challenging. More than a few business and IT leaders have learned the hard way that success depends on several key factors, including technology, processes and culture.
In addition, as emerging technologies mature and a digital footprint reaches deeper into organizations, the scope and nature of DevOps continues to evolve. So there's a need to expand and scale initiatives in order to boost the value further.
Reducing Complexity, Improving Agility
Most business and IT leaders understand that a successful DevOps initiative reduces complexity and improves agility. It helps an enterprise code faster and better; ensures that software more closely aligns with the needs of business units and customers; and helps match and scale IT to the organization's specific and often changing requirements.
"An organization can gain a level of automation and efficiency that is transformative," says Capgemini's Newberry.
Capgemini found that about 88 percent of CIOs surveyed are already using DevOps to some degree. "The big issue is the level of adoption, as there's still somewhat of an aversion to the concept," Newberry points out.
In some cases, business and IT leaders aren't entirely clear on what DevOps is, what it does and how it works. But there are also problems with how organizations roll out initiatives. Too often, there's a failure to identify and engage key stakeholders and understand the link between the organization's business agility needs and how, what and where DevOps can deliver.
A clear vision about what DevOps will deliver is paramount. This typically revolves around unlocking business value, which includes ways to deploy applications, software and IT services faster, smarter and better. In order to get to best practice results—including significant performance gains and organizational transformation—it's critical to include input from key stakeholders across the organization.
"A point that's often overlooked is that DevOps is heavily about the culture of the organization," Newberry says. "It requires different thinking and different behavior."
A culture of accountability is paramount, Flux7's Suleman says. This means establishing a governance model, along with KPIs, metrics and gateways, as well as fostering an understanding among teams that the sum of systems or code is greater than the individual parts.
"Development teams need to recognize that they are responsible for more than the one piece that they create and toss over to the next group," he stresses. "They need to be accountable in a more holistic way." This means thinking about the end product, focusing on code security, and understanding how the entire codebase interacts across a spectrum of products, services, APIs and other offerings.