What is NoOps and is It Real?

Is NoOps the logical conclusion of DevOps and what does this mean for software development and IT operations?

There are those who claim the evolution of DevOps culture will lead to widespread adoption of a software development strategy where developers are not reliant on IT operations. Interest in this “NoOps” strategy has been gathering a great deal of momentum lately, and you’d be forgiven for thinking its proponents were getting carried away.

While some organizations have started implementing NoOps, broad adoption isn’t going to happen overnight. Still, it’s good to keep an open mind about new approaches that claim revolutionary potential. And it’s certainly worth looking at how NoOps might benefit your software factory.

The Death of IT Operations?

Back in May, the WannaCry cyberattack hit organizations across the world. If there is one thing guaranteed to bring IT ops to the public’s attention it’s news about major attacks enabled by the failure of organizations to update or patch software known to have security vulnerabilities. Doubtless, some IT pros were feeling a little schadenfreude when the news broke.

Others may have been feeling a little defensive, seeing IT ops portrayed in such a negative light. This is to be expected in the context of DevOps culture, which appears to reduce the need for operations staff to such an extent that some people have started talking about the end of IT ops as we know it—NoOps being the buzzword they’ve come up with to describe this situation.

The term NoOps it is not as new as you might think. As far back as 2013, a GigaOm infographic declared it to be “the year of NoOps for programmers”. From this perspective, it seems like NoOps was always fated to be the logical outcome of DevOps. But are things so clear-cut? Does DevOps inevitably lead to the minimization of ops or is NoOps a buzzword and nothing more?

DevOps into NoOps

DevOps is now a well-established method for the continuous deployment of applications and patches, which is a requirement of staying competitive in the face of digital disruption. The ops part of DevOps deals with release management and configuration management but is seldom concerned with the running of datacenters or the deployment of virtual machines.

These tasks are mostly left to a cloud service provider. If there is still an in-house IT team charged with such tasks, it may well struggle to provide a serverless environment. This is the only way, developers can use DevOps methodologies to deploy applications in house with the same ease they would find in a cloud-based service if they have little traditional ops knowledge.

Making NoOps a Reality

So, is the next step application developers not having to deal with operations professionals at all? It is possible to automate the infrastructure creation and management tasks required to build and deploy application releases and—while this itself is a challenge—it would allow developers to manage and maintain live code in addition to their development code.

Questions remain about what technology would be required to make this scenario a reality (monitoring, feedback, root cause analysis and remediation spring to mind). And how about the continuing role of in-house ops? Is this scenario only achievable with cloud infrastructure or could it be realized on premises? Can on-premises deliver NoOps anyway?

While the concept of NoOps has been around a while, its widespread adoption seems a little way off—and doesn’t feel inevitable. Perhaps the real question is: If it happens, how effectively will it allow enterprises to maximize the productivity of their software factories? It will be interesting to see if and how the concept becomes reality over the coming months and years.

But it will be really interesting to see how early adopters benefit from this new reality and whether it will enable their much-needed digital transformation.

Peter Matthews
By Peter Matthews | August 03, 2017