From concept to business model: It’s time for DevOps to take that next step

Imagine you’re building a house. Different teams have different responsibilities, all entering and exiting the process at varying stages from beginning to end.

But the catch is, none of the teams are communicating with each other. The entire project is completed in siloes; teams are aware of the overall objective but they lack the direct channels between departments. Naturally, the job becomes 10 times harder.

This is a greatly simplified representation of the problem that development and IT operations teams have been tackling for years. Since Patrick Debois first coined the term ‘DevOps’ back in 2009, there have been debates around whether true convergence between these two departments has actually been achieved.

Regardless, huge strides have been made in bringing these two teams together, and this year is forecasted to feature milestone progress in this space. Naturally, the progress will extend far beyond one year, but we’re approaching a time where development and IT teams become one, unified by the governance of a process-based platform that enforces key technical and business requirements across the delivery lifecycle.

Here are four key trends that we can expect to gain momentum as the year progresses.

1. Closing the gap with low code

Low-code – using a development environment for application software through a GUI – has been gaining momentum across industries in recent years, and will be fundamental in the merge of DevOps. Removing the complexity of code makes the entire process accessible to teams, regardless of experience in writing code.

Furthermore, low code allows developers to play a more central role in writing configurations for software, thereby streamlining operations and reducing unnecessary risk as developers have full control over the configurations. Usually, IT operations would need to interpret the application based on the developer’s guidance and implement the configurations accordingly.

Naturally, where there are benefits, there are also challenges. For example, low code comes with the risk of undetected security vulnerabilities in preconfigured code modules, as well as potential performance limitations of software.

Despite these challenges, we expect to see continued uptake of low code solutions in the coming year, which will be paramount to the convergence of DevOps.

2. The rise of Kubernetes

Originally designed by Google, now maintained by the Cloud Native Computing Foundation, Kubernetes (K8s) triumphed in the race for dominance as the top performing container orchestration system in 2017. And it’s been on the rise ever since.

Its popularity amongst IT operators and developers stems from a number of sources. For developers, Kubernetes delivers a consistent, predictable way to run applications, meaning Dev teams need not worry about addressing as many variables and edge cases when writing code.

On the other side of the table, IT operators are given a systematic and reliable way of deploying applications at scale. Again, we see here how the two sides are converging because of this.

3. The next stage of automation

This year, we can expect to see an automation evolution.

Where automation has predominantly been used by DevOps for automating software delivery processes, like Continuous Integration (CI) and Continuous Deployment (CD) of applications, teams are now using the technology to extend other elements of IT management and operations. Businesses will be making heavier use of automated testing to ensure application quality, for instance, and security teams will adopt automation to help remediate risks.

Automation will help to bring even more velocity to both development and operation processes. Traditionally, workflows like software testing or security response could become bottlenecks for DevOps teams. By automating these processes as much as possible, businesses reduce friction and make DevOps even faster and more efficient.

4. The rise of AI and ML

Where automation as a tool has been well established in the world of DevOps, artificial intelligence and machine learning are comparatively under deployed. And those companies that have already invested have found that the AI-powered tools they have are relatively simplistic on a functional level, limited mostly to basic pattern and anomaly detection.

However, all this is changing now that AI-powered DevOps tools are capable of advanced functionality, like automatically rightsizing workload configurations or detecting security risks in complex access control settings.

By taking advantage of advancing AI and ML tools, Development and Operations teams can work together with greater efficiency because these tools address challenges that the two teams both tackle every day. The tools will also help all stakeholders in DevOps speak a common language and work towards common operations, security, and compliance goals – steps that are essential for achieving the complete convergence of Dev and Ops.

The convergence

After making steady progress towards achieving the intention behind ‘DevOps’, we’re nearing the point of complete integration. As both teams from development and IT operations continue to work closely together, using increasingly intelligent technology to plug the gap between the

two, DevOps will naturally evolve from being the concept that emerged over a decade ago, to a functional business model that achieves that which Debois had originally intended