Even though DevOps has only been around as a codified discipline for just over a decade, it has been so widely adopted it is now the default way most IT shops work with their developers.
DevOps is the blending of software development and IT operations into a cohesive unit that shares the responsibility for running and maintaining an organization’s software portfolio.
In the past, software developers built applications and IT managed them. Because developers and IT usually had conflicting goals, responsibilities, and marching orders, running and maintaining the software they created was a significant, time-consuming challenge for IT operations staff. With the arrival of DevOps, however, the 2 groups now work together to ensure production systems run smoothly.
Even so, the reasons DevOps has been adopted so quickly are mainly business focused. Software updates that used to take weeks or months to move into production can now be done in hours. This improves business agility and rapid business model transformation to meet ever-changing customer demands and requirements. And, by keeping downtime to a minimum, DevOps has been shown to increase business resiliency, as well.
What the future holds
According to analysts, the business-outcomes focus of DevOps will continue in the foreseeable future as it evolves from discipline centered on IT operations to one focused on business outcomes. To achieve this product-focused orientation, DevOps teams will increasingly be defined by purpose rather than function. These include:
- Business and end-user aligned features teams
- Coaching and internal consulting done by enablement teams
- Platform teams to support feature team developers
- Specialty teams (aka, infrastructure or platforms teams) focused on mainframes and other specialty hardware
How these teams are led will also change. Many organizations are experimenting today with less-hierarchical, collaborative leadership approaches that draw from all the different specialty areas that make up a particular team.
Practice vs. Process
Today, DevOps is still an internally process-focused discipline. Its KPIs center on team outputs, activities, and deliverables. Going forward, KPIs will focus more on measurable customer and stakeholder outcomes, aka, business results. DevOps practices will align value streams that are focused on speed and business agility.
In order for DevOps to scale, automation of mundane repeatable processes will continue to grow. This includes greater automation around governance, compliance, security, and standardized IT operations.
Tasks will be broken down into those that automation is well suited to execute and those that require a higher degree of understating that only a human can realistically apply. Automation also will free up highly-skilled and hard-to-find developer and IT talent to focus on more value-added activities such as exception management, solution design, and creative problem solving.
Cyber security takes center stage
Cyber security is a board room-level discussion today. No matter how much security is in place, no organization is safe from cyber crime. As Agile development practices continue to move more code into production faster than ever, ensuring it is secure at go-live has never been more important.
DevOps will continue its evolution into DevSecOps, where security best-practices are baked into the software release cycle. This will be achieved by inviting security teams to early application planning sessions and ensuring basic security tools such as multi-factor authentication are in place. This early-stage collaboration will increase overall security and organizational agility.
Code-based change management will take over from swivel-chair integration, where someone is manually entering change tickets into a service desk solution such as ServiceNow. In the future, DevOps pipelines will continuously integrate with service desk management platforms. This will cut down errors and accelerate release cycles by ensuring all change requests are complete before go-live.
Platforms and automation
As DevOps continues along this evolutionary path, automation will play an ever-larger role. DevOps teams will migrate away from point solutions by adopting integrated SDLC pipelines that run on dedicated platforms. These platforms may include MLOps capabilities, unified CI/CD/CDRA, low-code/no-code platforms, and will extend to the network edge.
Automation requires detailed, real-time data to effectively improve risk management. This is why machine learning algorithms will play a larger role in analyzing risk in future development pipelines.
DevOps will become more people orientated
At its core, DevOps is about people and improving how they work together. Organizations who successfully deploy DevOps at scale will lean into and adopt this mindset, not just DevOps best practices around CI/CD, automation, and software support and maintenance.
The cross-cultural nature of established teams will lend itself well to team-based problem solving and support. Different support tiers will start to collapse into a more horizontal structure where the concept of tier 2 support disappears in favor of more difficult problems escalating directly to the product teams and tier 1 support stretching to accommodate lower-level cases.