The Scaled Agile Framework (SAFe) is a great tool for establishing agile and Lean best practices across an enterprise. It provides an overarching architecture for aligning development, quality assurance and other functions to produce a faster workflow and to boost performance across the board.
There is an important missing link, though. To date the SAFe framework hasn’t incorporated site reliability engineering – a function of growing importance in today’s application-driven economy.
Site reliability specialists focus on the operational infrastructure so vital to keeping sites and services running. They work to improve availability, latency, performance, efficiency, change management, capacity planning and a host of other factors that influence service delivery and the user experience.
So why isn’t this important function included in SAFe? The framework focuses more on system development and delivery than on the operational end of the spectrum where site reliability resides. But times are changing. Progressive companies are shifting site reliability engineering to the left to support development and delivery.
Why Shift Left?
Site reliability specialists have valuable software engineering skills. They bring an “as-code” approach to configuration, testing and other tasks – reducing the effort involved in monitoring and improving operational metrics. They use these software skills to manage reliability, but they haven’t been positioned to use what they know to build reliability in from the beginning.
A shift left breaks down this functional barrier. It positions reliability specialists to work in concert with development and release teams to ensure architecture and configuration quality across the entire software lifecycle. It also makes the most of those underlying software engineering skills.
The payoff from a shift left can be significant. Your organization can better manage configuration changes, service levels and error budgets. You can establish a continuous cycle of feedback and governance – from initial design and development through to the launch and operation of new services. And you can better support and advance your agile and Lean objectives.
Mapping Reliability Engineering to SAFe
Though SAFe doesn’t address the role of site reliability engineering, you can easily map and integrate the function on your own to support a shift left. Focus on the following three points of synergy to integrate reliability engineering at critical junctures in your DevOps lifecycle.
1. At the application level
Integrate reliability engineers with your SAFe agile development team – the group tasked with defining, building, testing and delivering apps in sprint. These new team members can set up and track application-level service objectives, error budgets and DevOps pipelines. And they can help you ensure each new component and each new application will support reliability – not erode it.
2. At the system level
As you move further along the SAFe continuum, integrate reliability engineers with your SAFe system team to support release train activities for multiple components and applications. These specialists will be positioned to focus on launch coordination, governance of your system architecture, error budget tracking, systemwide service level objectives – and more.
3. At the enterprise level
Finally, integrate reliability engineers into the SAFe enterprise solution delivery function to oversee your enterprise system architecture and service delivery. Task them with establishing and running Centers of Enablement for reliability engineering, developing enterprise-level best practices and governance controls, improving business agility and promoting the reliability of complex architectures.
Selecting the right tools
This significant broadening of the site reliability engineering function can clearly deliver important new benefits. For optimal outcomes, though, you will also need to broaden your supporting toolset.
At the application level, team members will need to track the resolution of issues they uncover. At the system level, they will need to evaluate readiness and performance against specific service-level objectives. At the enterprise level, they will need a big-picture view of reliability that spans all your systems and services.
Fortunately, a new generation of solutions is emerging to support site reliability engineers as they make the shift left. These new platforms are tailor-built for the task at hand and powered by artificial intelligence, machine learning and intelligent automation.
One example: The Broadcom BizOps platform includes a Release Health and Risk Dashboard that delivers proactive insights into each new release before go-live. Reliability engineers can quickly pinpoint problems and track remediation. Once a service is in production, an Operations Dashboard helps engineers track availability, response times, error rates, and more. Importantly, the two dashboards interoperate so your reliability team can correlate release health data with production data and evaluate the quality of their release health predictions.
It’s time to get started
If you want to bake reliability into your systems and services from the start, consider broadening the role of your site reliability engineering team. Use the SAFe framework as your guide and align skilled talent at the application, systems and enterprise levels. Select the right tools to support your newly distributed team – arming them with analytics that can turn data into actionable insights. You will be poised to make significant strides in your continuous improvement journey.