In the cloud era, senior executives are investing in key capabilities like DevOps with the expectation this will accelerate innovation and bring new digital products to market faster. To deliver on these goals, enterprise organizations seek to deploy high-quality code faster and more frequently, but they face a organizational, process and technology complexity that can hinder this objective, slowing throughput and incurring high error rates. Often when DevOps is discussed, the focus is concentrated on the technology layer, particularly the latest tools like Kubernetes, and also on the productivity of each individual developer.
However, for many organizations, the main challenges will be with team practices and organizational models.
Writing for DevOps.com, Mike Vizard highlighted the reality of DevOps uptake; that most adoption issues are related to organizational challenges, not technology. The top two barriers to adoption are slow processes and speed of adaptation (29%) followed by budget and funding (21%). Only 18% identified technology limitations as an issue.
As Mike writes, it is the broader organizational transformation that is more challenging. ‘Addressing bottlenecks’ might be a function of confronting the situation of a department manager who zealously guards their turf and acts politically to resist any changes required.
Implementing DevOps for one team is a relatively straightforward exercise, but for enterprise organizations, they face a much larger magnitude of complexity, as they typically have multiple teams spanning multiple geographies and suppliers with hundreds of developers all contributing to the same software development process.
Therefore, a DevOps methodology that also addresses the organizational level is required, one that maps processes end-to-end and identifies the improvements needed to speed production across this entire life cycle.
Software Factories
Transformation leaders are turning to a ‘digital factory‘ approach to streamline this complexity into a production line model that yields consistently high throughput and quality using Lean principles derived from the manufacturing world.
McKinsey explored the dynamics and successes of the digital factory model, a new approach to organizing digital teams. This model involves a self-organizing structure rather than rigid departmental hierarchies, and they described the KPI improvements the digital factory enables: “We see reductions in management overhead of 50% for technology teams in the digital factory, 70% in the number of business analysts needed to write technology requirements, and, as test automation becomes the norm, a drop of 90% in the number of testers. Finally, we see top engineering talent performing at eight times the level of their peers, as measured with metrics such as code commits.”
In short, McKinsey described a development team that improves quality and finds efficiencies as a direct by-product of one that leverages the latest tools and team methods to produce better digital products faster. The concept of a ‘software factory’ is a very powerful metaphor because much of the insights are drawn literally from the world of manufacturing, where over many decades they have perfected the science of optimizing the efficiency of a production line.
To close the loop and map these development improvements to business value, management can leverage organizational performance practices, notably Six Sigma and the theory of constraints, to define their ‘DevOps algorithms,’ a systematic formula for quantifying and improving the rate of software development throughput.
DevOps Flow
DevOps Flow is a methodology for implementing high-performance, cloud-centric software development based on science developed by pioneers like Toyota to optimize factory lines and apply this to software engineering.
Organizations need the ability to measure the entire system end-to-end to understand how value flows and where it is constrained and, most importantly, to correlate those metrics with desired business outcomes. This approach allows for continuous optimization in the pursuit of delivering greater and greater value to the organization faster.
It can be applied to speed the deployment of new software releases, notably identifying and removing constraints, reducing batch sizes and eliminating waste. These are all steps taken within manufacturing to increase production throughput and are improvements that software teams can adopt to achieve equivalent benefits.