IET Event: Software Engineering at Scale - Conway's Law

This event is featured by: Institute of Engineering and Technology (IET)

This month we will be working alongside IET to look at overcoming software engineering inefficiencies at scale within the framework of DevOps by applying our understanding of Conway's Law.

DevOps teams at scale and rapid prototyping at scale:

Conway's Law propositions that "organisations design systems that mirror their communication structures." In the context of software engineering, this means that the architecture of a software system will reflect the way teams and departments within an organisation are structured and how they communicate.

Hierarchical structures and efficiency issues:

As organisations grow, they often adopt hierarchical structures with specialised teams. While this can promote expertise and efficiency at a smaller scale, it can become inefficient when scaled up. In large organisations, teams may be siloed, with limited cross-functional communication, leading to misaligned goals, slower decision-making, and fragmented systems.

In a large software project, this inefficiency manifests in the build and deployment pipeline. With many teams working in isolation, each with their own priorities and processes, integration and addressing cross-code issues becomes difficult. A lack of coordination can lead to delays, bottlenecks, and software quality issues.

Overcoming the scaling issue:

To address these challenges and ensure a smooth build and deployment pipeline at scale, several countermeasures can be implemented:

Cross-functional Teams: Organise teams around specific features or components rather than technical expertise alone. This promotes communication and alignment, ensuring the system's architecture aligns with the team's workflow.

Microservices Architecture: Breaking down large monolithic systems into smaller, independent services can improve scalability. Teams can work on different services without stepping on each other's toes, while still maintaining a cohesive system.

DevOps and Automation: Implementing DevOps practices such as continuous integration (CI) and continuous deployment (CD) automates much of the build and deployment pipeline, reducing friction and accelerating delivery.

Regular Communication and Feedback Loops: Encouraging regular touchpoints between teams, such as daily standups, sprint reviews, and retrospectives, helps identify potential bottlenecks early and improve the system iteratively.

By proactively applying these countermeasures, teams can overcome scaling issues and maintain efficiency, ensuring a smoother process from development to deployment even as the organisation grows.

By being mindful and having awareness of Conway's Law allows you to address those failures in prototyping by utilising small cross-functional teams.

The Room will be The Planet Suite held at the hotel for DevOps Group