Platform as a Product

Chapter 4. Fleet Thinking: Managing Platforms at Scale

Most platform engineering starts with a single use case. A developer needs a database, or a team wants to deploy an application. The platform team builds a workflow or template to support that need. At first, this works well. One template. One path. One environment. But what happens when your organization grows? What happens when you have hundreds of services, dozens of teams, multiple environments, and ever-changing compliance requirements?

This is where fleet thinking becomes essential.

Just as automotive and aerospace companies manage fleets of vehicles or aircraft, with different versions, maintenance states, and operating conditions, platform teams must learn to manage fleets of services, environments, and infrastructure across an organization. This mindset shift is critical for sustainable, scalable internal platforms.

Fleet Management 101

Fleet management is about managing a collection of similar but not identical things, each with its own lifecycle, dependencies, and constraints.

In platform engineering, your “fleet” includes:

• Applications across multiple environments

• Databases, caches, and queues provisioned by the platform

• CI/CD pipelines and workflows in use across teams

• Infrastructure components like clusters, virtual private clouds (VPCs), or cloud accounts

• Platform capabilities and plug-ins that evolve over time

Each of these elements has a lifecycle. It needs to be created, monitored, upgraded, ...