Uneven demand
A busy period can expose a database limit before a compute limit. Understand the part of the system constraining the experience.
SaaS & applications
Keep the application
moving forward.
Growing software changes the demands on its infrastructure. Connect capacity, deployment and operating support to the way your product is used.
The challenges
The pressure points change as you grow.
More releases
The environment needs a repeatable path from development to production, with a way back when a change fails.
An expanding team
Platform responsibilities become easier to miss as more people are involved. Make ownership and access explicit.
Our approach
Turn requirements
into decisions.
Start with the application
Map the services, dependencies, data flows and release process before choosing a larger server.
Match capacity to behaviour
Consider utilisation, storage performance, traffic and isolation together. Set out the signal that should trigger the next capacity change.
Carry the design into operations
Agree monitoring, maintenance, recovery and the boundary between the application team and infrastructure provider.
Common questions
What should we include in an initial brief?
The application stack, present environment, user locations, typical and peak demand, and the problem prompting the discussion.
Who manages the application?
The operating model is agreed for the particular environment. Define infrastructure, operating system, deployment and application responsibilities separately.
Can a migration be part of the discussion?
Yes. Include dependencies, data volume, acceptable interruption and rollback requirements so the relevant team can assess the scope.
The next conversation
Let’s look at
your environment.
Share what runs today and what you would like to improve. The next step is a scoped discussion with the relevant team.
Discuss this requirementFurther readingRead the related engineering perspective