Give critical work
a stable foundation.
Physical, virtual and cloud platforms designed for the workloads and people who depend on them.
See the approachSwipe or use arrow keys to explore the diagram.
Read the diagram explanation
Set workload, availability and recovery requirements.
Work through compute, storage, networking and placement.
Power, cooling, identity and recovery can affect several layers together.
Platform placement and capacity need to account for what can fail together, how maintenance will work and how the team will restore the service.
Start with the workload.
The service sets the requirements. Capacity, failure boundaries and recovery shape the platform underneath it.
Place workloads deliberately
Match performance, availability and support needs to the right physical, virtual or cloud environment.
Plan around maintenance and failure
Define which services can be interrupted, what can fail together and how the team restores them.
Carry the design into delivery
Work through racks, power, cooling, cabling and integration so the operating environment supports the design.
Part of the same system.
Connect the design decisions across these areas.
What needs to work better?
Start with the environment, the constraint and the change you need to make.
