How to deploy · Scenario 5 of 5
Three-layer namespace
MaestroHub at the line, the site and the group.
- Servers1 per line, 1 per plant, 3 for the group
- Survives a server failureThe group hub, yes.
- Keeps running without the WANYes, at every layer
- MaestroHub tiersTier 1 + 2 + 3
How it works
Select a step to light up its part of the drawing.
What you install
Recommended hardware per machine. The tiers are explained on How to deploy.
| How many | What | Where | Each machine |
|---|---|---|---|
| 1 per line | MaestroHub, Tier 1 edge | L3, at the line | 8 vCPU · 16 GB · 200 GB SSD |
| 1 per plant | MaestroHub, Tier 2 site hub | L3.5, the plant DMZ | 16 vCPU · 24 GB · 500 GB SSD |
| 3 | MaestroHub, Tier 3 group hub | L4, your data centre | 16 vCPU · 24 GB · 500 GB SSD each |
| 1 | Fleet Manager, optional | L5 | A small server |
Trade-offs
Good for
- One namespace and KPIs across every plant
- Every line and every site keeps working alone
- Each layer is sized for its job
Costs
- Three layers to run and upgrade
- The most machines in total
- Layers need an agreed topic contract
When something fails
- The WAN between a plant and the group drops
- The plant keeps working; group KPIs catch up when the backlog is delivered.
- A site hub fails
- Lines keep running and buffer toward it.
Best for
One namespace across plants. Lines and sites that must work alone. Groups with an IT platform team. Three layers to run and upgrade; agree who owns which part of the namespace first.
Every machine's full stack, ports and limits are in the documentation for this scenario.
Planning your deployment?
Tell us about your sites, systems and deployment goals. Our team can help you understand the available editions and coordinate the next steps for implementation planning.