Deployments
Model runtime environments, infrastructure nodes, placements, and connections
Deployments connect the logical architecture to the infrastructure where it runs. Open Model → Deployments.
Deployment Model
Capture:
- Environment and deployment purpose
- Infrastructure nodes, regions, and managed services
- Architecture elements placed on each node
- Runtime connections between nodes
- Capacity, cost, sizing, and operational notes
- Resource requests declared by deployed elements
Use explicit read and edit modes during review. A deployment canvas is versioned architecture evidence, not a live cloud inventory.
Auto-Connect
Auto-connect compares deployed architecture elements with canonical Relationships and proposes corresponding infrastructure connections. Review proposals before accepting them; logical interaction does not always imply a direct network path.
Read Capacity Evidence
Infrastructure summaries show Not specified when CPU, memory, or storage has not been recorded. Totals include only recorded values. Missing capacity is unknown, rather than evidence that a node has no resources.
Record CPU, memory, and storage for every node before interpreting capacity-planning guidance. The capacity-planning section remains unavailable while this evidence is incomplete; figures describe the modeled infrastructure rather than live utilization.
Review Questions
- Is every runtime-critical element placed somewhere?
- Are environment and region boundaries explicit?
- Do resource requirements fit the chosen nodes?
- Are queues, stores, gateways, and managed services represented at the useful level?
- Does a logical Relationship cross unexpected infrastructure boundaries?
Deployments can be embedded in Architecture Documents, selected for View documentation, published, and compared between versions.