Module reference · Marketplace & workloads
x/deployment — Deployment
Tenant deployment specifications and groups — what tenants ask the marketplace to run.
What it does
Inside x/deployment
The deployment module stores what a tenant wants to run: a deployment specification enumerating one or more groups of services, each with resource requirements (CPU, memory, storage), placement constraints, and pricing bounds. When a tenant creates a deployment, the module emits the orders that the market module opens for bidding.
Deployments are versioned, updatable state: tenants can update or close a deployment, and group-level granularity means a single deployment can fan out to multiple orders placed with different providers in different regions.
Why it exists
The design rationale
Tenants think in terms of workloads, not individual orders. The deployment module translates a declarative workload description into marketplace primitives, the same way a Kubernetes Deployment translates a desired state into pods — but with the exchange, not a scheduler, deciding who runs it.
Interactions
How it connects to the rest of the chain
No module stands alone — these are the protocol surfaces this module depends on or serves.
Key concepts
Terms that matter here
- Deployment
- A tenant's declarative description of the services they want the marketplace to run.
- Group
- A subdivision of a deployment that becomes an independently-biddable order with its own placement rules.
- SDL
- The stack definition language in which workloads are described — services, resources, and placement.
Keep exploring