Agentic AI Reference Architecture

Infrastructure · Data artifact

On-Demand Consumption Plan

Data artifactInfrastructureInfrastructurearc:OnDemandConsumptionPlan

A serverless hosting plan that scales automatically from zero and bills only per execution, accepting occasional cold starts and shorter maximum timeouts.

Responsibility. Provisions function capacity purely on demand.

Also known as: Consumption Plan

Variant of Serverless Hosting Plan abstract

When to choose. Choose for variable, unpredictable or infrequently triggered agent workloads.

specializesalternative toalternative toServerless Hosting Plan: specializesServerless Hosting PlanPre-warmed Capacity Plan: alternative toPre-warmed Capacity PlanReserved Capacity Plan: alternative toReserved Capacity Plan
Direct neighbourhood (hover for relationship types)

Relationships

alternative to variability

Quantitative guidance

As stated by the sources; verify before use.

Classification

Patterns
Scale-to-zero
Technologies
Azure Functions Consumption PlanAWS Lambda
Quality attributes
Cost efficiency
Risks mitigated
Idle capacity cost

Sources

  1. Ch4.2: T. Nguyen, "Deployment and Scaling," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 4.2. ISBN: 9798244538229.