Infrastructure · Infrastructure resource
Agent Hosting Platform
Infrastructure resourceInfrastructureInfrastructureVariation point (abstract)arc:AgentHostingPlatform
An abstract compute substrate on which agent services are packaged, executed and scaled, trading operational control and warm capacity against management overhead and idle cost.
Responsibility. Provides the execution environment in which deployed agent services run.
Also known as: Deployment pattern, Deployment architecture
Variants
| Variant | When to choose |
|---|---|
| Container Orchestrator | Choose when consistent latency matters more than cost (24/7 warm instances, no cold starts), Kubernetes already runs other services to amortise overhead, the team has strong Kubernetes expertise, continuous background processing is needed, or debugging requires persistent logs and long-term metric retention. |
| Serverless Function Runtime | Choose when traffic shows high variance or long idle periods, deployment velocity is critical, the team lacks infrastructure expertise, cost must scale with actual usage, or the system processes discrete events rather than continuous streams. |
Relationships
hosts structural
- Agent Controller abstract Ch4.2
Design guidance
- MUST select the hosting pattern from actual traffic patterns, team capabilities, latency requirements and total cost of ownership including operational overhead, not compute pricing alone.
- SHOULD NOT assume serverless is always cheaper or microservices always more complex; both claims fail under specific traffic and team contexts.
Quantitative guidance
As stated by the sources; verify before use.
- Recommendation worked example (50,000 sessions/day, 70% in a 4-hour peak, ~5 s/session, P95 SLO 8 s): microservices $222.80/month compute (~$420/month incl. managed Kubernetes) vs serverless $330.66/month; serverless chosen (Ch4.2).
Classification
- Patterns
- MicroservicesServerless
- Quality attributes
- Performance efficiency (ISO/IEC 25010)Reliability (ISO/IEC 25010 | NIST AI RMF: valid and reliable)Cost efficiencyMaintainability (ISO/IEC 25010)
- Risks mitigated
- Collapse when usage exceeds initial capacity estimatesDefaulting to a familiar but costly deployment approach
Sources
- 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.