Chapter Ten
Case Studies
Five workload shapes worked end to end — the same service catalog assembled five different ways. Each study names the services, the trade-offs, and the reasons one design beats another for that workload.
Core Terminology
These case studies reuse the services from earlier chapters. The recurring terms below frame the trade-offs.
Reference Architecture
A documented, opinionated design for a class of workload — the starting point you adapt, not a blueprint you copy blindly.
Serverless
A model where you deploy code or containers and pay for use, with no servers to manage — Functions, Container Apps, Cosmos DB serverless.
Event-Driven
An architecture where components react to events rather than calling each other directly, decoupled through Event Grid, Service Bus, or Event Hubs.
Scale Unit
A self-contained, replicable slice of an architecture that you add to grow capacity predictably.
Data Residency
The legal requirement that data stay within a geography — a constraint that shapes Region choice and replication for regulated workloads.
Strangler Fig
A migration pattern that incrementally replaces a legacy system by routing slices of traffic to new services until the old one is gone.
Topics in This Chapter
Topic 66
Serverless SaaS
A multi-tenant SaaS built on Functions, Container Apps, Cosmos DB, and Front Door — scale-to-zero economics with per-tenant isolation.
Topic 67
Fintech Platform
A regulated payments platform — strong consistency, auditability, encryption, and multi-Region resilience with a measured RPO.
Topic 68
E-commerce at Scale
A storefront built for traffic spikes — global edge delivery, caching, autoscaling compute, and a catalog and cart data model that holds up under load.
Topic 69
Analytics Pipeline
Ingestion to insight — Event Hubs, Data Lake Storage, Synapse, and Power BI — and where streaming gives way to batch.
Topic 70
Enterprise Migration
Moving an on-premises estate to Azure with landing zones, Arc, hybrid networking, and a strangler-fig sequence that avoids a big-bang cutover.