Azure Data Fundamentals

Part III. Nonrelational Data on Azure

Modern applications rarely live on relational storage alone. This part broadens your perspective to the specialized storage modalities that optimize cost, scalability, and latency for unstructured content, schema-less entities, asynchronous processing, and globally distributed operational data. You’ll learn how Azure Storage services provide foundational primitives and how Azure Cosmos DB delivers low-latency, globally replicated data with tunable consistency.

• Chapter 7, “Azure Storage Solutions”, classifies storage services—object (Blob), table, queue, and file—explaining access characteristics, durability guarantees, and when each pattern fits (or doesn’t fit). It emphasizes selecting the lightest viable abstraction rather than forcing database semantics into object stores.

• Chapter 8, “Azure Cosmos DB”, examines what changes when data distribution, partitioning, consistency trade-offs, and elasticity throughout become primary design drivers. You’ll see how partition key design and request unit (RU) budgeting influence both performance and spend.

After completing this part of the book, you should be able to map workload access patterns to their storage modality, articulate the consequences of a poor Cosmos DB partition strategy, and justify consistency level selection based on business tolerance for staleness versus latency.

Common mistakes surfaced here include adopting Cosmos DB without having a genuine need for global distribution, ...