Data Modeling and Partitioning in Azure Cosmos DB by Mark Brown

Session Title: Data Modeling and Partitioning in Azure Cosmos DB
Speaker: Mark Brown

Abstract: If you are new to Azure Cosmos DB and have only ever used relational databases, this session is for you! Azure Cosmos DB is unique in that a database that it can provide the same exact performance whether it is Megabytes in size or Petabytes in size. But the only way to get that kind of performance is understand how to model and partition data. Even if you’re not trying to design a database with ultimate scale, you may have other questions How should I structure my data? When should I co-locate data in a single container? Should I de-normalize or normalize properties? What’s the best partition key for my model?

In this demo-filled session, we discuss the strategies and thought process one should adopt for modeling and partitioning data effectively in Azure Cosmos DB. Using a real-world example, we explore Azure Cosmos DB key concepts—request units (RU), partitioning, and data modeling—and how their understanding guides the path to a data model that yields best performance and scalability. If you’re familiar with relational databases, and want to dive into the non-relational world, this is the session for you.