• How can I optimally partition my data in Cosmos DB for maximum performance and scale?
• What are the best practices for migrating from relational to NoSQL or using them together?
• How can I minimize request latency for geo-distributed applications using Cosmos DB?
• What are configurable data consistency levels in Cosmos DB and why should I care?

Non-relational databases like MongoDB and Cassandra have achieved mainstream adoption, and a growing minority of software developers appreciate the advantages of NoSQL and polyglot storage models. The rise of public cloud providers like Amazon and Azure brings even more choice in the form of “serverless” NoSQL like DynamoDB and Cosmos DB. But what are the pros and cons of a managed NoSQL service? And how do you minimize cost while also maximizing performance and scale?

In this practical, hands-on talk we’ll dive into real-world advice on how to maximize your use of Azure Cosmos DB. We’ll consider the implications of easy and fast geo-replication, schema-less data storage, configurable data consistency, seamless data partitioning, and more. We’ll discuss best practices for migrating from relational to NoSQL, or using them together; we’ll even talk about use cases where Cosmos DB is not a good fit. Prior knowledge of NoSQL and relational databases is recommended.