With the introduction of Microsoft Fabric comes a new, more efficient query engine based on the Polaris distributed SQL engine. We will dive deep into the internals of Fabric data warehouse's logging, storage, compute, and query layers. You will see the impact statistics and table definition have on an execution plan, look at the way nodes are dynamically allocated and dropped for maximum query performance, and see how we deliver top performance over files stored in the opensource Delta format. At the end of this session, you will walk away with best practices for designing your data warehouse tables properly from the beginning, know how to keep it performing great, and have a better understanding of what the SQL engine is doing in the background when running in high concurrency scenarios. Best of all, the information you will learn will also apply to querying Fabric's lakehouse using the SQL analytics endpoint.