Azure SQL Data Warehouse offers an Enterprise Data Warehouse in the cloud. It uses Massively Parallel Processing (MPP) combined with Azure Storage, to provide high performance and scalability. To create valuable insights into the data stored inside the Data Warehouse, Azure uses Hadoop/Spark and machine learning.
Data is stored in relational tables with columnar storage. When using columnar storage, the data is written and read in columns, instead of the rows used in traditional row-oriented databases. So, when you query the data, columnar storage skips all of the irrelevant data by immediately jumping to the appropriate column. This will make your queries run a lot faster. For instance, when you want to look up the average age of all of your customers, columnar storage will jump to the age column immediately, instead of looking at each row for the age...
