Microsoft introduced two services for relational and unstructured data to its Azure SQL Database service. Ready for public preview in June, Azure SQL Data Warehouse stores relational data in the cloud while Azure SQL Data Lake does so for unstructured data.
The data warehousing market is several decades old but is currently enjoying a cloud renaissance. Microsoft’s warehouse service is its answer to Amazon Web Services’ RedShift, released in 2013. There are also startups such as data warehousing startup Snowflake Computing, headed by a former Microsoft executive Bob Muglia.
In addition to the usual competition in terms of price, Microsoft competes with several features not a part of Redshift. Microsoft’s offering allows for independently adjusting the compute and storage while AWS Redshift is fixed. Microsoft also claims better elasticity which “grows or shrinks in seconds” rather than hours to days. It can also be employed in a hybrid scenario combining on-premises and, as the name suggests, has SQL support.
The company also made it easier to operate and control several databases through a feature called “Elastic Databases”. Users can pool resources and control price and performance across several different databases under an overarching budget.
Azure SQL Data Lake can handle streaming data and allows for individual petabytes-sized files to be stored.
Microsoft continues to expand its cloud database offerings, recently updating Azure DocumentDB, a managed, scalable document database service.
Longtime data warehouse vendors include Oracle, IBM and Teradata. Those providers are evolving their own offerings and often through partnerships as well, such as Teradata hooking into Hadoop providers Hortonworks and Cloudera.