A temporal table needs only one primary key. System Versioning must be turned on in order to construct a temporal table. The default naming scheme is applied to the history table if you don't supply a name for it.
Even if you need to absorb data from many sources, you still need to design a dimension model snowflake or/and star. The best practices advise excluding all natural keys from the dimension tables and only including integer surrogate keys or value encoding. To minimize the uniqueness of the values and enable considerably greater compression, you should decrease the cardinality.
The function external table() should be used to refer to an external table after it has been defined. An external table can be queried by any database reader or user.
Zone-redundant storage synchronizes the replication of Azure Storage data across three Azure availability zones in the main region. If a zone is unavailable, the data is still accessible with zone-redundant storage for both write and read operations.
Utilizing archive compression, COLUMNSTORE_ARCHIVE data compression is used to reduce the size of columnstore data.
Unstructured data can be stored and accessed at scale using block blobs with Azure blobs.
> When you want your apps to handle random-access and streaming scenarios, Azure blobs are the best choice.
> When you need to be able to access application data from any location.
> When you need to build an enterprise data lake on Azure and do big data analytics.
When the external data schema is unknown, the plug-in infer_storage_schema aids in inferring the schema based on the contents of the external file.
