This blogpost will cover some of the aspects of this – including:
- Primarily setup – how to get started
- Monitoring state of databases that are in ‘stretch mode’
- Daily work with stretch databases
- Backup – what’s good to know
With the release of SQL Server 2016, the new feature called stretch database is also released. The feature lets you as a database administrator, make databases stretch (read: copy old data) to an Azure environment. The data is still able to respond to the normal queries that are used, in other way; there is no need to change the current setup for existing applications and other data-contracts to use this feature.
So when is the stretch database something you should consider
- When you only sometimes need to query the historical data
- The transactional data that are stored needs all historical data
- The size of the database tables are growing out of control (but not an issue of bad design – then you need to take other actions…)
- The backup times are too long in order to make the daily timeslots for maintenance
If you have one or more marks on the above list, then you have a database that are candidate for stretching into Azure.
A typical database in stretch mode are a transactional database with very large amounts of data (more than a billion rows) stored in a small number of tables.
The feature is applied to individual tables in the database – but a need for enabling the feature on database level is a prerequisite.
– read more at http://www.sqlshack.com/the-dbas-guide-to-stretch-database/