Edit

Data retention overview

  • Article

Important

Azure SQL Edge no longer supports the ARM64 platform.

Collection and storage of data from connected IoT devices is important to drive and gain operational and business insights. However, with the volume of data originating from these devices, it becomes important for organizations to carefully plan the amount of data they want to retain and at what granularity. While retaining all data at all granularity is desirable, it's not always practical. Additionally, the volume of data that can be retained is constrained by the amount of storage available on the IoT or Edge devices.

In Azure SQL Edge, database administrators can define data retention policy on a SQL Edge database and its underlying tables. Once the data retention policy is defined, a background system task runs to purge any obsolete (old) data from the user tables.

Note

Data once purged from the table, isn't recoverable. The only possible way to recover the purged data is to restore the database from an older backup.

Quickstarts:

How data retention works

To configure data retention, you can use DDL statements. For more information, Enable and Disable Data Retention Policies. For automatic deletion of the obsolete records, data retention must first be enabled for both the database and the tables that you want to be purged within that database.

After data retention is configured for a table, a background task runs to identify the obsolete records in a table and delete those records. If for some reason, the automatic cleanup of the tasks isn't running or is unable to keep up with the delete operations, then a manual cleanup operation can be performed on these tables. For more information on automatic and manual cleanups, see Automatic and Manual Cleanup.

Limitations

  • Data Retention, if enabled, is automatically disabled when the database is restored from a full backup or is reattached.
  • Data Retention can't be enabled for a Temporal History Table
  • Data Retention filter column can't be altered. To alter the column, disable data retention on the table.

Next steps