Inventory scan uses Amazon S3 Inventory to back up and discover Amazon S3 data at scale. This scan type is designed for environments that contain millions or billions of objects.
Instead of enumerating objects by using Amazon S3 APIs, Commvault processes Amazon S3 Inventory reports. This approach reduces scan overhead and improves scalability for large buckets.
Requirements
The following resources are required:
-
Amazon S3 permissions to configure and access inventory reports.
-
s3:PutInventoryConfiguration
-
s3:GetInventoryConfiguration
-
-
Local storage for inventory report processing.
-
Network bandwidth for downloading inventory reports.
How inventory scan works
-
Commvault analyzes the configured subclient content for Amazon S3 buckets.
-
When the total object count exceeds 500 million objects, Commvault automatically creates an Amazon S3 Inventory configuration.
-
Inventory reports are stored in a destination bucket that uses the naming pattern
cvlt-region-name-uniqueid. -
Amazon S3 Inventory generates reports that contain object information and associated metadata.
-
Commvault processes the inventory reports instead of using Amazon S3 List APIs to enumerate objects.
-
Inventory scans depend on the scheduled generation of inventory reports and do not provide real-time object discovery.
-
Newly added or modified objects are detected only after the next inventory report is generated and processed.
Planning considerations
-
Inventory scan is intended for Amazon S3 environments that contain hundreds of millions or billions of objects.
-
Inventory scan relies on scheduled Amazon S3 Inventory reports and is not a real-time scanning method.
-
Changes to objects are detected only after the next inventory report becomes available.
-
Inventory report generation schedules affect how quickly newly added or modified objects are discovered.
Supported operations
-
Full backups based on inventory reports.
-
Object discovery using metadata from inventory reports.