DynamoDB Support

Supported Features

• Table data view (Scan-based)
• Filtering by table column
• Table structure view (sampled — DynamoDB is schemaless)
• Entity sidebar
• Editing data (insert / update / delete)
• Running PartiQL queries
• Read-only mode
• IAM key, profile, and AWS CLI credential resolution
• Connecting to DynamoDB Local via custom endpoint

Limitations

DynamoDB is not a relational database and many table-view features are mapped onto its API as best they can be. The following caveats apply:

Sorting

DynamoDB has no server-side ORDER BY for Scan, so column sorting in the table view is not available. Sorting would require pulling the entire table into memory and sorting locally, which is too expensive to do while browsing.

Filtering

Table-view filters are mapped onto DynamoDB FilterExpression. Filtering happens after Scan, so highly selective filters on large tables can still trigger long scans even though only a few rows are returned.

Schema discovery

DynamoDB only declares attribute types for indexed attributes. The column list is inferred by sampling a small number of rows (configurable via db.dynamodb.columnSampleSize), so attributes that appear only in unsampled rows may not show up in the structure view.

Other unsupported features

• ALTER (add/drop/rename column, etc.) — schemaless, not applicable
• Foreign key relationships
• Triggers, routines, materialized views
• Server-side TRUNCATE — drop and recreate the table instead
• Renaming tables