Amazon QuickSight Collected Data

This page lists what Sifflet imports from Amazon QuickSight and how it builds lineage.

Imported Assets

Sifflet imports the following assets from the AWS account and region configured in the source:

Asset type	Description
Dashboards	Published QuickSight dashboards (the published version of each dashboard)
Analyses	QuickSight analyses, except deleted ones
Sheets	The sheets of each dashboard and analysis
Datasets	QuickSight datasets used by dashboards and analyses
Table references	The physical tables, custom SQL queries, and S3 files that each dataset reads, used to link datasets to your warehouse tables

Collected Metadata

Dashboards: name, description of the published version, sheets, and datasets used.
Analyses: name, sheets, and datasets used.
Sheets: name.
Datasets: name, and the physical tables they read.
Table references: schema and table name, column names and types, and the SQL text of custom SQL tables.
QuickSight data sources: type and connection details (host, port, database, Snowflake warehouse), used only to resolve lineage.

Sifflet doesn't collect usage statistics (views) for QuickSight assets.

Lineage

Sifflet builds lineage from the QuickSight API: each dashboard and analysis is linked to the datasets it uses, and each dataset to the tables it reads. Sifflet then links these tables to the matching tables of your other Sifflet sources, by comparing the connection details of the QuickSight data source (host, port, and database) with those of the Sifflet source.

Lineage to warehouse tables is supported for QuickSight data sources of these types:

Amazon Redshift
Snowflake
PostgreSQL
MySQL
Microsoft SQL Server

For datasets defined with custom SQL on a Snowflake data source, Sifflet parses the query and links the dataset to each table it reads. Only schema-qualified table names (schema.table or database.schema.table) are linked.

Data Freshness

Sifflet refreshes QuickSight metadata on the source's schedule. See Integrations Management for default frequencies and scheduling.

Limitations

Lineage to warehouse tables isn't available for datasets based on other QuickSight data source types, such as Amazon Athena, Amazon S3, Google BigQuery, or Oracle. These datasets and their dashboards still appear in Sifflet, without upstream links to your warehouse tables.
For custom SQL on data sources other than Snowflake, the whole query appears as a single table reference instead of links to each table it reads.
The Sifflet Insights browser extension doesn't support QuickSight.

Data Exposure

Sifflet only reads QuickSight metadata: it never reads or stores the rows of your datasets. The SQL text of custom SQL datasets is stored to display lineage. See Security for more details on the data Sifflet stores.