Smart Data Source Optimizer

In many enterprise Spotfire deployments, reports connect to wide tables or views via JDBC or Infolinks, often bringing in 50–200 columns by default—even when only a fraction of these columns are actually used in visualizations, expressions, or scripts (blame user laziness for not making a good house cleaning job).

This leads to:

• Unnecessary data transfer during loading and refresh operations.

• Increased memory usage on client and server nodes.

• Slower performance on opening and recalculating data tables.

• Inefficiency in long-term data model management and governance.

Proposed Feature: Smart Data Source Optimizer

This feature would analyze the relationship between the data source schema (e.g., JDBC table or view) and the actual report usage, and surface intelligent recommendations to streamline column loading.

Key Capabilities

• Column Usage Audit: Automatically scan the report to detect which columns are actively used in:

  • Visualizations (axes, colors, tooltips)

  • Expressions (custom columns, calculated fields)

  • Data functions (Python/TERR scripts)

  • Filters or markings

• Unused Column Detection: Flag any columns from the original data source that are loaded but not used anywhere.

• Optimization Suggestion Panel (in Canvas/Data Panel):

  • "You are using 9 out of 70 columns from this data source."

  • "61 columns are unused and could be excluded to reduce load time and memory by ~80%."

  • Button to preview unused columns and selectively disable them in the data connection or Infolink.

• Integration with Infolink Customization:

  • Automatically pre-populate a new Infolink view with only the used columns.

  • Optionally rewrite the SELECT clause to include only needed fields (for custom queries).