Object Dashboard

The object dashboard provides the most detailed description of a single database object. Select a database tile from the Database Summaries section of a server's dashboard to open the database dashboard, then select a table, index, or query from one of the leaderboards to open the object dashboard for the selected object.

Reviewing Table Details

When you select a table name from the Table Leaderboard, the console displays details about that tables size, performance, and maintenance history. The dashboard summarizes the table using the latest pg_stat_all_tables snapshot and a set of time-series charts.

The header shows the fully qualified schema.table name above three sections; the Table Overview, Activity Charts, and Maintenance Info. Use the arrow at the right side of each section to open or close that section.

The Table Overview section presents four key statistics for the table:

• The Live Tuples tile shows the estimated number of live rows, formatted with thousands separators.
• The Dead Tuples tile shows dead rows with a % dead qualifier computed as dead tuples divided by live plus dead tuples. The dead tuple percentage is color-coded, marking values at or below 5% good, values through 20% warning, and higher values critical.
• The Table Size tile shows the on-disk table size in a human-readable form (for example, 480 kB).
• The Sequential Scans tile shows the cumulative sequential scan count for the table.

A statistic without a recorded value displays -- in place of a number.

The Activity Charts section renders three line charts that conform to the dashboard time-range selection and refresh with the dashboard:

• The Tuple Operations Over Time chart plots per-second rates for inserts, updates, deletes, and HOT updates.
• The Sequential vs Index Scans Over Time chart compares the per-second sequential scan rate against the index scan rate.
• The Dead Tuple Ratio Over Time chart plots the dead tuple percentage as a shaded area to highlight vacuum effectiveness.

Each chart shows a message such as "No tuple operation data available" when the selected time range contains no samples.

The Maintenance Info panel reports the four maintenance timestamps that PostgreSQL records for the table:

• The Last Vacuum field shows the time of the last manual VACUUM.
• The Last Autovacuum field shows the time of the last scheduled autovacuum run.
• The Last Analyze field shows the time of the last manual ANALYZE.
• The Last Autoanalyze field shows the time of the last scheduled autoanalyze run.

A field displays Never when the corresponding operation has not run since statistics were last reset.

Reviewing Index Details

When you select an index name from the Index Leaderboard, the console displays details about the indexes size and usage. The index detail summarizes a single index using the latest pg_stat_all_indexes snapshot and a time-series chart.

The header shows the fully qualified schema.index name, followed by a line naming the table that the index serves. Two sections follow: Index Overview and Scan Activity.

The Index Overview section presents four key statistics for the index:

• The Index Size tile shows the on-disk index size in a human-readable form, such as 176 kB.
• The Index Scans tile shows the cumulative number of scans started on the index, formatted with thousands separators.
• The Tuples Read tile shows idx_tup_read, the number of index entries returned by scans of the index.
• The Tuples Fetched tile shows idx_tup_fetch, the number of live table rows fetched by simple index scans that use the index.
• The fetched count usually trails the read count, because index-only scans and bitmap scans avoid fetching individual table rows.

A statistic without a recorded value displays -- in place of a number.

The Scan Activity section renders a line chart that follows the index activity for the duration of the database dashboard's time-range selection:

• The Index Scan Activity Over Time chart plots the per-second index scan rate as the Index Scans/s series.

The chart displays the message No index scan data available when the selected time range contains no samples.

Reviewing Query Details

When you select a query from the Top Queries section of a server dashboard, the console displays details about query execution. The query detail dashboard summarizes a single query using the latest pg_stat_statements data and a set of time-series charts.

The header includes the database name, the server name, and the text of the query. If you have configured AI in the Workbench, an AI Overview panel evaluates the query. See AI Query Analysis for details on the summary panel, the analysis report, and running suggested SQL from the report.

The header normalizes the query onto a single line, collapsing runs of whitespace into single spaces for a compact preview. Select the Show full query toggle to display a query that exceeds 120 characters; the preview ends with an ellipsis.

When you're finished, you can use the Show less toggle to restore the compact preview.

The Query Plan section displays a graphical interpretation of the PostgreSQL query plan:

The Visual diagram uses a left-to-right layout. Leaf scan nodes display on the left and the root node displays on the right. SVG bezier arrows connect each child node to its parent.

Each tile in the diagram displays the node type and the relation or index name. A colored left border indicates the cost ratio relative to the root node:

• A red border marks nodes that consume over 80 percent of the total cost.
• An orange border marks nodes that consume over 50 percent of the total cost.
• The default border color applies to all other nodes.

Click a tile to open a popover with comprehensive node details. The popover displays the following information:

• The cost range from startup cost to total cost.
• The estimated row count and row width.
• The output columns the node produces.
• The execution strategy and scan direction.
• The planned and launched worker counts.
• Any filter, join, or index conditions.

Select the Text tab to display the execution plan with costs:

The Workbench uses standard EXPLAIN without VERBOSE for a concise explanation of the execution plan.

The Query Statistics section presents four key statistics drawn from the latest pg_stat_statements snapshot for the query.

Four tiles provide statistics about the query:

• The Total Calls tile shows the cumulative execution count, formatted with thousands separators.
• The Total Time tile shows the total execution time across all calls, scaled to a readable unit such as microseconds, milliseconds, seconds, or minutes.
• The Mean Time tile shows the average execution time per call, using the same automatic unit scaling as Total Time.
• The Avg Rows/Call tile shows the total rows divided by the call count, formatted to one decimal place. The Avg Rows/Call tile displays -- when the query has recorded no calls, because the average is undefined.

The Performance Charts section renders two charts and a time-range selector in the section header. The selector offers 1h, 6h, 24h, 7d, and 30d ranges that both charts share:

• The Execution Time Over Time line chart plots the Mean Time (ms), Min Time (ms), and Max Time (ms) series across the range.
• The Calls Over Time bar chart plots the number of calls recorded in each interval as the Calls series.

The execution time chart shows "No execution time data available", and the calls chart shows "No call frequency data available", when the selected time range contains no samples.