Skip to main content

Integrate with YHBI

YHBI is a business intelligence and data analytics platform. Its JDBC support enables you to add TDengine as a data source, read time-series data directly, and build visual reports without custom data-conversion code.

Prerequisites​

Prepare the following environment:

  • Deploy and start TDengine v3.3.2.0 or later (Enterprise or Community Edition).
  • Start taosAdapter. For details, see the taosAdapter Reference.
  • Install and start YHBI.
  • Download a recent TDengine JDBC connector from Maven Central. This example uses taos-jdbcdriver-3.4.0-dist.jar.

Configure the Data Source​

  1. In YHBI, choose Add Data Source, and select GENERIC under SQL data sources.

  2. Choose Select Custom Driver. In Driver Management, choose + next to the driver list and enter a name such as MyTDengine. Upload taos-jdbcdriver-3.4.0-dist.jar, select com.taosdata.jdbc.ws.WebSocketDriver, and confirm the driver.

  3. Copy the following value into the URL field:

    jdbc:TAOS-WS://127.0.0.1:6041?user=root&password=taosdata&conmode=1&varcharAsString=true

  4. Under Authentication Method, select No Authentication.

  5. In the advanced data-source settings, set Quote Character to a backtick (`).

  6. Test the connection. After the test succeeds, save the data source with a name such as MyTDengine.

  7. Choose Add Data Source, expand the data source you created, and browse the TDengine supertables.

  8. Load all data from a supertable or import a subset with custom SQL.

  9. To avoid caching TDengine time-series data in YHBI, enable In-Database Computing. YHBI then sends SQL requests directly to TDengine for processing.

Analyze Data​

After importing data, YHBI classifies numeric columns as measures and text columns as dimensions. In a TDengine supertable, ordinary columns typically represent measures and tag columns represent dimensions, so you might need to adjust column properties when creating a dataset.

You can create parameters in YHBI and use them in SQL that is executed manually or on a schedule. The following query reads real-time data from TDengine and supports dynamic table, time-range, and interval parameters:

SELECT _wstart AS ws, count(*) AS cnt
FROM supertable
WHERE tbname = ?{metric} AND ts = ?{from} AND ts < ?{to}
INTERVAL(?{interval})

The parameters and result columns are as follows:

  1. _wstart is the start time of each window.
  2. count(*) is the aggregate value for the window.
  3. ?{interval} specifies the downsampling interval. For example, 1m creates one-minute windows.
  4. ?{metric} specifies a table name. You can bind it to a drop-down parameter component whose ID is metric.
  5. ?{from} and ?{to} specify the dataset time range and can be bound to text parameter components.

In the Edit Parameter dialog, configure each parameter's data type, range, and default value. You can then change the values dynamically in the visual report.

To create a report:

  1. Choose Create Report in YHBI and create a canvas.
  2. Drag a visualization component, such as a table, onto the canvas.
  3. In the Dataset sidebar, select a dataset and bind its dimensions and measures to the component.
  4. Save the report and view the result.