Plot time series from data model views

Documentation Index

Fetch the complete documentation index at: https://docs.cognite.com/llms.txt

Use this file to discover all available pages before exploring further.

​

Create a dashboard

1

Sign in and create a dashboard

Sign in to your Grafana instance and create a dashboard.

2

Select time series for your dashboard

Use the query tabs below the main chart area to select time series for your dashboard.

If your query tabs are missing, check the feature flags in the data source settings.

• CogniteTimeSeries - search for time series from data model views. Select a View and start typing to filter and select a time series.
  Optionally, set a custom label and use the format {{property}} to pull data from the time series. You can use all the available time series properties to define a label, for example, {{name}} - {{description}} or {{metadata.key1}}.
  Use this tab when your time series are defined in data models. The View dropdown lists views that implement the CogniteTimeSeries container and shows the view name, space, and version. After you select a view, search by name or description to find time series instances. This tab filters out string time series, and selections use the instance ID (space and external ID).

If a time series has a unit, you can select a target unit in the query. The target unit dropdown appears only when a unit is defined, shows compatible units in the same quantity, and displays the current unit for reference.

• Data models - run a GraphQL query to fetch instances from a data model and plot their time series. See Display time series from the Data models tab below.

3

Review and adjust the timeframe

The time series matching your selection will be rendered in the chart area. Adjust the timeframe as necessary to display relevant data.

​

Display time series data from data model instances

1

Select the data model and version

On the Data models tab, select the data model and version.

2

Specify the query

Add __typename and externalId below the fields that contain the time series. In this example, below temperature and pressure:

{
  listPump {
    items {
      temperature {
        __typename
        externalId
      }
      pressure {
        __typename
        externalId
      }
    }
  }
}

Grafana automatically detects time series in your query results. If your query returns fields with type: "numeric" that include space and externalId, Grafana treats them as time series and fetches data points automatically. Keep __typename and externalId in the query to support older detection behavior and queries that return mixed instance types.