Skip to Content
This documentation is provided with the HEAT environment and is relevant for this HEAT instance only.
RunnersSystem Utilsheat-system-next-dimension

heat-system-next-dimension

Platform reserved (system-utils only). Aggregates upstream JSON containing canonical $heat-dataservice, writes it to object storage, and creates or updates a session dimension with DashboardVersion="Next" for ui/dashboard.

When to use it

Use in shipped platform session templates (for example tabular CSV dashboards or Resource Monitor) where HEAT owns the layout and payload pipeline. Integrators building custom analytics should use dashboard-v2 on dashboard-utils instead.

Limitations

  • Does not target ui/legacy (use heat-system-legacy-dimension).
  • Payload and layout rules match dashboard-v2; invalid realms or layout fail the node.
  • Set layoutFromParentOutput: true to take suggestedLayoutConfiguration from system-tabular-to-dataservice when emitSuggestedLayout is enabled.

Scenario instance and users

Scenario handling follows dashboard-v2 so both dimension kinds agree on a session:

  • One scenario per session. With autoCreateScenarioInstance and no scenarioName, the first dimension node to run creates the scenario and records it in session metadata (ScenarioInstanceId); later nodes reuse it. Name precedence: scenarioName, session metadata scenario_name, then scenario_title, then auto-dashboard. Templates set scenario_name from their own analytics (for example a system-arbex-js node via ctx.session.metadata.set).
  • Span. scenarioFromIsoKey / scenarioToIsoKey paths in the parent payload, else the session’s own startTimeUtc / closedAt metadata (open sessions use timeWindowSeconds), else an ISO-8601 scan of the payload. The scenario this node created is re-inferred on every run, so a span guessed before data arrived follows the data afterwards.
  • Users. dashboard_users in the payload (HEAT Auth or capture-origin GUIDs) is mapped onto the session’s registered users; unknown ids are ignored. Session users are registered upstream by the runners that know the participants (see dashboard-v2 upstream contract).