naming-tagging
Naming Convention for Microsoft Fabric
A clear naming convention in Microsoft Fabric improves manageability, governance, and automation. Fabric resources are logically grouped within domains, workspaces, and items, so naming should reflect their structure and purpose.
Naming Pattern Components
| Component | Description | Example |
|---|---|---|
| Org | Short identifier of the organization or team | contoso |
| Domain | Functional domain or capability area | finance, hr |
| Workspace | Logical grouping of items and artifacts | forecasting |
| Item Type | Abbreviation of the Fabric item type | lh, dw, ml |
| Item Name | Name of the item, unique within a workspace | salesdata |
| Environment | Lifecycle stage: dev, test, prod | prod |
| Region (opt.) | Short Azure region code if needed | weu, eus |
Abbreviations for Fabric Artifacts
| Experience | Artifact Type | Abbreviation |
|---|---|---|
| Power BI | Dataset | pbi-ds |
| Dataflow (Power BI) | pbi-dfl | |
| Datamart | pbi-dm | |
| Data Factory | Pipeline | df-pl |
| Dataflow (Data Factory) | df-dfl | |
| Data Engineering | Lakehouse | de-lh |
| Notebook (Eng.) | de-nb | |
| Spark Job Definition | de-sj | |
| Data Science | Experiment | ds-exp |
| Model | ds-mdl | |
| Notebook (DS) | ds-nb | |
| Data Warehouse | Warehouse | dw-wh |
| Real-Time | Eventstream | rta-es |
| Database | rta-db | |
| Queryset | rta-qs | |
| Data Activator | Trigger/Rule Set | da-act |
Hinweis: Die Abkürzungen sollten konsistent in Namen von Artefakten verwendet werden, z. B.
de-lh-100-bronze-sales, um Ordnung, Auffindbarkeit und Automatisierung zu erleichtern.
Naming Format
Workspaces
ws-<org>-<domain>-<purpose>-<env>
Example: `ws-contoso-sales-forecasting-prod`
Items (e.g., Lakehouse, Warehouse, ML Model)
<type>-<org>-<domain>-<workspace>-<item>-<env>
Example: `de-lh-contoso-sales-forecasting-salesdata-prod`
Notebooks, Pipelines, Datasets (optional suffixes)
nb-<workspace>-<purpose>-<env>
ds-<workspace>-<entity>-<env>
pl-<workspace>-<task>-<env>
Examples:
nb-forecasting-cleaning-prod
ds-forecasting-orders-prod
pl-forecasting-dailyload-prod
Best Practices
- Use lowercase letters and hyphens only.
- Keep names under 63 characters if possible.
- Avoid sensitive data in names.
- Apply the convention uniformly to support automation and search.
- If an item does not clearly fall into one of the defined types or structures, use the fallback format:
gen-<org>-<domain>-<workspace>-<item>-<env>. This ensures consistency while allowing flexibility for future artifact types.