Current Template Version: 2025.1.1
Large data solutions typically generate disparate and complicated data that can be difficult to converge into a single view. Workspace Monitoring provides a seamless, consistent experience with end-to-end visibility across Fabric workloads. Fabric Workspace Monitoring enables users to correlate events from origination to subsequent operations and queries across Fabric experiences and services.
Workspace Monitoring is a built-in solution that enables users to:
- drive root-cause analysis
- analyze historical logs
- detect anomalies
When Workspace Monitoring is enabled at the workspace level, the workspace will start sending the selected diagnostics to a KQL database within Fabric that can be queried using KQL or SQL.
for Workspace Monitoring
This module of the Fabric toolbox provides a set of pre-built reports (Real-Time Dashboard and Power BI Report) on top of the Fabric Workspace Monitoring feature to be able to easily monitor your workspace from day one.
There are two monitoring template options available:
| Real-Time Dashboard | Power BI Report |
|---|---|
| This template allows users to track workspace activities in real-time. It provides an operational tool for users to analyze issues within seconds, and zoom in to exact time periods that an event occured. | This template allows users to configure connections to the Monitoring Eventhouse to retain detailed historical activity data. |
This template, in combination with Workspace Monitoring features, allows users to track workspace activities in real-time. It connects directly to the underlying monitoring Eventhouse KQL database, providing an overview of items, operations, and users. Users can also compare DAX improvements between semantic models. Users can deep dive an analyze the Query and Ingestion patterns of Eventhouses.
The Real-Time Dashboard template is structured on the following way:
The Power BI allows users to configure connections to Monitoring Eventhouse where they can retain detailed historical activity data. This repo hosts Power BI report templates (.pbit) that you can point to your Monitoring Eventhouse databases to load data and get insights.
We have ported and enhanced the 'Fabric Log Analytics for Analysis Service Engine report template', which retrieved the data with the BYOLA approach.
Now, you can seamlessy connect and track your workspace items, operations, visuals etc. without leaving the SaaS experience from Microsoft Fabric.
The Power BI Report template is structured on the following way:
We recommend to use the Power BI report template for the new Workspace Monitoring feature in Microsoft Fabric, whenever you are migrating your workspace monitoring solution from (BYOLA) Log Analytics Workspace monitoring.
Before you begin: Workspace Monitoring features must be enabled for the Workspace under the Workspace settings.
The following step-by-step guide describes the initial configuration step of the dashboard/report template.
- Download the 'Fabric Workspace Monitoring Dashboard.json' template file from the repository
- Navigate to your workpace
- Click on 'New item'
- Create a new 'Real-Time Dashboard' item
- Provide a dashboard name
- Click on 'Create'
- Navigate to the Dashboard and select the 'Manage' tab
- Click on 'Replace from file'
- Select the downloaded 'Fabric Workspace Monitoring Dashboard.json' template
- Click on 'Data sources'
- Click on 'Add' on the side bar
- Select the OneLake data hub
- Select the preferred Monitoring KQL database from OneLake catalog
- Click on 'Connect'
-
Provide a data source name
-
The data source shall be connected to several tiles, parameters and base queries
-
Select the initial value of the Workspace parameter on the menu bar
- Optionally: Set Default value for Workspace parameter
- Navigate to 'Parameters' on the 'Manage' tab
- Click on 'Edit' at the 'Workspace' parameter
- Select a 'Default value'
- Optionally: Set Default value for Workspace parameter
-
'Save' the Dashboard
-
'Refresh' the Dashboard
-
Switch the UI experience to 'Viewing' mode
Congratulations! You are ready to go!
The following parameters are defined in the template:
| Parameter | Description |
|---|---|
| Query URI * | Globally unique identifier uri of the Eventhouse Monitoring database containing the Semantic model logs. |
| Days Ago To Start * | Load data from the specified day to the time the call was initiated. The maximum value you can select is 30 days. However, your Premium capacity/Fabric SKU memory limits apply to this parameter. If those limits are exceeded, the template might fail to refresh. |
| Days Ago To Finish * | Load data up to the specified number of days ago. Use 0 for today. |
| UTC Offset Hours * | An hourly offset used to convert the data from Coordinate Universal Time (UTC) to a local time zone. |
| Fabric Item Id | Optionally enter a fabric artifact (semantic model) Id. |
| RangeStart | Optionally enter the start date for incremental refreshes |
| RangeEnd | Optionally enter the end date for incremental refreshes |
- Download the 'Fabric Workspace Monitoring.pbit' file from the repository
- Open the report in Power BI Desktop
- Paste the URI of the Monitoring Database to the first parameter
- Sign-in with your Microsoft Account
- Click on 'Load'
- Save the report (for instance as a '.pbix' file)
- Publish the report in a preferred workspace
- Navigate to the settings of the semantic model
- Edit the credentials of the data source in Fabric
- Trigger the first initial refresh in Fabric
- Optionally - Schedule the semantic model refresh, if preferred
- Once the refresh has been succedeed, open the report
Optimize Power BI refreshes for the template Additionally, you can optimize the Power BI report template usage, you can configure the incremental refresh for each table.
Power BI report template
- Eventhouse Query Limits
- If the template refresh fails due to the data volume, you can filter the data by configuring the Fabric Item Id parameter.
Recommendation: We recommend to filter the detail pages for a subset of semantic models for a short time period to avoid the 50.000 row result set soft limit.
The Q&A report pages provides difference approaches to analyze the 'SemanticModelLogs' and 'EventhouseQueryLogs' with AI & ML.
We have prepared three approaches, which helps you in ad-hoc analysis scenarios:
The SM Query Diagnostic is a drill-through analyzer page, which focuses on the details of a given query (one OperationId).
Capacity utilization: Please note that the report templates and the items of the Workspace Monitoring are utilizing CU of your capacity. The Workspace Monitoring feature is not charged during public preview.
CPU & Memory metrics in Monitoring Eventhouse and templates: Please note that the CPU and Memory related metrics are not aggregated, smoothed or bursted, like the operations in the Microsoft Fabric Capacity Metrics App. A one-to-one comparison is for different reasons not possible. Billable capacity utilization and storage information are shown in the Microsoft Fabric Capacity Metrics App. https://learn.microsoft.com/en-us/fabric/enterprise/metrics-app
Please test the solution on a non-production Fabric SKU first without impacting other workloads on your tenant.
These templates (Real-Time Dashboard template and Power BI template) are not official Microsoft services.
Ideas/Suggestions: Submit your ideas and suggestions as an issue in this repository.
Bug Reports: We maintain a backlog in the project issues page. Let us know if you run into any problems or share your suggestions by adding an entry into the issues section.
Important Please, do not open a support ticket in case of an issue regarding the templates. In case of any questions, issues regarding the templates, please create an issue in this repository.
For more information, please visit the main page of this repository or the learn.microsoft.com website.
This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft trademarks or logos is subject to and must follow Microsoft's Trademark & Brand Guidelines. Use of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship. Any use of third-party trademarks or logos are subject to those third-party's policies.