In today’s data-driven world, effective monitoring and governance of data environments are crucial for organizations. This is where FUAM comes into play.
This open-source project, developed by Microsoft cloud solution architects Gellert Gintli and Kevin Thomas, is designed to provide a high-level overview of activities within Fabric and Power BI tenants. Let’s dive into what FUAM is, why it’s needed, and how it can benefit your organization.
What is FUAM?
FUAM stands for Fabric Unified Admin Monitoring. It is not an official solution but an open-source project available on GitHub. The primary goal of FUAM is to offer a comprehensive solution for monitoring and governing Fabric and Power BI environments, which includes detailed insights into tenant activities, capacity usage, reporting usage and complete list of inventories, helping Fabric administrators manage their data environments more effectively.
Overview in FUAM Core report
Why Do We Need FUAM?
Managing a Fabric or Power BI tenant involves overseeing a wide range of activities, from backend activities like pipeline and notebooks creation to reporting activities like report distribution and consumption. These activities can range from citizen developers to CEOs and CIOs, with reports being used for daily tasks or critical business decisions. However, existing tools for monitoring these environments have limitations, particularly in terms of data retention and insights that they can address. Refer to this blog for an intro to Admin Monitoring workspace.
Out-of-the-box reports for Admin Monitoring
Current available admin dashboards in Power BI, such as Content Sharing, Purview Hub, and Feature Usage and Adoption reports, offer valuable insights but have short data retention periods (only up to 90 days depending on the data source). This makes long-term auditing and trend analysis challenging. FUAM addresses these limitations by providing a solution that collects and stores data for future use, enabling more comprehensive monitoring and governance.
Key Features of FUAM
FUAM is not just a dashboard; it is a full solution that starts from raw data collection and builds a comprehensive monitoring system
Overview of the FUAM architecture
Data Collection and Storage
FUAM collects raw data from various sources, stores it in a Lakehouse, and uses Medallion architecture. This ensures that historical data is retained for future auditing and analysis and provide the potential for self-service analysis.
Comprehensive Reports
FUAM comes with multiple reports that provide detailed insights into tenant activities:
- Core Report: Offers insights related to tenant settings, capacity metrics, activity logs, inventory and more.
- Semantic Model Analyzer: Analyzes selected semantic model.
- SQL Endpoint Analyzer: Provides insights into SQL endpoints.
- Item Analyzer: Examines individual items within the tenant.
- Gateway Monitoring: Monitors gateway activities through gateway log.
Modular Architecture
FUAM’s modular approach simplifies scheduling processes. Each module contains the end-to-end logic for a specific data injection, and all modules are orchestrated in a central pipeline. This ensures efficient data collection and processing.
Benefits of FUAM
FUAM offers a lot of benefits for Fabric administrators or the BI team in managing their Fabric and Power BI tenant
Pages available in the Core Report
Historical Data Retention
By storing data in a Lakehouse, FUAM addresses the issue of short data retention periods in existing tools. This enables long-term auditing and trend analysis, helping organizations make informed decisions based on historical data. Particularly in the capacity metrics and audit log data.
Ease of Deployment
As long as all the prerequisites are fulfilled, FUAM can be set up in less than a day, making it easy for organizations to start collecting and analyzing data quickly.
Cost of FUAM
The solution is open-source, with no other licensing costs except for Fabric capacity.
Challenges and Considerations
While FUAM offers many benefits, there are some challenges and considerations to be aware of.
Overwhelming Insights
The reports can be overwhelming due to the amount of data it provides. It is recommended to start with the Core report before exploring other reports to avoid feeling overwhelmed. Understanding your own challenges and requirements first, will also help in navigating these reports.
Workspaces Limitations
FUAM currently focuses on workspaces and inventory on Fabric capacity. Personal workspaces or shared capacity workspaces may be missing from the data collection. However, with some tweaks in the data pipeline, you may be able to bring these data into the Lakehouse.
Possible Higher Capacity Costs for Large Organizations
Larger organizations may face higher consumption costs due to the increased data collection and processing requirements. We are not particularly concerned about the backend activities cost which is smoothed by default, instead more on the XMLA connector that is being used for extracting Capacity Metrics data. XMLA read falls under interactive operation, which means the capacity usage will spike that during data extraction. Refer to the red spikes denoted by the black arrows.
Capacity usage for FUAM pipeline on daily refresh. Capacity SKU is F2.
Missing Lineage Information
FUAM does not currently include lineage information, which may be a limitation for some organizations. In the process of governing data sources, lineage information would be useful to understand how your data is being distributed to different means of consumption. Currently we are looking into integrating our own ETL process to address this gap.
A sample of lineage view
FUAM + Power BI
FUAM is a powerful admin tool to monitor and govern your Power BI environment and analyze the usage of your Fabric Capacity.
It provides insights into your Fabric inventory, report views, capacity usage and many more. This in turn will allow Fabric admins to monitor and manage the data environment effectively.
Despite the challenges mentioned, we highly recommend that FUAM is implemented so your organization can start collecting data now for future analytical use.
If you would like to implement FUAM, please refer FUAM on Github:
https://github.com/microsoft/fabric-toolbox/blob/main/monitoring/fabric-unified-admin-monitoring/how-to/How_to_deploy_FUAM.md
If you need help implementing FUAM, please feel free to reach out to us.