Eventhouse monitoring

Eventhouse monitoring in Fabric offers comprehensive insights into the usage and performance of your eventhouse by collecting end-to-end metrics and logs for all aspects of an Eventhouse. It's part of workspace monitoring that allows you to monitor Fabric items in your workspace. Eventhouse monitoring provides a set of tables that you can query to get insights into the usage and performance of your eventhouse. Use these queries to optimize your eventhouse and improve the user experience.

Prerequisites

A workspace with a Microsoft Fabric-enabled capacity
An eventhouse in your workspace.

Enable Eventhouse monitoring

From the Eventhouse explorer pane, select Monitoring.

You can also access the monitoring eventhouse from the workspace where your eventhouse is located. From the workspace where your eventhouse is located, select Workspace Settings and select the Monitoring tab.
Select + Eventhouse and wait for the success message confirming that the monitoring eventhouse was added to your workspace.

Workspace activity logging is automatically enabled.
Select the Monitoring database link to open the Monitoring Eventhouse in a new tab.
In the Monitoring KQL database overview page, view the query insights and query the data.

Tip

You can also analyze data with an SQL analytics endpoint or notebook, or create a standalone KQL queryset to analyze the monitoring data.

Eventhouse monitoring tables

When monitoring your eventhouse, you can query the following tables:

Capacity throttling and eventhouse size

Capacity throttling and eventhouse size monitoring logs provide eventhouse users and workspace admins with detailed visibility into throttling events, sub-optimal sizing, and scale-out activities to proactively manage performance and costs. These logs let you set alerts on throttling states and scaling reasons.

Capacity throttling logs track throttling states with updates every 30 seconds to 1 minute, helping users understand impacts such as interactive delays or query rejections.
Sub-optimal size logs report when an Eventhouse remains undersized after throttling exit, prompting actionable recommendations like adding capacity or reducing load.
Scale-out event logs aggregate reasons for scale-out over 24 hours, including high CPU or memory use and ingestion load, enabling users to optimize workloads and prevent unexpected cost spikes.

Operations

Operations logs provide eventhouse users with detailed visibility into queries, commands, ingestion, and data operations to monitor usage patterns, troubleshoot issues, and optimize performance.

Metrics logs track ingestion, materialized views, and continuous export activity.
Command logs record commands run on your eventhouse KQL databases, including performance and resource use.
Data operation logs capture internal data operations such as batch ingestion, update policies, and materialized views.
Ingestion results logs report the outcome of queued ingestion operations, including failures.
Query logs record queries run against your eventhouse databases, enabling analysis of performance and usage patterns.

Sample queries

You can find sample queries in the workspace-monitoring folder of the fabric-samples GitHub repository.

Monitoring built-in templates

The templates let you create monitoring dashboards that track workspace activities in real-time by connecting directly to the underlying monitoring Eventhouse. In the dashboard, you can monitor semantic models, KQL database queries, and ingestions.

Follow the Visualize your Workspace Monitoring Data guide to create the templates or download the templates from the workspace-monitoring-dashboards GitHub repository or from the links in the following section.

The repository contains instructions on how to prepare the environment and how to use the templates.

Two monitoring templates are available: