It's Monday morning. The executive dashboard is slow, last night's pipeline failed without anyone noticing, and the capacity is being throttled. By the time someone raises a ticket, the business has already lost trust in the data.

This session shows how to catch these problems before your users do. We'll look at Fabric's new observability features:

Monitor hub: One place to see job runs, alerts, capacity and agent activity across your Fabric estate.
Workspace monitoring: Collects logs and metrics you can query, alert on, and pull together from many workspaces into one view.
Operations agent: When a pipeline fails, it investigates the run and suggests likely root causes.
We'll also look at how capacity smoothing and throttling actually work, and why repeated heavy workloads, not one-off spikes, are usually behind a slow report. Finally, we'll cover what's coming next: capacity health in the Monitor hub, and monitoring for Fabric data agents.

You'll leave with a practical blueprint for proactive monitoring, including the design decisions to make on day one, before they become expensive to change.

What you'll learn

How the Monitor hub, workspace monitoring and Operations agent fit together
How to set up central monitoring across multiple workspaces, and the limits to plan around
How to read capacity signals and spot throttling risk early
How to turn telemetry into alerts so the right people know before the report goes red
Which features are available now, in preview, or coming soon