We’re continuing to evolve Microsoft Fabric Data Factory to make pipelines more intuitive to author, easier to operate, and more reliable at scale.
In this release, we’re introducing enhancements across authoring, observability, and data operations:
- Auto-populate base parameters for Notebook activity
- Activity-level (L2) workspace monitoring for pipelines (Preview)
- AI-powered Operations Agent for Pipelines (Preview)
- Refresh Materialized Lake View (RMLV) activity (Preview)
To learn more, refer to the Notebook activity in Fabric Data Factory documentation.
Gain deeper visibility with L2 workspace monitoring for pipelines (Preview)
Understanding pipeline behavior requires both high-level insights and detailed diagnostics. With L2 (activity-level) workspace monitoring, pipelines now provide deeper visibility into execution.
What’s new
Workspace monitoring now captures activity-level execution details.
L2 activity-level pipelines workspace monitoring table.
This builds on existing L1 monitoring, which tracks pipeline-level status such as overall success, duration, and system diagnostics.
L1 item-level pipelines workspace monitoring table.
Why it matters
With L2 monitoring, you can:
- Drill down from a pipeline run to specific failed activities.
- Access detailed diagnostics for root cause analysis.
- Debug issues without relying on external tools or APIs.
Workspace monitoring continues to store logs centrally in a monitoring event store, enabling query and analysis across all pipeline runs.
This unlocks a more complete observability experience—moving from reactive monitoring to actionable insights.
To learn more, refer to the Enable Workspace Monitoring in Microsoft Fabric documentation.
Improve Pipeline Reliability with Operations Agent for Pipelines (Preview)
Managing production data pipelines often requires manually reviewing run history, logs, and diagnostic information to understand failures or performance issues. Operations Agent for Pipelines helps simplify this process by continuously monitoring pipeline health and providing actionable recommendations.
What's new
Operations Agent is an AI-powered capability built directly into the pipeline experience that helps you:
- Monitor pipeline health across your workspace.
- Detect failures and performance issues proactively.
- Analyze run data and operational signals automatically.
- Receive actionable recommendations to resolve issues faster.
- Reduce the time spent manually investigating pipeline executions.
Configuring the Operations Agent in Pipelines.
Rather than requiring users to sift through logs and diagnostics, Operations Agent helps surface relevant insights and guidance so teams can maintain reliable data operations at scale. It follows a continuous workflow of observe → analyze → recommend → act, helping move pipeline operations from reactive troubleshooting to proactive monitoring.
Why it matters
- Accelerate troubleshooting by identifying likely root causes of failures.
- Reduce manual effort spent gathering logs and investigating issues.
- Improve operational visibility with continuous health monitoring.
- Increase reliability through proactive recommendations and remediation guidance.
Operations Agent is integrated directly into the Fabric Data Factory experience, making it easy to configure monitoring for individual pipelines or broader workspace scenarios.
Operations Agent for monitoring pipeline health.
To learn more, refer to the Operations Agent for Fabric Data Factory Pipelines documentation.
Operationalize data transformations with RMLV activity (Preview)
Maintaining and refreshing materialized data structures is a common requirement in modern analytics workflows. The Refresh Materialized Lake View (RMLV) activity, now in Public Preview, brings this directly into pipelines.
What’s new
The RMLV activity enables you to:
- Trigger refresh operations for materialized lake views directly within pipelines.
- Eliminate the need for manual refresh configuration.
- Integrate lake view maintenance into end-to-end orchestration.
Refresh materialized lake view activity in the pipeline canvas.
Why it matters
By bringing refresh operations into pipelines:
- Data stays fresh and consistent without manual intervention.
- Orchestration becomes fully automated end-to-end.
- Operational overhead is reduced for data teams.
This is part of a broader effort to make pipelines not just orchestrators—but the control plane for data operations.
To learn more, refer to the Refresh Materialized Lake View Activity in Fabric Data Factory Pipelines documentation.
Looking ahead
Together, these enhancements make it easier to author, monitor, and operate pipelines in Microsoft Fabric Data Factory. Whether you're reducing manual configuration, improving troubleshooting, or automating operational tasks, these capabilities help streamline end-to-end orchestration and support more reliable data operations at scale. Explore the documentation for each feature to get started.
Microsoft Employee