Microsoft Fabric slashes enterprise storage costs with general availability of OneLake lifecycle management 

Source: Microsoft 

Microsoft Fabric now enables automated storage tiering and lifecycle management in OneLake, allowing enterprise data teams to significantly reduce TCO for long-term retention without sacrificing data availability. By automatically moving stale or historical datasets from hot to cool or cold tiers, organizations can optimize consumption while keeping all data fully online and accessible across the enterprise estate. 

Architectural flexibility with multi-tier storage 

OneLake now natively supports hot, cool, and cold storage tiers, transforming how enterprise data lakes handle long-term retention. The hot tier remains the baseline for active ETL pipelines and high-frequency reporting, while cool and cold tiers deliver substantial cost reductions for archival or compliance data. 

Transitioning data between tiers requires no data migration or schema changes. Files remain accessible online within OneLake, ensuring seamless query continuity across the platform while shifting transaction costs only when cooler data is accessed. 

Granular control and dynamic auto-tiering 

To eliminate manual maintenance, platform administrators can configure customized lifecycle rules tailored to organizational processing patterns. For instance, data engineers can set automated rules to transition raw “bronze” data to cool storage 30 days after processing ends. 

For unpredictable workloads, the AutoTierFromCoolToHot policy dynamically manages access spikes by moving data to cool based on inactivity and automatically promoting it back to hot upon access. Administrators can also set workspace-level default tiers for dedicated historical archives to ensure new incoming files inherit lower-cost storage settings immediately. 

Strategic impact for enterprise platform leaders 

For enterprise CDOs and Data Platform Managers operating large-scale Fabric environments, storage expenses can quickly accumulate if left unmanaged. OneLake lifecycle management turns passive data retention into an automated governance policy, aligning operational expenditure directly with business value. 

By establishing systematic tiering rules early, enterprise architects can ensure predictable capacity consumption while keeping historical datasets readily available for downstream analytics and enterprise AI models. 

Learn more on Fabric Updates Blog 

Microsoft Fabric simplifies workspace cost control with OneLake storage reporting 

Source: Microsoft 

Microsoft Fabric now enables workspace administrators to instantly audit and manage data volume across distributed items using the newly generally available OneLake storage report. By replacing manual inspections with single-click visibility into active, system, and soft-deleted data, enterprise data teams can proactively eliminate storage sprawl, enforce retention policies, and optimize platform consumption costs directly within the Fabric portal. 

Granular visibility into distributed storage footprints 

Managing data estates across dozens or hundreds of items in a single workspace often leads to unmonitored storage expansion and unexpected consumption charges. The general availability of OneLake item-size reporting addresses this operational challenge by providing an immediate, item-level breakdown of storage consumption. Workspace administrators can search, sort, and analyze items by size, while distinguishing between user-visible data, system-managed overhead, and soft-deleted assets. This granular breakdown helps data engineers and platform owners pinpoint data bloat and identify unneeded artifacts that unnecessarily drive up operational costs. 

Operationalizing storage lifecycle management and governance 

Beyond surface-level metrics, the report provides critical insights into the billing status and active footprint of individual Fabric items. By surfacing refreshed storage calculations directly inside workspace settings, platform managers can establish structured lifecycle management policies tailored to their highest-volume assets. Instead of spending valuable engineering hours writing custom scripts or conducting manual checks, administrators can now direct their governance efforts toward specific high-impact workloads, ensuring that capacity usage aligns directly with organizational value. 

For enterprise data leaders overseeing large-scale Fabric deployments, storage governance is directly linked to cost efficiency and platform sustainability. As data teams scale their analytics and AI operations across multiple business units, having immediate, centralized oversight of OneLake capacity prevents wasteful allocation and keeps cloud spend predictable.  

Learn more on Fabric Updates Blog 

User Data Functions turn Microsoft Fabric APIs into a secure enterprise Gateway 

Source: Microsoft 

Microsoft Fabric now enables data teams to centralize and secure platform automation by using User Data Functions (UDFs) as a governable API gateway. By combining service principal authentication with Azure Key Vault, enterprise organizations can now trigger pipelines, manage workspaces, and automate capacity management without exposing hardcoded credentials or scattering complex integration logic across multiple applications. 

Centralized, credential-free automation 

Calling Microsoft Fabric REST APIs directly across scattered notebooks, reports, and third-party applications often introduces security risks and maintenance overhead. The introduction of User Data Functions (UDFs) solves this by providing a dedicated, reusable logic layer. 

A UDF securely retrieves service principal secrets from Azure Key Vault via Fabric generic connections, instantiates a scoped Fabric client, and executes administrative or operational tasks. This architecture abstracts the underlying REST API calls, allowing workloads to trigger data pipelines, manage Fabric items, or read workspace metadata through a single controlled interface. 

Enterprise governance and operational scale 

For organizations managing complex data ecosystems, hardcoded credentials and sprawling deployment scripts represent a significant governance vulnerability. By routing API interactions through a UDF backed by a service principal, access control remains strictly defined by the identity’s permission scope rather than the invoking application. 

This separation of concerns ensures that business logic, identity management, and platform execution remain isolated. Furthermore, the pattern applies universally across the platform. A single UDF implementation can automate end-to-end operational flows, ranging from spinning up new workspace capacities to executing scheduled Data Factory pipelines. 

Onboarding automation for data platform leaders 

This capability delivers immediate value to Data Platform Managers, Lead Data Engineers, and Enterprise Architects seeking to streamline platform operations. A practical scenario includes automated project onboarding: when an internal team requests a new data environment, an enterprise app can invoke a UDF with metadata parameters like project name and capacity target. The function provisions the workspace, assigns capacities, and sets up initial assets automatically. 

The business bottom line 

By turning platform management into a governed API service, enterprise IT teams reduce credential exposure, lower administrative burden, and enforce consistent security policies across all Fabric workloads. Organizations can safely open platform capabilities to business-line applications while keeping core assets secure. 

Learn more on Fabric Updates Blog 

Microsoft Fabric bridges real-time streaming and operational AI for instant business action 

Source: Microsoft 

Microsoft Fabric has expanded its real-time capabilities by combining Change Data Capture (CDC), DeltaFlow, and PySpark AI functions within Eventstreams. This integration allows enterprise data teams to classify, interpret, and route operational database changes as they happen, moving beyond simple numeric thresholds to true semantic, context-aware automation. 

Shifting from reactive thresholds to semantic event routing 

Traditional real-time architectures rely on static, rule-based alerts triggered by numeric thresholds, such as inventory dropping below a specific quantity. The integration of AI functions directly into Spark Structured Streaming micro-batches changes this dynamic entirely. Operational databases capturing continuous inserts or updates can now process unstructured text alongside structured fields in real time. 

By embedding logic like ai.classify or ai.extract into the streaming pipeline, organisations can evaluate the actual meaning of incoming operational logs, routing high-priority requests or anomalies to downstream systems within seconds without operating a dedicated model server. 

Streamlining end-to-end architectures for low-latency operations 

From an architectural perspective, this pattern eliminates the complexity of managing separate ingestion engines, standalone inference endpoints, and external event orchestrators. Fabric Eventstreams captures change feed data via native CDC connectors, while DeltaFlow reshapes raw Debezium envelopes into analytics-ready tables. 

Source: Microsoft 

The streaming notebook enriches these events in flight and publishes them as first-class Business Events. From there, automated workflows can trigger Microsoft Teams alerts, open service tickets via Power Automate, or update Real-Time Dashboards in Eventhouse simultaneously, maintaining predictable end-to-end latency across high-throughput scenarios. 

Why enterprise data leaders should care 

For enterprise organisations operating complex logistics, field services, retail concourses, or facility operations, this capability closes the gap between operational databases and immediate action. 

Instead of relying on manual triage or delayed batch reporting, platform leaders can deploy automated operational loops that categorize incidents and dispatch resources in real time. The underlying data remains stored in Eventhouse for post-event auditing and continuous taxonomy refinement, delivering both instant responsiveness and long-term analytical value.  

Learn more on Fabric Updates Blog 

Power BI date picker slicer prevents misleading trends and reduces report overhead 

Source: Microsoft 

Report authors have long struggled with date filters that break over time or display misleading trends as new data streams in. The Public Preview of the new date picker slicer visual in Power BI solves this by introducing dynamic relative date ranges anchored to your actual dataset, significantly reducing ongoing report maintenance for enterprise BI teams. 

Automated dates that stay alive on dashboards 

Power BI now includes a native date picker slicer that dynamically anchors relative date windows (such as “last 24 full months”) directly to the maximum or minimum available date in your model, today’s date, or custom offsets. Unlike legacy slicers, these relative rules persist when visuals are pinned to dashboards, keeping tiles automatically refreshed as underlying data loads. Viewers retain full control to override defaults using a compact, overlay-style calendar UI without altering the baseline report configuration. 

Ending manual maintenance and broken trendlines 

For data platform leads, static or brittle date logic generates constant report maintenance and misleads decision-makers with incomplete partial-period drops. This feature enforces data integrity by automatically resolving partial periods using simple measures (“is not blank” filters), ensuring executive scorecards show true side-by-side performance. Crucially, because relative rules stay intact when visual tiles are pinned to dashboards, centralized BI teams no longer need to manually rebuild or adjust executive dashboards at the end of every reporting period. 

Faster reporting and friction-free analytics 

This update delivers a cleaner reporting footprint and better self-service capabilities across enterprise environments: 

• Reduced report maintenance: BI teams save hours previously spent manually adjusting static date filters or fielding user tickets about partial-month trend declines. 

• Optimal canvas real estate: The footprint compresses to a single line without sacrificing calendar, slider, or reset features, leaving more space for high-value metrics. 

• Frictionless executive exploration: Business stakeholders can toggle between automated defaults and ad-hoc ranges without permanently breaking published views. 

Learn more on Power BI Updates Blog