dataMinds News Round up –
April 2025
Author: Hugo Lu
Hugo explores how Azure Databricks and Microsoft Fabric together enable seamless integration of big data workflows. It highlights the role of Apache Iceberg in improving data analytics performance and scalability, sharing real-world use cases where organizations have leveraged Fabric and Databricks to unify their data lakes for real-time processing.
Author: Kevin Feit
Kevin zooms in on the innovative capabilities of the SQL Database in Microsoft Fabric, which now supports operational databases alongside analytical workloads. The article explains how automatic index optimization and native integration with Microsoft Purview sensitivity labels enhance governance and performance.
Author: Gina Shaw
Highlights from FabCon 2025, focusing on breakthrough announcements like OneLake Security and enhanced Fabric Data Agents. The article provides an overview of key sessions and workshops that showcased the future of Microsoft Fabric.
Author: Samu Niemelä
Recap the excitement from FabCon 2025 in Las Vegas at the MGM Grand. This article covers key highlights such as the unveiling of OneLake Security, developer experience enhancements, and vital networking sessions that set the stage for the next generation of Microsoft Fabric innovations.
Author: Sandeep Pawar
This article demonstrates how Fabric AI Functions can be used to convert unstructured PDF invoice data into structured, actionable information. It outlines a streamlined process where key invoice elements—such as invoice numbers, dates, vendor details, and total amounts—are automatically identified and extracted. The resulting structured data facilitates seamless integration into downstream systems for automated processing and analytics, dramatically reducing manual effort and improving data accuracy.
Author: Meagan Longoria
This article examines the inherent trade-offs associated with adopting low-code solutions for application development. It discusses how these platforms can significantly accelerate development cycles and empower non-technical users by simplifying the creation of business applications. However, the article also highlights potential drawbacks such as limitations in customization, risks of vendor lock-in, scalability challenges, and integration complexities with existing systems. Overall, it provides a balanced perspective to help organizations determine when low-code platforms may offer a strategic advantage—and when a traditional coding approach might be more suitable for long-term success.
Author: Eugene Meidinger
This article investigates the performance of the DAX aggregation functions SUM and SUMX, demonstrating through examples and benchmarks that, in many scenarios, they achieve nearly identical performance. Despite their conceptual differences—where SUM is a simple aggregation and SUMX iterates over a table expression—the underlying engine optimizations often level the playing field. The piece offers insights into when to use each function, helping practitioners make informed decisions in their data modeling and performance tuning strategies.
