The Promise of Data Mastering at Scale

What you'll learn

Master data management (MDM) software turned 15 years old this year. Originally launched in 2004 by SAP, master data management systems aimed to help resolve the data unification problem by creating a central source of standardized references to customers, products, employees, suppliers, physical assets and other data across their many IT systems. MDM is valuable, but it's also slow, labor intensive, and costly. As the scale of MDM projects increases to millions of entities and hundreds or thousands of data sources, the traditional methods often fail. Mike Stonebraker will share his view on how MDM technology and MDM organizations must change to fulfill the promise of MDM at scale. He'll discuss how MDM at scale requires machine learning (ML) models, how MDM ML requires humans-in-the-loop, how MDM at scale often requires real-time model update, and how future data mastering innovations will come to market as ML models. You'll learn how MDM technology and MDM organizations must change to fulfill the promise of MDM at enterprise scale.

Table of contents