| Master Data Management |
| The quality of data is the key factor in efficiently managing business activities and decision making processes in an organization. In most organizations, the data exists in OLTP systems, ERP systems, enterprise data warehouses (EDW)/ data marts and various business applications. This leads to difficulty in having accurate and consistent data in all the systems - "The Single Version of Truth". The Master Data represents the business critical data that supports the OLTP, EDW (decision support systems) and other applications, resulting in smooth operation of business activities end-to-end. The Master Data Management (MDM) uses applications and technologies to achieve the creation of accurate, high quality of data at the source and mainenance of data to be used across various processes (ETL), applications ( EAI, ESB, SOA, DW, BI etc.), and systems in an organization. |
| Master Data Management Processes |
| Data Profiling - Data has to be constantly profiled in or near realtime to maintain quality, accuracy, prevent deviation from master data, and cleansing as required thereby keeping the data synchronized with the master data. |
| Data Consolidation - The master data has to be conslidated into a central repository for use across all systems and applications. |
| Data Governance - The master data governance defines business rules, data definitions, standards, privileges, access rights, quality, workflow, management and compliance aspects. |
| Data Sharing - The master data has to be shared across the organization and be the source for all applications, resulting in being the "Single Version of Truth". |
| MDM Architecture |
| 1. Local MDM - Implementation of MDM for a single business unit |
| 2. Hub MDM - Implementation of MDM using a MDM hub that serves all applications (OLTP, ERP, EDW/BI etc.) of the entire orgnization |
| 3. Federated database MDM - Implementation of MDM for ERP and centralized IT systems |
| 4. Hybrid MDM - Implementation of MDM where a main or central MDM hub controls the department/application MDM hubs. This architecture is used in IT environments. |
| Reference |
| 1. Master Data Management, David Butler, An Oracle White Paper [018876], September 2011 |
| 2. Essential Elements of a Master Data Management Architecture, Joseph M. Bugajski, Burton Group, The Fourth MIT Information Quality Industry Symposium, July 14-16, 2010 |
| 3. Master Data Management (MDM) Hub Architecture, Roger Wolter, Microsoft Corporation, April 2007 |