Analyst Rick F. van der Lans recently released a Whitepaper, sponsored by Denodo, titled "Developing Bi-Modal Logical Data Warehouse Architecture Using Data Virtualization". The paper introduces a Logical Data Warehouse as an agile architecture that decouples data consumers and data stores, bringing data in together as a single logical database. There are two key points to summarize the paper. First, only a Logical Data Warehouse has the flexibility to meet modern business intelligence requirements. Second, Data Virtualization is required to make the Logical Data Warehouse successful.
However, the Whitepaper needed to make one final point to bring together the full picture of a Logical Data Warehouse (LDW). Agile Master Data Management is needed to govern your data, clean your data, and make sure the Logical Data Warehouse is successful.
Need for a Logical Data Warehouse
The whitepaper does an excellent job of laying out the need for a logical data warehouse architecture that can both deliver the robustness needed for standard forms of reporting as well as the ability to easily handle new sources of data like big data. Key features of Data Virtualization enable the existence of this architecture and position the TDW to handle evolving needs in the future. However, the whitepaper misses a potential key contributor to overall data architecture that can have a synergistic relationship to best set up your data architecture for the future - Evolutionary Master Data Management.
The whitepaper states that the rigidity of traditional BI systems leaves your business ill-equipped to keep pace with evolving BI requirements. These traditional systems will not be able to handle new data sources and the requirements of those sources alongside new analytical requirements. Yet, migration to new systems may be difficult as the existing workload of your solution cannot be compromised. Thus, van der Lans argues, you need new architectures for your BI systems that are capable of quickly handling your old and new forms of BI.
The answer, according to the whitepaper, is to implement an architecture that can support a bi-modal BI system capable of handling both old and new simultaneously. The structures of the LDW are hidden from data consumers, and those consumers only see relevant data. This de-coupling creates the flexibility needed for your architecture to be able to handle old BI requirements alongside new and evolving data sources and requirements.
MDM's Necessary Role
Now, your Logical Data Warehouse will effectively bring your data together and deliver it to your data consumers, but the architecture as described is missing an important piece - Master Data Management. Master Data Management is key to ensuring sound, dependable data in any architecture. The right, agile MDM solution will work with your Logical Data Warehouse to bring together critical data, clean it, and work with the LDW to deliver it to your consumers.
Denodo's Data Virtualization, van der Lans argues, is the best tool to achieve a LDW. However, the most efficient and flexible LDW architecture combines the power of Data Virtualization with an agile MDM tool. While DV gives you the power to dynamically deliver needed data to your consumers, MDM plays a key role in governing and cleaning the most necessary data to ensure that the right data reaches your consumers at the right time.
Not just any MDM, Agile MDM
Traditional MDM systems lack the flexibility needed to fully integrate with an evolving Logical Data Warehouse, but Semarchy's Evolutionary MDM™ has the agility to keep pace with your Logical Data Warehouse and fully unleash it's potential. If your MDM isn’t able to handle every-changing demands, it will hamper the LDW. Here DV and MDM work together to deliver a truly agile architecture that will evolve alongside your needs.