Moving To a New Data Warehouse Risks Reporting Continuity

There are a number of reasons to migrate a data warehouse.

  • One is cost savings. Many enterprises are finding that data warehouse appliances can significantly reduce data warehouse total cost of ownership.
  • Another is mergers and acquisitions. In this case, duplicate financial data warehouses need to be rationalized.
  • A third is standardization. Here an enterprise or government agency may want to rationalize various warehouses based on disparate warehouse technology platforms by moving to a standard platform.

Regardless of the reason for the migration, in every case the reporting and analysis supported by the migrating data warehouse must continue to run seamlessly.

Use Data Virtualization to Insulate Reporting Users during Data Warehouse Migrations

You can use Composite data virtualization to insulate reporting users from the impact of data warehouse migrations. Composite data virtualization removes reporting risk by inserting a virtual reporting layer between your warehouse and your reporting systems. Decoupling these systems enables the reporting to continue before, during and after the migration. The integration pattern shown below depicts the original state prior to migration.

Data

Original Data Warehouse Prior to Migration

In the integration pattern shown in the figure below, the Composite Data Virtualization Platform is implemented between the reports and the warehouse. It is used to build and host the reporting data virtualization layer, a new set of views and data services that the reporting systems can use to query their data. While this requires the extra effort to create these views, this investment not only reduces immediate migration risk, it pays off by providing long run flexibility.

You must also modify the reports to query the virtualization layer, rather than the warehouse directly.  But the reporting queries would need to be rewritten anyway to query from the new data warehouse. The benefit of this approach is that it allows for a controlled migration of reports in advance of the data warehouse migration.

Data

Virtual Reporting Layer Added to Decouple Reports from Original Warehouse


The integration pattern shown in the figure below shows the next step in the process where the new warehouse and supporting ETL are brought on line and the old warehouse is retired.

Composite data virtualization insulates your reporting users at this step by enabling a controlled migration of the views in the reporting layer. Each existing view can be cloned, modified to point at the new warehouse, and tested before the actual cutover, thereby insulating the reporting users from undo risk. Further, the virtual reporting layer is easily extensible for adding more sources or supporting new reporting solutions.

Data

Virtual Reporting Layer Serving Data from New Data Warehouse

Selected Examples

  • Data Warehouse Appliance Migration – To reduce data warehousing total cost of ownership by migrating to a data warehouse appliance, this large technology company used Composite data virtualization to decouple their reporting from their data warehouse. Data warehousing cost reductions were achieved while reporting was successfully migrated without interruption.