Data Integration and Management

I thought I would reference an excellent blog that I saw the other day entitled: "Comparing ODI and OWB Data Quality and Profiling Options" by Mark Rittman.

Mark outlines the unification of Oracle Data Integrator Enterprise Edition and the resulting deployment options customers will be able to select from for achieving consistency and quality for their data-centric architectures. Specificallly he describes how to discern between using Data Quality options from either Oracle Data Integrator or Oracle Warehouse Builder Enterprise ETL options.

He walks through several examples using each component for profiling data, how to apply cleansing for data flows, and extending business rules.

In his summary, he concludes that the two tools are useful for different scenarios. Oracle Data Integrator’s Data Quality options are easy to use and offer the flexibility for both non-Oracle as well as Oracle sources, whereas Oracle Warehouse Data Quality options are uniquely fitted to Oracle targets, which offer nice options for auto-generating corrections and suggest business rules the auto-detected metadata.

For more information on Data Quality using Oracle Data Integrator, take a look at Oracle Data Quality on OTN for more detail. You can download the same tutorials that were used in this article from Mark Rittman.

I never liked the term EII much to describe federated data services such as Oracle Data Service Integrator, so when I read the blog entry “Whatever happened to EII?” by Rob Karel and James Kobielus from Forrester, I felt confirmed in my thoughts.

I agree with the blog in that federation aka EII should not live in isolation; it is best included in both SOA and Data Integration strategic dimensions, as it is one of multiple DI tools to prepare raw enterprise data for consumption, besides data consolidation/replication, as well as data quality and master data management. The other dimension is its placement in a SOA landscape, as it is one of several data service providers from the disciplines of data integration (Data access, data federation, data movement, master data, data quality) to serve the business services in your enterprise architecture.

Just like an ESB or a business process manager are more than EAI, federated data services are more than EII.

In that sense data federation is not a monolithic answer to a problem, but rather as a piece in the puzzle for full data integration.