Building Data Integration Solutions

Chapter 4. Models, Architectures, Methods, and Patterns

Unfortunately, there is not a standard, consistently used lexicon within data integration and literature involving integration topics. Some organizations and individuals will use the term method to indicate the implementation of certain types of data-related processes, while other organizations and individuals may use the term technique or pattern to indicate the same processes. But depending on who you ask, a method is not necessarily the same thing as a technique or pattern, and because there are so many viable pieces to the data integration puzzle, the inconsistent use of terms can be very confusing.

Note

What I’ve learned from doing a rather extensive, albeit informal, survey of semantics within media on the topic of data integration is that there are many contextually overlapping meanings of the terms model, architecture, method, pattern, technique, strategy, and blueprint—among others. Also, there are subtle semantic differences between terms with similar but distinct descriptors, such as data architecture and data integration architecture. The discrepancy is not always readily apparent and is largely dependent on an individual’s interpretation. Data architecture is broad and encompasses the overall management of data assets. Data integration architecture is narrower and focuses on the movement and transformation of data. Table 4-1 tabulates the differences.

Table 4-1. Differences between data architecture and data ...