Enterprise data integration includes a variety of unique -- yet related -- technologies known by a slew of acronyms. The terms defined below help clarify the broad practice of data integration and introduce some of the methods that support an enterprise data integration project.
-
Enterprise data integration (EDI) (not to be confused with electronic data interchange, also commonly known as EDI) is the broad set of technologies that help combine, or integrate, two or more data sets. EDI typically involves merging data from disparate systems and delivering that combined data in such a way that it can be manipulated and analyzed to support information management activities, such as business intelligence or reporting. Read about the costs and benefits of data integration software versus hand coding.
-
Extract, transform, load (ETL) refers to three separate database functions used to collect data for analysis, reports or other uses. First, the requested data is read and extracted from a source database. Next, it is it transformed using business rules into the desired state. Finally, the transformed data is loaded into a target database or data warehouse where is can be accessed by other applications or users. Learn about the pros and cons of custom code versus the use of an ETL tool or read a study that reports ETL tools are transforming.
-
Enterprise application integration (EAI) enables data propagation and business process execution across various applications to support complex operational business functions such as fulfilling a customer order. It aims to allow changes in one application to propagate across all systems across an enterprise as if they were one global application – its goal is to ultimately deliver one consistent version of the data. Read about EAI's role in SOA and decentralized IT systems.
-
Enterprise information integration (EII) is a strategy for making enterprise-wide data operative and manageable. This integration technology uses query federation to allow information gathered from various sources to be accessible on demand and in real-time, without storing it on a disk, a process often referred to as "on the fly" transformation. EII enables a "virtual" data warehouse, repackaging the data collected across systems into an easy-to-view representation. As Wayne Kernochan, of Infostructure Associates, writes in this article, "EII provides a "database veneer" or service that allows administrators, developers, and end-users to treat a broad array of data sources as if they were one large database or data service." Learn how ING Worldwide uses EII or listen to a podcast about data integration, which includes a discussion of EII technology.
-
Service oriented architecture (SOA) is an architecture in which software components can be exposed as services on the network, enabling re-use by different applications and business processes. In an SOA, data and applications are loosely coupled, rather than tightly integrated. Ultimately, SOA supports integration efforts since it frees data from individual applications, making it easier to share data and reuse common services across the enterprise. It provides a flexible architectural model that permits the enterprise to respond to business and market changes quickly and efficiently. Listen to a podcast about SOA and data integration, read a sample chapter on SOA and business rules management or find out why one expert thinks that SOA will fundamentally change data integration.
-
Master data management (MDM) is is a discipline involving people, process and technology, which helps enterprises define, link and maintain consistent definitions of "master" business objects such as customers, suppliers or products. To learn more, review seven master data management best practices, read how Gartner and Forrester suggest evaluating MDM technology or for skeptics, check out this article: Is master data management all hype?
- Customer data integration (CDI) is, as CDI expert and author Jill Dyche describes it, "the collection of processes, controls, automation and skills necessary to standardize and integrate customer data from different sources." It involves eliminating data redundancy, cleansing, normalizing and establishing one version of customer data that is consistent across the enterprise. Listen to a podcast about customer data integration with expert Jill Dyche or check out how analyst firm Gartner ranks customer data integration vendors.
This was first published in August 2010