1.10 Data Interfaces

A Data Interface in EDQ is a template of a set of attributes representing a given entity, used to create processes that read from, or write to, interfaces rather than directly from or to sources or targets of data.

Data Interface Mappings are used to map source data to a Data Interface, or target data from a Data Interface. The actual mappings to be used (when a process that uses Data Interfaces is run) are defined in the job. This allows different jobs to use the same processes on different data; for example, to use a process in both batch mode (using a staged data mapping) or real-time (using a web service or JMS mapping).

It is also possible to specify which mapping(s) to use when running a standalone process that uses Data Interfaces. See Process Execution Preferences for further details.

Note:

Data Interfaces replace the concept of 'Views' which existed in Oracle Enterprise Data Quality versions 8.1 and earlier. Views were simpler forms of Data Interfaces that could only be used for reading data into processes, and which only supported a single mapping of source data to the View at any one time. If importing a View from a package file created using an older version of EDQ, the View should be dragged onto the Data Interfaces node. This will import the View attributes as a Data Interface. If importing a whole project that uses Views, the Views will be migrated to Data Interface, and the mapping used to map data to the View will be added as a Data Interface mapping. Please note, however, that any jobs that previously read data from a View will need to be edited to specify which mapping to use for the Data Interface (even if there is only one).

Data Interfaces are often used to create process templates, which can be used by System Integrators and Consultants who may deal with many different clients with a similar type of data (such as Customer, Operational, Financial, and so on). The process templates are not defined against any specific source or target of data, so new data from new clients can be quickly mapped and run through processes created from the templates.

Data Interfaces also enable effective process reuse: they make it possible to subdivide a potentially complex process into a number of reusable processes, each performing a specific function. These processes can then be chained together as required when configuring a job.

For example, the Oracle Enterprise Data Quality Customer Data Services Pack includes processes for standardizing addresses, individuals and entities that are used in both batch and real-time matching services (with different match processes).