Repositories
A repository is a repository of master and reference data. You can store different types of records in one repository or create more than one repository. A repository allows you to manage records, establish relationships between records, search data, and integrate data with other enterprise systems.
A repository defines the structure of your master data. It is often referred to as metadata. The structure of the repository is defined by your company and is very specific to your business requirements. It consists of:
- Attribute definitions and the arrangement of attributes in groups.
Attributes define the structure of a repository or a synchronization format. For more information on attributes, refer to Attributes.
Attributes can also be arranged in groups. For more information on attribute groups, refer to Attribute Groups.
- Classification schemes for records.
When browsing the repository, classification schemes allow you to view data hierarchically and to drill-down to a specific classification. You can define custom classification schemes as well, once the repository is created. Refer to Classification Schemes for details. The classification description can be loaded using Filewatcher. Refer to TIBCO MDM Customization for details.
Predefined classification schemes are used to categorize records based on schemes already used within your company.
- For the MDM edition, no predefined classification schemes are supplied.
- For the GDSN edition, the predefined classification schemes are UDEX and GPC. Predefined classification schemes are defined by standards bodies, such as GS1 with GPC.
For more information on classification schemes, refer to Classification Schemes.
- Input maps to import data from external sources.
Input Maps are used to import a large amount of data from external sources. They use data sources as primary input and allow you to map the data source attributes to repository attributes. You can combine more than one data source to map all or only some of the repository attributes. You can define an expression while mapping to transform the data.
For more information on input maps, refer to Input Maps.
- Output maps to extract data to synchronize data with other systems or partners.
Predefined output maps define a set of attributes required to publish data using the output map. When one or more such maps are selected, all attributes required to publish data using the map are automatically selected and added to the repository. If two or more output maps are selected, a union of all attributes required for all selected output maps is added. Custom output maps can be defined only after a repository is created.
For more information on output maps, refer to Output Maps.
- Rules for validating access control and the cleansing of data.