|
Support different Architecture Scenarios
Metadata tool should be able to support different metadata architectures. It should be able to support centralized as well as 'hub and spoke' architecture. An evolved organization may be adopting a single tool and yet allow different groups to develop their own repositories. A metadata tool should be able to:
- Assign a repository as a central/HUB or a spoke repository.
- Should be able to support:
- Single physical repository containing all metadata
- Hub and spoke repository with metadata being distributed across the Hub and Spoke.
- Truly federated hub and spoke, with metadata residing in spokes and Hub containing only common models, information about the metadata location and metadata exchange definitions.
- Store metadata in a centralized, shared location or in a distributed location from storage servers' perspective. You can have a single metadata repository, but stored in multiple servers.
- Dependent repositories should be able to automatically reuse foundation repository definitions.
Architecture Layers
A metadata tool should either have its own front-end, middle (processing layer) and database layer. It will allow it to be more flexible and open. It should also be able to work with other middle ware tools.
Web-enablement Architecture
- Product should be web-enabled and provide a user-friendly interface
- Product is fully administrable through browser.
Strategic Partnership
Vendor's strategic partnerships with following category of tools:
- CASE and Data Modeling tools
- ETL vendors
- ERP and conventional business application packages
- OLAP
- Reporting
- Data Quality Assurance and Monitoring Tools
- DBMS
- Data Integration Tools
|
