Sunday, August 10, 2014

Common Data model

List different back ends
Analyze their data model
Make a repository
Have change management in place for on going changes in CDM


Field
Name
Type
Length
Annotation
Default value
Enumeration
Structural constraint
Dependent Schema
Dependent services
Field with equivalent definition



Considerations:
Technical quality
standards organizations
business Alignment

Best practices:

organization and Annotations
default values
Constraints
simple Constraints
Datatypes
Enumerations
structural constraints: Structural constraints typically should be adopted from the starting schema without modification
semantic consistency rules

Service version
Service schema version
Complex type version
Simple type version

Complex integration with many integration(backend and frontend), performance is not key: Source data model->CDM source->CDM target->Target data model

Approach 1:
WSDL "contains" Service request/response type
Service request/response type "defined in" Service schema
Service schema "defines" simple types to base business type
Base business is "defined" based on base simple types
Base simple type, base business type are "defined" in simple type schema

Approach 2:
WSDL "contains" Service request/response type
Service request/response type "defined in" common complex schema
Service schema "defines" simple types to base business type
Base business is "defined" based on base simple types
Base simple type, base business type are "defined" in simple type schema

Challenges:
Developer availability and training issues
Organization adaptation




Maintenance :
A unified metadata format for all services enables more complete analysis and minimizes developer availability and training issues.
Reports to understand and review new schemas and models to be added to the landscape.
Difference analysis that highlight changes to schemas.
Impact analysis to understand the interrelationships between the various source models and their maps between the models.
Change management tools that allow analysts to accept/reject changes to the models in much the same way as editors use “Track Changes” in Microsoft® Word.
Interactive, easy to use testing tools applied in-line with the development process that use test data maintained in the exchange model create higher quality components, save significant time, and minimize the impact on the existing production environment


No comments:

Post a Comment