ONE Desktop ONE DQG ONE MDM ONE RDM ONE Runtime Server
latest (15.1.0) 14.5.x 13.9.x 13.8.x 13.7.x 13.6.x 13.5.0 13.4.x
User Community Service Desk Downloads
If you can't find the product or version you're looking for, visit support.ataccama.com/downloads

Impact of Changes to the MDM Project

Some of the model-building actions described in MDM Project configuration have impact on other parts of the solution. This chapter explains what impact particular actions (or sub-actions) can have. Deployment and Change Management aspect is not taken into the account.

New Connected System

Procedure

Add a new system as described in Connecting a System.

  • If the new system is connected for batch processing, add a batch load operation as described in Full Load Operation.

  • If the new system is connected for online processing, connecting a system is enough.

Impact

Logical Model

It might be necessary to extend the Logical Model because:

Reference Data Mastering

Transformations

  • Cleanse

    • Cleansing might require some specific rules or stream for a newly connected system.

  • Match

    • Unification rules might be modified or extended.

    • Selection rules (center or survivor) setting up priorities might need to be redefined.

    • Unification structure might require some extension.

  • Merge

    • Selection rules setting up priorities might need to be redefined.

Export Operation

  • Some new export or system-oriented export might be required - see Input and Output Interfaces > Adding an Export Operation section in the documentation.

  • Newly processed data might be required in existing export operations - adjust the current setting.

  • Newly processed data might not be required in existing export operations - adjust the current setting.

Event Handler & Publisher

  • Data changes might be required to be newly published - see Adding a Data Event Handler and Publisher.

  • Data changes of a new connected system might be available and published in existing publishers - adjust the current setting.

  • Data changes of a new connected system might not be available and published in existing publishers - adjust the current setting.

Native Services

  • Some requests and responses will be enriched by a new origin and/or source system definition.

New Instance Entity

It is usually necessary to add a new instance entity or dictionary after adding a new connected system or modifying an existing one.

Procedure

  1. Add and configure an entity on the Instance Layer as described in Creating an Instance Layer Model.

  2. Complete related Transformations (cleanse plan).

New Relationship on the Instance Layer definition below is usually a part of this use case.

Impact

Connected Systems

  • Some of existing Connected Systems (Model and Source Mapping) might need to be changed.

  • Consequently Batch load operation might need to be changed.

MDM Web App

Reference Data

Export operation

  • New export operation might be required - see Input and Output Interfaces > Adding an Export Operation section in the documentation.

  • Newly processed instance entity might be required in existing export operations - adjust the current setting.

  • Newly processed instance entity might not be required in existing export operations - adjust the current setting.

Event Handler & Publisher

  • Data changes might be required to be newly published - see Adding a Data Event Handler and Publisher.

  • Data changes of the new instance entity might be available and published in an existing publisher - adjust the current setting.

  • Data changes of the new instance entity might not be available and published in an existing publisher - adjust the current setting.

Native Services

  • New services might be required - see Native Services.

  • Entity-based services will be added for a new instance entity, for example, search, list, getById.

  • Some of the WSDLs will be changed (for example RW services, Traversal, Identify).

  • Some responses will be enriched by a new instance entity definition (for example, get statistics and model).

Enabling Matching on an Instance Entity

The matching feature is a prerequisite for creating a new master entity.

Procedure

  1. Enable Matching on the Instance Layer as described in Creating an Instance Layer Model, step 6.

  2. Complete related Transformations (the matching plan).

Impact

Logical Model

  • It might be necessary to add a Copy Columns definition on relationship on Instance Layer - see Creating an Instance Layer Model, section Adding a Relationship between Instance Entities.

Native Services

  • Some of the WSDLs will be changed (for example, RW services, Traversal, Identify).

  • Some responses will be enriched by a new column definition (for example, get model).

  • Identify service might need to be configured - see Native Services.

New Master Entity

A new master entity is usually added because mastering is required. To add a master entity, an existing related instance entity on the Instance Layer with enabled matching is required. If there is such related instance entity, see the use cases in New Instance Entity and Enabling Matching on an Instance Entity. It is also possible to add a (virtual) instance entity to access data from the Instance Layer.

Procedure

  1. Create a master entity on the Master Layer - see Creating a Master Data Layer Model.

  2. Complete related Transformations (the merging plan).

New Relationship on the Master Layer definition below is usually a part of this use case.

Impact

MDM Web App

  • It might be necessary to display the new master entity and relate it to surrounding entities in the MDM Web App.

Reference Data

Export operation

  • New export operation might be required - see Input and Output Interfaces > Adding an Export Operation section in the documentation.

  • Newly processed master entity might be required in an existing export operations - adjust the current setting.

  • Newly processed master entity might not be required in an existing export operations - adjust the current setting.

Event Handler & Publisher

  • Data changes might be required to be newly published - see Adding a Data Event Handler and Publisher.

  • Data changes of the new master entity might be available and published in an existing publisher - adjust the current setting.

  • Data changes of the new master entity might not be available and published in an existing publisher - adjust the current setting.

Native Services

  • New services might be required - see Native Services.

  • Entity-based services will be added for a new instance entity, for example, search, list, getById.

  • Some of the WSDLs will be changed (for example, RW services, Traversal, Identify).

  • Some responses will be enriched by a new instance entity definition (for example, get statistics and model).

New Relationship on the Instance Layer

A relationship to be created between two existing instance or dictionary entities, where the parent entity has at least one attribute set up as a primary key and the child entity has at least one attribute set up as a foreign key. Both single-attribute keys and compound keys are supported.

Procedure

  1. Add and configure a relationship between instance (and dictionary) entities on the Instance Layer as described in Creating an Instance Layer Model, section Adding a Relationship between Instance Entities.

Impact

Load plan

  • The mapping logic might need to be changed to fill relationship attributes by appropriate data - adjust the current setting.

MDM Web App

Transformation plans

  • Based on the CopyColumn definition, new columns are added into one and/or both entities.

Native Services

  • Some of the WSDLs will be changed (for example, RW services, Identify).

New Relationship on the Master Layer

A relationship to be created between two existing master or (virtual) instance entities, where the parent entity has one attribute set up as a primary key (must be the id attribute) and the child entity has one attribute set up as a foreign key (user defined, usually <parent_entity_abbreviation>_id). Only single-attribute keys are supported.

Procedure

  1. Add and configure a relationship on the Master Layer as described in Creating an Instance Layer Model, section Adding a Relationship between Instance Entities (the procedure is similar on the Master Layer).

Impact

Logical Model

  • (Virtual) instance entity might need to be added to be able to create a relationship between the master and mentioned (virtual) instance entity - see Creating a Master Data Layer Model.

  • Enabled Matching on a related instance entity might be required to be able to create a relationship between both two master entities and master and (virtual) instance entity - see Enabling Matching on an Instance Entity above.

MDM Web App

  • Link between entities might be added to be shown in the MDM Web App.

Native Services

  • Some of the WSDLs will be changed (for example, Traversal).

New Reference Data

New reference data definition might be required when creating a new instance entity (see New Instance Entity) to translate source dependent values to master codes and when creating a new master entity (see New Master Entity) to provide name and description for each master code.

Procedure

  1. Add and configure Reference Data as described in Working with Reference Data.

Impact

Logical Model

  • Use the already defined Reference Data for an appropriate attribute - see Using Reference Data below.

Native Services

  • Entity based services might be added for a new reference data entity, for example search, list, getById.

  • Some of the WSDLs might be changed when reference data dictionary defined (RW services, Traversal).

Using Reference Data

The already defined Reference Data can be used on both the Instance and Master Layer, but in different contexts.

Procedure

  1. Use a dictionary for an attribute on the Instance layer (translation context) and/or use a dictionary on the Master Layer (master-data-providing context) as described in Creating an Instance Layer Model and Creating a Master Data Layer Model respectively.

Impact

Load operation

  • Reference data dictionary might generate a Lookup step into the plan - plug it into the existing data flow.

Transformation

  • Cleansing plan

    • Reference data dictionary might generate a Lookup step into the plan - plug it into an existing data flow and map the dic_ output to the appropriate std_ and/or cio_ value

    • A dic_ attribute might need to be mapped to the appropriate std_ and/or cio_ value.

Native Services

  • Some of the WSDLs will be changed by using a master dictionary on the Master Layer (Traversal).

Was this page useful?