Building Data Movement Diagrams

A data movement diagram provides a high-level graphical view of the movement of your information, including data sources, replications, and ETL operations.

You can create the following types of data movement diagrams:

Replication diagram – lets you model the replication of data from source to remote databases via replication processes (see Replication Processes (DMM)). Replication processes contain publications and article definitions that define which data are replicated. Scripts can be generated for Replication Server and MobiLink (see Working with MobiLink) engines.

In the following example, data contained in the New York primary database is replicated by the Europe replication process into the Paris, Berlin, and Madrid remote databases:
Transformation diagram – lets you model ETL and EII transformations of data from input to output sources via transformation processes (see Transformation Processes (DMM)). The transformation is specified in detail in one or more data transformation diagrams (see Data Transformation Diagram Basics) which can be linked together in transformation control flow diagrams (see Transformation Control Flow Diagram Basics).

In the following example, multiple input sources are transformed by the Data Fusion and Reorganization transformation process, and then loaded to the Giant Corp data warehouse:

Data Movement Diagram Objects
PowerDesigner supports all the objects necessary to build data movement diagrams.
Creating a Data Movement Diagram
You can create a data movement diagram in an existing DMM in any of the following ways:
Replication Processes (DMM)
A replication process is an instance of a replication engine that copies data from one or more source databases to one or more remote databases or other replication processes.
Transformation Processes (DMM)
A transformation process is an instance of a data transformation engine that extracts data from input sources, transforms it, and loads it to output sources. Input and output sources can be databases, flat files, XML documents, or business processes.
Databases (DMM)
A database can serve as an input to or output from a replication process or a transformation process. The structure of the database is modeled in one or more Physical Data Models (PDM) that can, in turn, be linked to a live database.
XML Documents (DMM)
An XML document can serve as an input to or output from a transformation process. The structure of the XML document is modeled in an XML model (XSM) which is associated with the XML document.
Business Processes (DMM)
A business process can serve as an input to a transformation process. The structure of the business process is modeled in a business process model (BPM) which is associated with the business process.
Flat Files (DMM)
A flat file can serve as an input to or output from a transformation process. A flat file is a text file containing data in which each line holds one record.
Servers (DMM)
A server is a network device to which a database, a replication process or a transformation process is deployed. You should assign each of these objects to a server to ensure correct script generation.
Publications (DMM)
A publication specifies the data to replicate and enables remote databases to subscribe to them as a group. You must specify a data connection between the source database and the replication process before you can create a publication. Publications have no symbol in the diagram but are listed on the Publications tab of a replication process property sheet.
Articles (DMM)
An article is the basic unit of replication and contains either a table or a view to replicate. Articles are gathered together into a publication to be replicated by the replication process. Articles have no symbol in the diagram, but are listed on the Articles tab of a publication property sheet.
Columns (DMM)
An article column contains a table or view column to replicate. Article columns belong to articles which are gathered together into publications to be replicated by the replication process. Article columns have no symbol in the diagram, but are listed on the Columns tab of an article property sheet.
Procedures (DMM)
A procedure contains a stored procedure to replicate from a source to a remote database. Procedures have no symbol in the diagram, but are listed on the Procedures tab of a publication property sheet.
Subscriptions (DMM)
A subscription specifies where a publication must be replicated. Subscriptions can include Where clauses to filter the data to be replicated to the remote database. Subscriptions have no symbol in the diagram, but are listed on the Subscriptions tab of a replication process property sheet.
Users (DMM)
A user is a person or a group who is allowed to log onto the replication process, and act as a replication system administrator. Users have no symbol in the diagram, but are listed on the Users tab of a replication process property sheet.
Event Scripts (DMM)
An event script is a stored procedure that is run on a replication process (for global actions) or an article (for specific actions) when the associated event occurs. The event scripts available to your specific replication process are specified in the extension file attached to your model. For example, you can create a global event script which handles connection errors and that will be executed whenever the replication process encounters a SQL error.
Data Connections (DMM)
A data connection sends data between a database or other data store and a replication process or transformation process.
Data Connection Groups (DMM)
A data connection group allows you to specify a set of data connections, in which one active database is supported by any number of backups. In case the active database fails for any reason, RepServer can switch to a backup database, and resume operations.
Process Connections (DMM)
A process connection sends data between two replication processes to let you model a data replication system with more than one replication process.
Migrating Deprecated Model Container Objects into a Project
Until v15.0 you could use an information liquidity model (the former name of the data movement model) to display groups of models and the generation and mapping links between them. Now PowerDesigner projects are used to organize your models and project diagrams provide improved visibility for the interconnections between your models.