Architecture of ODI

All ODI components are divided into :-

  1. Design Time Components :- ODI Studio
  2. Run Time Components :- Agents
  3. A special components for monitoring based on Web Application :-  ODI Console

Components and Architecture of ODI
 ODI Architecture and Components (References:

ODI Studio : It provides a graphical navigation to manage ODI artifacts.

Further it is divided into Four parts:

  1.  Designer :  It provides declarative rules for data transformation and data integrity . All project development takes place in this component.It is used to retrieve metadata(in Model) , develop projects and release scenarios.  
  2.  Operator : Is used to monitor ODI jobs and sessions, basically  manages data integration process in production.  It  shows execution logs with error counts, the number of rows processed, execution statistics, the actual code that is executed, and so on. At design time, developers can also use the Operator module for troubleshooting purposes.
  3.  Topology : It is used to configure the infrastructure of ODI projects  i.e. data servers(source/target), physical schema(source/target), logical schema and Agent configuration.
  4.  Security :  It used to manages user profiles, roles and their privileges. Security can also assign access authorization to objects and features.

Agent : Agent is multi-threaded java programs, used to execute ODI scenarios. It retrieves the code from ODI repositories, connects to different source/target systems and executes the data integration process.

There are three types of agents in  ODI 12c.

1. Standalone Agents : It can be installed on source and target and requires a JVM.

2.  Co-located Standalone Agents : They can be installed on the source or target systems as well. They can be  managed using Oracle Enterprise Manager and must be configured with an Oracle WebLogic domain.

3. J2EE Agents :  They are deployed on Oracle WebLogic Server and  can be managed using Oracle Enterprise Manager.

All ODI components (design time, run time, ODI console) are connected to each other through a repository.

Repository :- It is a central component in ODI architecture,  basically it stores

  1. Information about Infrastructure (i.e Source/Target, source/target server name, source/target schema ).
  2. Information about Agents.
  3. Information about ODI users, roles and responsibilities.
  4. All metadata information of ODI project i.e. Projects,  Mappings, Procedures, Source Tables and Target Tables.
  5. Execution Logs.

There are two types of repository in ODI :

  • Master Repository
  • Work Repository

 Master Repository contains :

  • Security Manager :- Users, roles, profiles and there privileges(developers, testers etc).
  • Topology Manager :- Source/Target servers and their schema, Agents.

 Work Repository contains :

  • Data models: metadata, constraints, references, data lineage, etc.
  • Projects: interfaces, packages, procedures, folders, knowledge modules, etc.
  • Scenarios, scheduling information and logs.


Types of Work Repository:

  1. Development work repository : Used for developing all ODI projects and execution logs.
  2.  Execution work repository : Used for execution in production environment.

Note: Repository consists one or more Master repository, and each master repository can have one or more work repository. Each different work repository can process the work of different environment i.e. One for Development, one for Testing, one for Production etc.

Why multiple Master repository  : In corporate environment, to gain the isolation of production environment to rest of other information system, we can use one separate Master repository for the Production.


2 Replies to “Architecture of ODI”

Leave a Reply