Deployment Architecture and Components

This page describes the standard deployment architecture for Oracle Utilities Analytics Warehouse (OUAW).

Product Overview

Oracle Utilities Analytics Warehouse has four basic components:

• Database
• ETL
• Dashboard
• Administration Tool

The following is a high-level diagram of OUAW's deployment architecture.

Database

The database component of OUAW consists of all required schemes and objects in the database. This section provides a high-level description of the necessary database schemes and their requirements.

Release Administration

This is a metadata schema generated at the time of the database component's installation. It is composed of the database objects that store the product's version number and the information regarding the component's installation.

Metadata

This is a metadata schema generated at the time of the database component's installation. It is composed of the database objects used to store and maintain the ETL configuration. Its layer contains information about the configuration details for each source system, ETL job configuration, scheduling information, execution logs, and others to facilitate the ETL process.

For details about the configurations in the metadata schema, see Metadata Model.

Journal

This schema is generated in the source application database at the time of integrating source with OUAW. It contains the database objects required for the data extraction process. This isolates all data warehouse specific objects from the main transactional system and has read-only access to the base application database objects.

Replication

This schema is a separate layer in the target data warehouse where data from different source systems will be stored. One replication schema is created for each source instance while integrating the source system with OUAW. The table structures of the replication are similar to the source's, but include the timestamps and the operation type columns to facilitate the ETL process.

The key benefits of this layer are:

• Its trickle feed replication process reduces the load on the source system.
• It permits to retain history.
• It facilitates the debugging or tracking back from target to replication.
• It simplifies the loading process by merging data from multiple tables together into a unified view that emulates that of the target fact or dimension structure.

Staging

This schema, generated at the time of installing OUAW, is a separate layer in the target data warehouse. Data will be transformed here before loading into the target schema.

Only one staging schema is created for OUAW, and at the time of deployment there will be no object into this schema. As this layer is to hold intermediate data, objects will be created at the time of loading target dimensions and facts. As data is not persistent in this layer and it will be required to analyze the ETL job failures, it is recommended that business users do not have access to it. A configurable data retention period is implemented to ensure that the volume of data retained is manageable and meaningful. The default retention period is seven days.

Target

This schema, generated at the time of installing OUAW, is a separate layer in the target data warehouse where the final transformed data is stored. Business users will have access to this layer to run their appropriate analytical queries. The schema contains the fact and dimension tables that make up the dimensional model to support users' analytical queries. This is also called a presentation schema. This schema contains all target dimensions, facts, and materialized views based on which pre-built analytics are built.

ETL

This is a core component of Oracle Utilities Analytics Warehouse (OUAW) that consists of the ETL objects required for extraction of data from the source database into their respective replication schema. It converts and loads final transformed data into target star schemes. This component is built using Oracle fusion middleware technologies such as Oracle Data Integrator and Oracle GoldenGate.

OUAW is usually installed after the transactional systems have been operational for a few years, which implies that during implementation the data warehouse needs to be synchronized with the historical data that has been created in the source system. Extraction of initial source data is performed by the Oracle Database DBMS_DATAPUMP feature. OUAW uses Oracle GoldenGate to change data synchronization.

OUAW uses Oracle Data Integrator for the integration of data and then loads the final transformed data into the target schema. Oracle Data Integrator keeps in the database all objects required for the data integration into schemes, also called repositories, and there are two types of repositories in Oracle Data Integrator: Master and Work.

Master Repository

The Oracle Data Integrator master repository for OUAW stores the following:

• Connection Information: JDBC URLs, LDAP information, usernames, and passwords for source or target connections.
• Security Information: Usernames, passwords, ODI users privileges, and profiles information.
• Version Information: When a new version of an object is created in the ODI, that information is also saved in the Master repository.

Information contained in the master repository is maintained in ODI Studio with the topology and the security navigators. Data contained in the master repository is mostly static data and will be going through minimum changes whenever any topology or security information is added or updated. ODI components access the master repository whenever they need the topology and security information data stored in it.

Work Repository

The Oracle Data Integrator work repositories for OUAW stores the following:

• Project Information: Folders, packages, mappings, procedures, variables, sequences, functions, knowledge modules, and so on.
• Models: Data store structures, metadata, fields, columns, constraints, and so on.
• Load Plans and Scenarios: Scenarios to transform and load data into target entities in the data warehouse.
• Operations: Execution details, scheduling information, and logs.

The contents of a work repository are managed by using Designer and Operator navigators. They are also accessed by the agent at run time.

Apart from the master and work repository schema, Oracle Data Integrator requires a few more schemes in the target data warehouse database for the WebLogic domain and agent. These schemes are created by the repository creation utility (RCU) at the time of creating and configuring the WebLogic domain and the agent for ETL jobs. This is covered in more detail in the Install and Configure Oracle Utilities Analytics Warehouse section. For more information about ETL framework jobs and their usage, see ETL Project and Framework Jobs.

Dashboard

This is a visual component of OUAW that is built using Oracle Analytics Server. It contains pre-built analytics and visualizations for the utilities industry and populates data from the target data warehouse. The key benefits of this component are:

• It contains analytics for the most important business use cases.
• It supports user customization.
• Customers can extend this component and build their own dashboards and analytics for their specific needs.

Note:

Customers must follow a certain approach to extend this component for a smooth upgrade of OUAW. How to extend this component is explained in the Oracle Utilities Analytics Warehouse Developer Guide.

Administration Tool

This tool is an administration component of OUAW that lets you perform tasks such as modifying default metadata configuration, adding or modifying storage parameters, or configuring ETL jobs. See Administration for more information.