Business intelligence architecture

Business Intelligence Architecture

Need to define business intelligence architecture for an information management project and want an overview of key architectural components?

What is a data warehouse?

A data warehouse is the foundation for information management and involves:

  • Extracting data from operational systems;
  • Moving it into data warehouse structures;
  • Reorganizing and structuring the data for analysis purposes; and
  • Moving it into reporting structures called data marts.

Business Intelligence Architecture

Data warehouses are usually intended to store very large amounts of data. Key design considerations include data load strategies and daily operations.

What architecture components are involved with warehouse design?

Data warehouses are comprised of:

  • System processes;
  • Process architecture; and
  • Hardware architecture.
What are system processes?

System processes include:

  • Extract and load processes;
  • Transformation and data cleansing processes;
  • Data backup and archiving processes; and
  • Reporting management processes.
What is process architecture?

Process architecture includes the process required to handle and manage daily operations. They include:

  • Extract and load management process, which includes all of the software and utilities required to:
    • Extract source system data and move it to the warehouse environment;
    • Fast load data into a staging area where it can be subsequently manipulated; and
    • Complete basic transformation to ensure that non-essential data is eliminated and other data is converted to appropriate data types.
  • Warehouse manager, which includes:
  • Software and utilities required to perform management functions such as:
    • Ensuring referential integrity;
    • Transforming staging area data into formats suitable for loading into the data warehouse;
    • Dropping indexes, if needed to optimize load performance, and recreating them when the load is complete;
    • Creating de-normalized data structures if needed for reporting performance;
    • Generating and maintaining data aggregations; and
    • Moving data to archival storage.
  • Reconciliation routines, which include all of the routines, needed to ensure that data is reconciled with itself;
  • Transforming data into a star schema, which involves:
    • Creating fact tables;
    • Creating dimensions and handling slowly changing dimensions; and
    • Summarizing data if required for reporting performance.
What is hardware architecture?

Hardware architecture includes the components needed to support the entire data warehousing initiative and includes:

  • Server hardware; which require:
    • Large configurations to support very large data volumes; and
    • Server management tools to help database administrators and systems administrators manage things like:
      • Disk space utilization;
      • Locked processes;
      • Hardware failure; and
      • Excessive CPU usage.
  • Network hardware, which:

    • Needs to be robust enough to handle business source system extracts, data warehouse and data mart loading and business intelligence reporting, and
    • Requires network management tools to help network administrators monitor performance
Roles and Responsibilities

A data architect usually works with systems and network administrators to define the overall architecture. This is then documented in an architecture document, which includes the data warehouse architecture and design specifications.

Business intelligence architecture checklist

A data architect will need to consider most of the following while establishing business intelligence architecture:

Capacity planning;

Conceptual database design;

Data backup storage;

Data base program;

Data bases;

Data compression;

Data information management;

Data management;

Data management information;

Data management software;

Data management solutions;

Data management system warehouse;

Data management tools;

Data management warehouse;

Data marts;

Data model;

Data storage strategy;

Data systems;

Data warehouse architectures;

Data warehouse design solution;

Data warehouse program;

Data warehouse report;

Data warehouse systems;

Database and data warehouse;

Database architecture;

Database design;

Database design software;

Database development;

Database management;

Database management system;

Database partitioning;

Database schema;

Database servers;

Database software;

OLTP database versus data warehouse;

Design specifications;

Development environment;

Dimensional model;

Enterprise architecture framework;

Enterprise architecture planning;

Enterprise architecture tools;

Enterprise business intelligence solutions;

Enterprise data warehouse;

Enterprise data warehouse solutions;

ERP and business intelligence;

Estimates for creating data warehouse;

Extract transform load;

Hardware architecture;

Information architecture;

Information management architecture;

Information management design;

Information management technology;

Information management tools;

Information strategy;

Logical database design;

Operational data store (ODS);

Offsite data storage;

Physical data model;

Relational database;

Software architecture;

Starflake design;

Storage management information;

Storage systems;

System architecture;

Technology strategy; and

Warehouse data management,


Data warehousing is the foundation for information management and involves extracting data from operational systems, moving it into data warehouse structures, reorganizing and structuring the data for analysis purposes, and moving it into reporting structures called data marts.

A data warehouse architect usually has overall responsibility for defining and sizing the correct data warehouse and business intelligence architecture.

Similar articles: