Online analytical processing (OLAP) is a data analysis technique used in business intelligence (BI) that involves using multidimensional databases for rapid reporting and trend analysis. OLAP cubes relate measures and dimensions, and OLAP databases are often smaller and faster than data warehouses. OLAP specialties include multidimensional, relational, and hybrid. BI also includes data mining, reporting, operational performance management, and predictive analytics. OLAP is used for ad hoc reporting and can be used in finance, operations, sales, and marketing departments for budgeting and forecasting. More sophisticated tools may be required for predictive analytics and business analytics.
Online analytical processing (OLAP) is a method of using multidimensional databases to support rapid reporting, often involving trend analysis. The primary query language for OLAP is called Multidimensional Expressions (MDX). Its name comes from the class of program known as online transactional processing (OLTP). Online analytical processing is a data analysis technique used in the field of business intelligence (BI).
BI involves using technology to analyze an organization’s internal processes and data to support decision making. When using online analytical processing for BI, historical data is often the subject of analysis, but BI can also include analysis of current and future states. Along with OLAP, other data management techniques that fall under the realm of BI include data mining, reporting, operational performance management, and predictive analytics.
Online analytical processing is often used for ad hoc reporting and typically generates reports in pivot or matrix format. Departments that can use OLAP include finance, operations, sales and marketing. Usage types can include budgeting and forecasting.
One of the defining features of online analytical processing is the OLAP cube. The concept of a cube relates the elements known as measures and dimensions, which describe the metadata of the various measures. Tables in the star or snowflake schema of a relational database can be the source of metadata. An example of a cube is using a company’s accounts receivable amount as the measure, with a due date as the dimension.
OLAP uses databases designed with multiple dimensions. These databases can be smaller than those needed for data warehousing capabilities often used for business intelligence. Compared to other types of analysis, less transaction details are generally required in online analytical processing. Not only are OLAP databases often smaller than data warehouses, but accessing OLAP databases is often faster than accessing relational databases.
There are various specialties of online transaction processing. Many of the most frequently used specialties include multidimensional, relational, and hybrid. Multidimensional OLAP stores data in multidimensional arrays, relational OLAP uses relational databases, and hybrid OLAP uses a combination of relational and specialized tables.
While online transactional processing is an important technique in BI, more sophisticated tools or improvements to OLAP may be required for organizations interested in predictive analytics and business analytics. Predictive analytics is often used to predict events such as customer buying behavior. Business performance data is usually the focus of business analytics.
Protect your devices with Threat Protection by NordVPN
