Since 1992, Edgewater Technology has been providing its clients with the tools and services needed to begin their Business Intelligence (BI) implementations and leverage their existing technology investments with either an enterprise data warehouse strategy or a smaller process-specific data mart. Data warehouses and data marts are intended to solve the problem of disparate data sources and the incompatible goals of transaction processing and BI applications.
What is a Data Warehouse?
The purpose of a data warehouse is to provide a central repository where data from one or more transactional systems can be consolidated into a single integrated and consistent data source. The data warehouse is designed to optimize report generation and data mining on large numbers of database records. Because the intent of the data warehouse is to be read-only, data duplication is allowed in the warehouse in order to support reporting needs.
What is a Data Mart?
Data marts serve more focused and immediate reporting needs for a subset of business users. Each data mart is designed to most efficiently satisfy the needs of the intended user group. It’s all about slicing up a data warehouse or large data source into manageable data marts. The large data source remains in the background and these special-purpose data marts give faster response to targeted questions.
Over the last several years, traditional database vendors have been adapting and tuning their products to support data warehouse and data mart applications to meet the growing demand for BI applications. The leading databases including Microsoft SQL Server, Oracle, and IBM DB2, offer data warehouse and data mart capabilities including pre-built industry specific data marts for a wide range of vertical markets. Edgewater has extensive experience in all of these database systems and their data warehousing functionality.
On-Line Analytical Processing (OLAP)
BI solutions are not simply about building a data warehouse. The value is the ability to analyze data to provide insight needed to make decisions. OLAP is a key enabler of BI that provides sophisticated analysis capabilities far superior to traditional query and reporting.
Traditional query and reporting tools often limit the range of analysis since the query logic must be determined upfront before a user has access to the reporting tool. Given only the reporting tool and the complex transactional schema, there is too great a learning curve for the average business user to overcome to create the desired report. Even if this issue could be addressed sufficiently, the query that is developed may have performance implications that would limit the timeliness of the report delivery and potentially impact the response time of others using the system.
OLAP provides far greater capabilities, while also hiding the complexity of the underlying query language. It addresses the need to support a wider range of queries than traditional reporting tools. OLAP allows users to actively explore data by formulating queries “on the fly”, providing the user a rich set of flexible queries.
To achieve this flexibility while maintaining performance, OLAP utilizes multi-dimensional analysis. In this approach, data is conceptually viewed as a cube or other multi-dimensional object. The cube consists of quantitative values, known as measures, and descriptive categories, known as dimensions. The multi-dimensional approach stores data similar to the way organizations manage data along dimensions such as product lines, customers, geographic areas, and time. The intuitive cube concept allows users to easily formulate queries along hierarchies of dimensions to generate subsets along any “slice” of the cube and drill down to more detailed levels or roll up to higher levels.
Edgewater is a proven provider of data conversion, cleansing, and consolidation services. Let us assist you to integrate your valuable data into a data warehouse that can provide added insight and actionable BI to give you the competitive advantage.