Project REAL: Data Lifecycle - Partitioning
Published: September 1, 2005 | Updated : January 5, 2007
Partitioning is a method of splitting up a large data set into smaller, more manageable chunks. This white paper addresses the partitioning of SQL Server tables and Analysis Services measure groups. The paper focuses on horizontal partitioning, which segments rows of data based on values in a partitioning column. This is a common strategy in large data warehousing implementations, primarily as a means of managing data, cube loading, and maintenance. There are also benefits when querying the data, because both SQL Server and Analysis Services provide a means of discerning which partitions are relevant to a query, assuming the query uses the partitioning column to filter the data.
Included in this document:
Introduction
Data Lifecycle Overview
Relational Partitioning
Cube Partitioning
ETL Implementation
References
Conclusion
Download
Project REAL Data Lifecycle Partitioning.doc
676 KB
Microsoft Word file
Updated : January 5, 2007