Professional Microsoft® SQL Server® Analysis Services 2008 with MDX

21.1. Proactive Caching

Traditionally, OLAP refers to fast access of aggregate or summarized data, with the source data retrieved from a relational data warehouse and stored in a storage format called MOLAP (Multidimensional OLAP). Relational databases are nonetheless really helpful for detail-level data and also helpful for reporting. You learned that MOLAP storage is optimal and provides the best performance, whereas ROLAP storage provides immediate access to the latest data but does not have the same performance as that of MOLAP storage. Proactive caching is an important feature in SSAS 2008 because it aids the UDM in delivering the best of relational and OLAP worlds, most importantly real-time data access with near MOLAP performance. Proactive caching helps UDM to achieve real-time data by providing controls that help in data propagation from the source data to the UDM, which is then available for users' queries. When appropriate proactive caching properties are set, Analysis Services starts building a new MOLAP cache when data in the underlying data source changes. While the cache is being built, Analysis Services serves users from the existing MOLAP cache. As soon as the new MOLAP cache is built, users are served from the new MOLAP cache and start seeing the new data. You will see this concept in action throughout this chapter.

As mentioned earlier, the UDM merges the relational and OLAP worlds. We consider proactive caching to be a management feature that helps the administrator ...