One of the main reasons for using Analysis Services centers on performance with complex data retrieval. The design of the storage within Analysis Services, therefore, becomes very important when trying to achieve the query and processing performance expected. In order to understand storage design, this section first describes what modes of storage are available within Analysis Services. Next, you will look at the configuration of partitions. Last, you will learn how to design aggregations.
Analysis Services permits configuring dimensions and measure groups using the following storage modes: Multidimensional OLAP (MOLAP), Relational OLAP (ROLAP), and Hybrid OLAP (HOLAP):
Multidimensional OLAP: MOLAP storage mode is the most aggressive because it stores all the aggregations and a copy of the source data with the structure. Additionally, the structure stores the metadata required to understand the structure. The real benefit of this structure is query performance, as all information needed to respond to queries is available without having to access the source data. Periodic processing is required to update the data stored within the structure, and this processing can be either incremental or full. As a result, data latency is introduced with this storage mode. Also as a result of this structure, storage requirements become much more important due to the volume of information that the system requires.
Relational OLAP ...