Page tree
Skip to end of metadata
Go to start of metadata

BI Accelerator Index


A BI accelerator index is a redundant data store of a BI InfoCube on the BI accelerator server.


BI accelerator enables quick access to any data in the InfoCube with low administration effort and is especially useful for sophisticated scenarios with unpredictable query types, high volumes of data and a high frequency of queries.


BI Accelerator Index

A BI accelerator index contains all the data of a BI InfoCube in a compressed but not aggregated form. The BI accelerator index stores the data at the same level of granularity as the InfoCube.

It consists of several, possibly split indexes that correspond to the tables of the enhanced star schema and a "logical" index which, depending on the definition of the star schema, contains the metadata of the BI accelerator index.

BI Accelerator Server

The BI accelerator server is a TREX system as an installation of a BI accelerator engine. The data of the BI InfoCube is kept and processed entirely in the main memory of the BI accelerator server.

The BI accelerator engine is the part of the analytics engine that manages the BI accelerator index. This software allows the system to read data from the BI accelerator index, add data to the BI accelerator index, or change data. The BI accelerator optimizer is the part of the BI accelerator engine that ensures the best possible read access to a BI accelerator index.


Maintenance Processes for BI Accelerator Indexes

With the BIA index maintenance wizard you can create, activate, fill and delete BI accelerator indexes.

Like relational aggregates, a BI accelerator index is a redundant downstream data source that is used to improve query performance. For this reason, hierarchy and change run processes and processes for rolling up data are derived from aggregate maintenance.  

BI Accelerator Index as InfoProvider for Reporting

At query runtime, analytical engine functions such as aggregation, filtering, selection and some cell-based sorts are performed on the BI accelerator server.

  • No labels