Data Warehousing and Knowledge Discovery

Volume 6263 of the series Lecture Notes in Computer Science pp 117-129

Speeding Up Queries in Column Stores

A Case for Compression
  • Christian LemkeAffiliated withSAP AGIlmenau Univ. of Technology
  • , Kai-Uwe SattlerAffiliated withIlmenau Univ. of Technology
  • , Franz FaerberAffiliated withSAP AG
  • , Alexander ZeierAffiliated withHasso-Plattner-Institute

* Final gross prices may vary according to local VAT.

Get Access


BI accelerator solutions like the SAP NetWeaver database engine TREX achieve high performance when processing complex analytic queries in large data warehouses. They do so with a combination of column-oriented data organization, memory-based processing, and a scalable multiserver architecture. The use of data compression techniques further reduces both memory consumption and processing time. In this paper we study query operators like scan and aggregation on compressed data structures implemented in TREX.