A Multidimensional OLAP Engine Implementation in Key-Value Database Systems

摘要

This paper tries to explore the capabilities of MapReduce-like execution engines for multidimensional data analytics through implementing a Multidimensional Online Analytical Processing (MOLAP) engine with cube model on the Hadoop ecosystem. The cube storage module converts dimension members into binary keys and leverages a novel distributed database to provide efficient storage for huge cuboids. The bit encoding sparse index is used to compress the cube data and the dimension bit encoding key with maximum members is used to achieve cube data sharding and distributed aggregation computing. We discuss how to match the star schema with cube model databases, and how to use resilient distributed data-sets for MDX-like queries executing on key-value systems. Finally, some queries of TPC-DS benchmark are adopted to validate the prototype implementation of the MOLAP engine. The results indicate that designed scenarios based on TPC-DS are suitable for various big data analytics operation benchmarking.