A Case for Inverted Indices in In-Memory Databases

Download Now
Provided by: IARIA
Topic: Data Management
Format: PDF
Recent database research has focused on in-memory databases, which can be used in mixed workload scenarios, enabling OLTP and OLAP queries on the same database engine. These compressed column-oriented database systems use a differential store concept to enable fast inserting and require a merge process to compact the data periodically in a compressed main partition that is changed by the merge process only. This characteristic feature calls for a re-evaluation of the performance of inverted indices. In this paper, the authors present a use case for an inverted index in a column-oriented in-memory database system to reduce the total costs of query processing in a mixed workload environment
Download Now

Find By Topic