A cache-sensitive hash indexing structure for main memory database

  • Authors:

  • Xiaoqing Niu;Xiaojia Jin;Jing Han;Haihong E;Xiaosu Zhan

  • Affiliations:

  • Computer Department, Beijing University of Posts and Telecommunications, Beijing, China;China Mobile Group Hebel Co., Ltd., China;Computer Department, Beijing University of Posts and Telecommunications, Beijing, China;Computer Department, Beijing University of Posts and Telecommunications, Beijing, China;Computer Department, Beijing University of Posts and Telecommunications, Beijing, China

  • Venue:
  • ICPCA/SWS'12 Proceedings of the 2012 international conference on Pervasive Computing and the Networked World
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

To satisfy the need of data processing speed, and increase cache hit rate of traditional hash indexing structure in Main Memory Database (MMDB), a page-based cache-sensitive hash indexing structure is proposed in this article. While maintaining the processing speed, storage efficiency, this structure greatly enhances the processor cache hit rate. This article describes the design and implementation of this cache-sensitive hash indexing structure in detail. After that, theoretical analysis and simulation experiments are performed. We come to a conclusion that for real MMDB, this new hash structure improves cache hit rate greatly and the whole database becomes more effective than traditional ones.