An efficient quad-tree based index structure for cloud data management

  • Authors:

  • Linlin Ding;Baiyou Qiao;Guoren Wang;Chen Chen

  • Affiliations:

  • Key Laboratory of Medical Image Computing, Ministry of Education and College of Information Science & Engineering, Northeastern University, China;Key Laboratory of Medical Image Computing, Ministry of Education and College of Information Science & Engineering, Northeastern University, China;Key Laboratory of Medical Image Computing, Ministry of Education and College of Information Science & Engineering, Northeastern University, China;College of Information Science & Engineering, Northeastern University, China

  • Venue:
  • WAIM'11 Proceedings of the 12th international conference on Web-age information management
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

Recently, as a new computing infrastructure, cloud computing is getting more and more attention. How to improve the data management of cloud computing is becoming a research hot. Current cloud computing systems only support key-value insert and lookup operations. However, they can not effectively support complex queries and the management of multi-dimensional data due to lack of efficient index structures. Therefore, a scalable and reliable index structure is generally needed. In this paper, a novel quad-tree based multi-dimensional index structure is proposed for efficient data management and query processing in cloud computing systems. A local quad-tree index is built on each compute node to manage the data residing on the node. Then, the compute nodes are organized in a Chord-based overlay network. A portion of local indexes is selected from each compute node as a global index and published based on the overlay routing protocol. The global index with low maintenance cost can dramatically enhance the performance of query processing in cloud computing systems. Experiments show that the proposed index structure is scalable, efficient and reliable.