An index structure for parallel processing of multidimensional data

  • Authors:

  • KyoungSoo Bok;DongMin Seo;SeokIl Song;MyoungHo Kim;JaeSoo Yoo

  • Affiliations:

  • Department of Computer Science, Korea Advanced Institute of Science and Technology, Daejeon, Korea;Department of Computer and Communication Engineering, Chungbuk National University, Chungbuk, Korea;Department of Computer Engineering, Chungju National University, Chungbuk, Korea;Department of Computer Science, Korea Advanced Institute of Science and Technology, Daejeon, Korea;Department of Computer and Communication Engineering, Chungbuk National University, Chungbuk, Korea

  • Venue:
  • WAIM'05 Proceedings of the 6th international conference on Advances in Web-Age Information Management
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

Generally, multidimensional data require a large amount of storage space. There are a few limits to store and manage those large amounts of data in single workstation. If we manage the data on parallel computing environment which is being actively researched these days, we can get highly improved performance. In this paper, we propose an efficient index structure for multidimensional data that exploits the parallel computing environment. The proposed index structure is constructed based on nP(processor)-n×mD(disk) architecture which is the hybrid type of nP-nD and 1P-nD. Its node structure increases fan-out and reduces the height of an index tree. Our proposed index structure gives a range search algorithm that maximizes I/O parallelism. The range search algorithm is applied to k-nearest neighbor queries. Through various experiments, it is shown that the proposed method outperforms other parallel index structures.