An Efficient Query Execution Plan for Multi-Way Joins in Shared-Nothing Database Environment

スポンサーリンク

概要

• 論文の詳細を見る

• In we have introduced an analytical model for the parallel processing of multi-way joins in shared-nothing database environment, with analysis of the hash-join algorithm applied to the linear trees represcnted in the left and right-deep tree structures. We have found that the length of the pipeline segment that depends on the structure of the query execution tree, and the utilization of the system resources (amount of main memory,disk and interconnection network bandwidth) are important factors in the efficient parallel processing of multi-way joins. In this paper we develop a scheme to determine the query execution plan based on the system resources balance so that the query can be efficiently executed in a shared-nothing database environment. As shown by our simulation, the proposed approach leads to query plans of significantly better performance than those achievable by the previous using left-deep and right-deep trees.

• 一般社団法人情報処理学会の論文
• 1993-03-01

著者

• Harada Lilian

  Fujitsu Laboratories

• Aakaboshi Naoki

  Fujitsu Laboratories

関連論文

• On Parallel Hash-Join Processing with Skewed Data
• An Efficient Query Execution Plan for Multi-Way Joins in Shared-Nothing Database Environment
• Evaluation Results of Multi-Way Joins in Shared-Nothing Database Environment : The Case for Left-Deep and Right-Deep Execution Query Trees

スポンサーリンク