吉林大学学报(理学版)

• 计算机科学 • 上一篇    下一篇

林业资源数据集群的快速索引设计

李惺颖1,  谢阳生1,  唐小明1, 罗鹏1, 黄龙1,2   

  1. 1. 中国林业科学研究院 资源信息研究所, 北京 100091; 2. 北京林业大学 水土保持学院, 北京 100083
  • 收稿日期:2013-04-27 出版日期:2013-11-26 发布日期:2013-11-21
  • 通讯作者: 谢阳生 E-mail:Xieys@caf.ac.cn

Design of Fast Index for Forest Resources Data Cluster

LI Xing ying1, XIE Yang sheng1, TANG Xiao ming1, LUO Peng1, HUANG Long1,2   

  1. 1. Research Institute of Resource and Information, Chinese Academy of Forestry, Beijing 100091, China;2. School of Soil and Water Conservation, Beijing Forestry University, Beijing 100083, China
  • Received:2013-04-27 Online:2013-11-26 Published:2013-11-21
  • Contact: XIE Yang sheng E-mail:Xieys@caf.ac.cn

摘要:

设计一种快速索引以管理分布存储的林业资源数据, 通过大量独立数据节点共同负载数据请求, 减少数据访问过程的互斥, 实现了空间与属性数据的分布式存储并行访问. 实验结果表明, 基于目前的应用服务器集群, 应用该快速索引方法, 能以较低的改进成本提高整个数据集群的快速统计和更新能力.

关键词: 林业资源, 集群数据存储, 多库存储, 并行索引

Abstract:

This paper describes the design of a fast index for distributed storage of forestry resource data. According to the load data request of a large number of independent data nodes to reduce the mutual exclusion during data access, parallel access was realized for distributed storage of spatial and attributive data. Based on the common application server cluster, this fast index can be used by making the minor modifications of data access process so as to improve the performance of statistics and update of the whole data cluster with low cost.

Key words: forestry resource, cluster data storage, multiple database storage, parallel index

中图分类号: 

  • TP311