A COMPARISON OF HEURISTICS FOR SCHEDULING SPATIAL CLUSTERS TO REDUCE I/O COST IN SPATIAL JOIN PROCESSING
In spatial join processing, a common method to minimize the I/O cost is to partition the spatial objects into clusters, and then to schedule the processing of the clusters in the spatial join processing such that the number of times the same objects to be fetched into memory can be mi nlmized. A key issue of this clustering-and-scheduling approach is how to produce a better sequence of clusters to guide the cluster scheduling thus to reduce the total I/O cost of spatial join processing. This paper describes three cluster sequencing heuristics. An extensive comparison among them has been conducted, and simulation results have shown that, while using the cluster sequences generated to guide the cluster scheduling can significant reduce the I/O cost in fetching spatial objects in spatial join processing, their performance differs.
Spatial databases Spatial join processing maximum spanning tree Ant colony optimization scheduling Match
JI-TIAN XIAO
School of Computer and Information Science, Edith Cowan University, 2 Bradford Street, Mount Lawley, WA 6050,Australia
国际会议
2006 International Conference on Machine Learning and Cybernetics(IEEE第五届机器学习与控制论坛)
大连
英文
2455-2460
2006-08-13（万方平台首次上网日期，不代表论文的发表时间）