頁籤選單縮合
題 名 | Multiple Disk Allocation Based on Round-Robin Hilbert Curve for Spatial Data |
---|---|
作 者 | Chang,Chin-chen; Jiang,Ji-han; | 書刊名 | International Journal of Information and Management Sciences |
卷 期 | 9:1 1998.03[民87.03] |
頁 次 | 頁11-23 |
分類號 | 494.59 |
關鍵詞 | R-tree range query; Hilbert curve; Shortest spanning path; Declustering; |
語 文 | 英文(English) |
英文摘要 | In this paper, we present a technique to arrange R-tree nodes on multiples disks. Our goal is to maximize the throughput of range queries by increasing the parallelism of query processing. Our approach adopts a distance-preserving mapping, the Hilbert curve, to map multi-dimensional points into one-dimensional values. Given a group of spatial objects, we construct a shortest spanning path In minimizing the total Hilbert distance and build the R-tree bottom up. Then, we distribute the R-tree nodes into disks by the way of round-robin. The good distance-preserving property of the Hilbert curve gives an interesting result that each disk contains R-tree nodes far away in the linear ordering. Our experimental results on range queries show that the proposed method achieves a very good declustering performance for spatial data. |
本系統中英文摘要資訊取自各篇刊載內容。