论文信息 - A parallel spatial join framework using PMR-quadtrees

A parallel spatial join framework using PMR-quadtrees

Spatial join is the most important and complex operation in spatial databases. Therefore, there is a lot of ongoing research presenting algorithms and data structures to efficiently process spatial joins. In this work we present a parallel solution for spatial join processing with a dynamic load balance strategy. We have implemented PMR-quadtrees (Samet, 1990) on top of persistent object structures using an object oriented database server.

Marta Mattoso | Claudio Esperança | Alexandre A. B. Lima

[1] Hanan Samet,et al. Performance of Data-Parallel Spatial Operations , 1994, VLDB.

[2] Hans-Peter Kriegel,et al. The R*-tree: an efficient and robust access method for points and rectangles , 1990, SIGMOD '90.

[3] Jano Moreira de Souza,et al. A Raster Approximation For Processing of Spatial Joins , 1998, VLDB.

[4] Walid G. Aref,et al. Hashing by proximity to process duplicates in spatial databases , 1994, CIKM '94.

[5] Hanan Samet,et al. A qualitative comparison study of data structures for large line segment databases , 1992, SIGMOD '92.

[6] Hans-Peter Kriegel,et al. Parallel processing of spatial joins using R-trees , 1996, Proceedings of the Twelfth International Conference on Data Engineering.

[7] David J. DeWitt,et al. Partition based spatial-merge join , 1996, SIGMOD '96.

[8] Hanan Samet,et al. The Design and Analysis of Spatial Data Structures , 1989 .

[9] Hans-Peter Kriegel,et al. Efficient processing of spatial joins using R-trees , 1993, SIGMOD Conference.

[10] Hanan Samet,et al. Benchmarking Spatial Join Operations with Spatial Output , 1995, VLDB.