Concurrent accesses to R-trees

Abstract

Access to spatial objects is often required in many nonstandard database applications, such as GIS, VLSI and CAD. In this paper, we examine the R-tree as an index structure, and modify it to allow concurrent accesses. We investigate three different locking methods for concurrency control. The first method uses a single lock to lock the entire tree, allowing concurrent searches but only sequential updates. The second method locks the whole tree only when the splitting or merging of nodes in the tree is required. The third method uses the lock-coupling technique to lock individual nodes of the tree.

We discuss in detail three common user operations, search, insert, and delete, on the R-tree. A system operation, maintain, which may be invoked when there is overflow or underflow at some nodes is also discussed. Concurrency control algorithms using the three locking methods mentioned above are described in this paper. They have also been implemented in the SR distributed programming language. We report on our preliminary results comparing their relative performance.

This work was supported in part by the Natural Sciences and Engineering Research Council of Canada under the Networks of Centres of Excellence Program and in part by the Advanced Systems Institute of British Columbia.