We present a strategy for geographically distributed spatial query optimization that involves multiple sites. Previous work in the area of geographically distributed spatial query processing and optimization focused only on strategies for performing spatial joins and spatial semijoins, and geographically distributed spatial queries that only involve two sites. We propose a strategy for optimizing a geographically distributed spatial query, which uses spatial semijoins and can involve any number of sites in a geographically distributed spatial database. It identifies and initiates spatial semijoins from the smaller spatial relations in order to reduce the larger spatial relations. By doing so, the data transmission and I/O costs are significantly reduced. We compare the performance of our strategy against the naïve approach of shipping entire spatial relations to the query site. We find that our optimized strategy minimizes the data transmission cost and I/O cost in all cases, and significantly in specific situations. In addition, the CPU cost is not significantly affected by our strategy.

中文翻译：

---

在分布式空间查询处理中在多个站点上使用空间半连接

我们提出了一种涉及多个站点的地理分布式空间查询优化策略。以前在地理分布式空间查询处理和优化领域的工作仅关注执行空间连接和空间半连接的策略，以及仅涉及两个站点的地理分布式空间查询。我们提出了一种优化地理分布式空间查询的策略，该策略使用空间半连接并且可以涉及地理分布式空间数据库中的任意数量的站点。它从较小的空间关系中识别并启动空间半连接，以减少较大的空间关系。这样做可以显着降低数据传输和 I/O 成本。我们将我们的策略的性能与将整个空间关系传送到查询站点的幼稚方法进行比较。我们发现我们的优化策略在所有情况下都最大限度地降低了数据传输成本和 I/O 成本，并且在特定情况下显着。此外，CPU 成本不受我们策略的显着影响。