Processing of Multiple Long-Running Queries in LargeScale Geo-Data Repositories

Abstract

One of the main challenges of large-scale information mediation and warehousing is the efficient and resource-effective processing of continuous queries. Continuous queries monitor streams of incoming data and produce results on-the-fly. They are usually long-running, yet need to be removed from the system when they become obsolete. We present a framework and the corresponding algorithms and data-structures for the efficient evaluation of multiple long-running spatial queries over unbounded streams of spatial data and the management of obsolete queries. Using both real-life and synthetic datasets and workloads, we show that our proposed approach achieves significant improvements over standard approaches