ALGORITHMS FOR EFFICIENT TOP-K SPATIAL PREFERENCE QUERY EXECUTION IN A HETEROGENEOUS DISTRIBUTED ENVIRONMENT

摘要

Top-k spatial preference queries allow searching for objects on the basis of their neighbourhoods' character. They find k objects whose neighbouring objects satisfy the query conditions to the greatest extent. The execution of the queries is complex and lengthy as it requires performing numerous accesses to index structures and data. Existing algorithms therefore employ various optimization techniques. The algorithms assume, however, that all data sets required to execute the query are aggregated in one location. In reality data is often distributed on remote nodes like for example data accumulated by different organizations. This motivated developing algorithm capable of efficiently executing the queries in a heterogeneous distributed environment. The paper describes the specifics of operating in such environment, presents the developed algorithm, describes the mechanisms it employs and discusses the results of conducted experiments.