CAJ | 학술논문

现有的时空数据库能够支持对大规模移动对象数据的时空聚集查询。但移动对象数据量巨大，且目前多采用单线程进行查询计算，其查询速度无法满足应用需求。由于时空聚集查询具有良好的可并行性，提出了一种基于区域划分的多线程并行时空聚集查询方法。该方法在服务器端对时空域进行格网划分，形成多个时空立方体，采用多线程并行的技术方法在每个时空立方体上进行时空聚集查询，最后将查询结果整合。通过对真实数据集进行的实验分析，该方法在查询速度上有明显提升。
현유적시공수거고능구지지대대규모이동대상수거적시공취집사순。단이동대상수거량거대，차목전다채용단선정진행사순계산，기사순속도무법만족응용수구。유우시공취집사순구유량호적가병행성，제출료일충기우구역화분적다선정병행시공취집사순방법。해방법재복무기단대시공역진행격망화분，형성다개시공립방체，채용다선정병행적기술방법재매개시공립방체상진행시공취집사순，최후장사순결과정합。통과대진실수거집진행적실험분석，해방법재사순속도상유명현제승。
Existing spatiotemporal database can support spatiotemporal aggregation query over large-scale data of moving objects. However, as the data volume is huge and single-thread processing is used, the query speed cannot satisfy the application requirements. Taking consideration of the parallelism of spatiotemporal aggregation, a spatiotemporal aggregation querying method is proposed by using multi-thread parallel technique based on region division. Concretely, the spatiotemporal domain is divided into several spatiotemporal cubes on the server, spatiotemporal aggregation over each cube is computed by using the method of multi-thread parallel technique, and all the querying results are integrated. By testing and analyzing on the real datasets, it is demonstrated that this method has enhanced the querying speed significantly.