• Citer

Distributed moving object data management in MobilityDB

par Sayed Mohamed Bakli E., Mohamed ;Sakr, Mahmoud ;Zimanyi, Esteban
Référence BigSpatial '19: The 8th ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data(November 03 - 05, 2019: Chicago, Illinois, USA)
Publication Publié, 2019-11-05

Publication dans des actes

• Accès en ligne
• Détails
• Contenu
• Statistiques

Résumé :

The availability of moving object data that is being collected nowadays, and the demand of using them in applications, have generated the need for spatiotemporal data management systems. MobilityDB is an open source moving object database system. Its core function is to efficiently store and query moving object trajectories. It is engineered up from PostgreSQL and PostGIS, providing spatiotemporal data management via SQL. In order to store and analyze the massive datasets of trajectories, a scalable version is required. In this paper, we present a solution to distribute MobilityDB using Citus. Citus is a PostgreSQL extension for distributed query processing. We report on the integration architecture, and the types of queries that can be distributed out of the box. The experiments prove the feasibility of the solution, and show a significant speed up in queries.