Geospatial Data Streams

Abstract

The recent rapid development of wireless communication, mobile computing, global navigation satellite systems (GNSS), and spatially enabled sensors enables the exponential growth of available spatiotemporal data produced continuously at high speed. Owing to these advancements, a new class of monitoring applications has come into focus, including real-time intelligent transportation systems, traffic monitoring, and mobile objects tracking. These new information flow processing (IFP) application domains need to process huge volumes of geospatial data arriving in the form of continuous data streams. IFP applications are pushing traditional database technologies beyond their limits due to massively increasing data volumes and demands for real-time processing. Data stream management systems (DSMSs) have been developed by the database community to query and summarize continuous data streams for further processing. Owing to of pure relational paradigms, DSMSs have rudimentary geospatial processing capabilities. Geospatial stream processing refers to a class of software systems for processing high-volume geospatial data streams with very low latency, that is, in near real time. DSMSs are oriented toward processing large data streams in near real time. Despite the differences between these two classes of management systems, DSMSs resemble DBMSs ; they process data streams using SQL, SQL-like expressions, and operators defined by relational algebra. Geospatial data streams, that is, real-time, transient, time- varying sequences of geospatial data items, generated by embedded positioning sensors demonstrate at least two Big Data core features, volume and velocity. Increasingly, a dominant approach is to leverage in-memory computing over a cluster of commodity hardware. Similar to centralized DSMSs, existing distributed in-memory query engines and their processing models are predominantly based on relational paradigms and continuous operator models without explicit support for geospatial queries. There is a clear need for a highly scalable data stream computing framework that can operate at high data rates and process massive amounts of large geospatial data streams. The goal of this chapter is twofold. First, to give an insight into geospatial stream processing at the conceptual level, that is, exclusively from the user’s perspective, using a declarative, SQL- based approach. Second, to present a novel, in-memory parallel, and distributed prototype that supports real-time processing and analysis of large geospatial data streams.