Search

Reactive Systems, especially in the Big Data Era, are undergoing a revolution. The increasing interest is due to their responsiveness, elasticity, resiliency and message-driven trait. In particular, a reactive message-driven system has one core feature asynchronous communication assuring a non-blocking behavior. A particular case of such systems is the use of actors to assure encapsulation and support fast decisions, failure handling and progess. The Actor Model meets the requirements for buiding a highly conccurent, scalable, distributed system, and Akka's tools provide the means to build it. However, to the best of our knowledge the effect of intra-actor asynchronous message processing on the routing strategies and the load balancing issue has not been addressed in the literature yet. The aim of this study is to analyse the behavior of the routers and routees over the existing routing strategies when each routee processes its messages asynchronously. Moreover, we study how the load is balanced among the routees and propose our own routing strategy that can reduce the processing time overhead by considering the complexity of the messages. For that, we build a distributed system to offline process massive trajectories, initially stored in HDFS, with Akka, the Play Framework, and MongoDB. Specifically, we leverage the Geolife projectâArZs GPS trajectory dataset.

The proliferation of indoor and outdoor tracking devices has led to a vast amount of spatial data. Each object can be described by several trajectories that, once analysed, can yield to significant knowledge. In particular, pattern analysis by clustering generic trajectories can give insight into objects sharing the same patterns. Still, sequential clustering approaches fail to handle large volumes of data. Hence, the necessity of distributed systems to be able to infer knowledge in a trivial time interval. In this paper, we detail an efficient, scalable and distributed execution pipeline for clustering raw trajectories. The clustering is achieved via a fuzzy similarity relation obtained by the transitive closure of a proximity relation. Moreover, the pipeline is integrated in Spark, implemented in Scala and leverages the Core and Graphx libraries making use of Resilient Distributed Datasets (RDD) and graph processing. Furthermore, a new simple, but very efficient, partitioning logic has been deployed in Spark and integrated into the execution process. The objective behind this logic is to equally distribute the load among all executors by considering the complexity of the data. In particular, resolving the load balancing issue has reduced the conventional execution time in an important manner. Evaluation and performance of the whole distributed process has been analysed by handling the Geolife project&rsquo, s GPS trajectory dataset.