Towards Autonomous Load Balancing for Mobile Data Stream Processing and Communication Middleware Based on Data Distribution Service

Intelligent Transportation Systems, Fleet Management and Logistics, and integrated Industrial Process Automation share the requirement of remote monitoring and high performance processing of huge data streams produced by large sets of mobile nodes. The deployment and operation of infra-structures enabling such mobile communication and data stream processing have two key requirements: they must be capable of (a) handling large and variable numbers of wireless connections to the monitored mobile nodes regardless of their current use or locations, and (b) automatically adapt to variations in the volume of the mobile data streams. This article describes a mechanism for load balancing of mobile data streams based on the autonomic reference model MAPEK. The autonomic capability has been incorporated into a scalable middleware system based on the OMG DDS standard and aimed at real-time and adaptive handling of mobile connectivity and data stream processing for great sets of mobile nodes. Besides explaining the autonomic extension of this middleware (MAPE-SDDL) and the generic load balancing approach implemented, we also present results of initial tests and discuss the potential benefits of using this model for general dynamic adaptivity in distributed middleware. Keywords-Load balancing; Data Stream Processing; Autonomic computing; DDS; mobile communication middleware.