Towards a Congestion – Aware Sensor-Cloud Gateway Architecture for Real-Time Smart Living Applications
R. R. Darwish
Smart living the most modernized application nowadays availed oneself from the emergence of the sensor-cloud ontology. In particular, this work is targeting sensor-cloud based real-time smart living applications. The presented study is specifically focusing on magnifying the sensors’ requests servicing value at the gateway based sensor-cloud platforms. Resolving the possible requests’ congestion scenarios as a part of criticality biased sensors’ requests’ serving approach, to the extent of the author’s knowledge, has not been thoroughly investigated in the literature. Thence, the overwhelming objective of this study is guaranteeing criticality-aware gateway architecture with a sustainable throughput even during congestion conditions. The proposed gateway architecture encompasses two functional building blocks namely: dynamic traffic control module and autonomic real-time sensors’ requests’ scheduling module. The proposed traffic control module aimed at building a dynamic approach for consolidating the complementary incoming sensors’ requests in order to minimize the workloads to be dispatched to the scheduling module. On the other hand, the presented requests’ scheduling module assures a prior predictability for the overloading situations. Also, enforces a dynamic overloading resolving contingency scheduling plan, through which the timeliness of critical sensors’ requests during overloading is defended. Moreover, the presented scheduling module permits an efficient resources reclaiming behavior. Extensive simulations and comparisons have been implemented in order to validate the effectiveness of the presented gateway architecture. The conducted simulations confirm the proposed architecture’s optimal performance during normal condition as well as the architecture’s robust performance guarantee during overloading situations. Additionally, it has been proved that the timeliness of critical requests even during overloading conditions is ensured. Also, it has been indicated that the proposed gateway architecture capable of achieving a preferable scalability behavior without sacrificing the performance requirements. Finally, the conducted comparisons reveal the superiority of the proposed architecture against the state of art counterparts.
Keywords: Agglomerative clustering, criticality awareness, dynamic clustering, congestion management, sensor-cloud, smart living.