Scheduling Correlated IoT Application Requests Within IoT Eco-System: An Incremental Cloud Oriented Approach

Abstract

In the era of adopting Internet everywhere, the recent modernization through incredible Internet of Things (IoT) makes the world more shrewd and automated. Thousands of users are connected every day in order to get benefits from the promising IoT solutions. Subsequently, a colossal amount of requests demanding different types of services for different IoT eco-system applications are very frequent. This creates a big challenge to the IoT service providers because IoT has a limited resources for storing and processing of huge data generating every second. Therefore, managing these resources while maintaining a good service quality is also an obvious challenge for the service provider as the ultimate goal is to satisfy user expectations. In this context, to abolish the restraints of IoT, it is, no doubt, a perceptive idea to fuse the powerful data analysis technique (data mining) with the compelling data management technique (cloud computing). Our paper proposes an Incremental Scheduling of Correlated IoT Requests (ISCIR) mechanism that makes a novel attempt to resolve the problem of scheduling massive IoT requests. It allocates adequate resources to the requests through cloud computing and data mining techniques while assuring a better service quality to the IoT application users. A monitoring system is used to collect the relevant data about the service requests and an appropriate information extraction process is used to get the useful patterns from the collected data. Thus, an incremental scheduling approach is proposed which enhances the knowledge about the system in its each iteration and uses that knowledge for managing the resources. An analytical analysis of the proposed approach considering the heterogeneity of IoT requests is derived. Finally, the simulation is done to validate the analytical analysis and the results show significant improvement in the performance of the system based on some prominent Quality of Service (QoS) parameters.