iz

Internet of Things (IoT) along with smart processing devices used to provide edge computing environment supports several daily life applications including smart healthcare, city traffic management, automated cars, and smart power grids for efficient supply of electricity. These applications usually process the data using cloud computing resources that increases application latency. Application execution on the Edge computing environment allows to perform application processing inside the edge network hence reducing the latency and increasing the utilization of local and available computing machines. This also helps to avoid heavy infrastructure cost or renting out the cloud resources. The problem of processing big size data and computational oriented applications can be resolved by combining the resources of devices inside the edge network executing resource intensive task. The traditional clustering methods are centralized have overhead of creating and maintaining the cluster. For this paper a decentralized grouping of edge nodes is provided that groups devices on arrival using an organizer node who have some high end job, and handle the execution in parallel on group devices by distributing the tasks to them. The proposed method reduces traffic amount communicated to cloud and allow application processing in the edge network resulting in reduction of latency as well as the cost in terms of the infrastructure. The decentralized group formation proposed here along with task partitioning works in the edge computing as the results are provided for this. Detailed results ensure reducing overall latency, limiting bandwidth and use of local resources hence this results in saving the economic cost.

Keywords: Computing, IOT, Processing, Effect, Application.