Document Type : Primary Research paper
Authors
1
Associate Professor, Department of Electronics and Communication Engineering, Erode Sengunthar Engineering College, Perundurai, Erode, Tamilnadu, India – 638057.
2
Associate Professor, Department of Electrical and Electronic Engineering, Kumaraguru college of Technology, Coimbatore-641049.
3
B.Tech, School of Computer Science and Engineering, VIT University Vellore 632014.
4
Department of Computer Science & Engineering, B.T. Kumaon Institute of Technology, Dwarahat, Uttarakhand – 263653.
5
Professor, Department of Electrical and Electronics Engineering, J.K.K. Munirajah College of Technology, T.N.Palayam, Erode, Tamilnadu, India - 638 506.
6
Associate Professor, Department of Electronics and Communication Engineering, Chaitanya Bharathi Institute Of Technology, Proddatur, Andhra Pradesh, India.
Abstract
Millions of mobile devices, such as sensors, healthcare devices, and smartphones, are now connected to the internet using IoT, allowing them to exchange data with one another. However, all IoT mobile devices have limited battery power and are unable to perform heavy task execution. To perform heavy task execution, all IoT devices used cloud services, but due to mobile mobility (changing location), large amounts of data were lost. To address the issue of latency, FOG computing was introduced, in which FOG terminals with massive resources such as computing power and storage will be deployed on various locations, and whenever a mobile offloads a heavy computation task, a close FOG terminal will accept the offloaded request, which will then be assigned to a VM, which will execute the request and send a response back to the mobile. The above technique of selecting a close FOG terminal does not focus on needed Resource Provisioning (which means how many VMs must be assigned to complete request processing in order to reduce response time) and Mobile Power Control (which requires controlling mobile transfer rate in order to reduce mobile power consumption). If the mobile transfer rate is reduced, the application must dynamically select a VM with a high processing performance to reduce response time latency (delay). When we choose a high-processing VM over a low-processing VM, the response time decreases while the cost (price) of the VM decreases. Because the high-processing VM is chosen, the execution is faster and the VM consumption time decreases. To address the issue of choosing a close FOG terminal, the authors propose a combination approach using Resource Provisioning (assigning Fog resources to mobile requests) and Mobile Power Control (lowering the mobile transfer rate to save the battery).
Keywords
)