PROJECT TITLE :

Mobility-Aware and Delay-Sensitive Service Provisioning in Mobile Edge-Cloud Networks

ABSTRACT:

Mobile edge computing, also known as MEC, is a relatively new technology that has emerged as a potentially useful tool for bringing the cloud to the network's periphery and bringing network services closer to mobile users. When users are served from edge clouds, service latency can be reduced, operational costs can be reduced, and network resource availability can be improved. In addition to the MEC technology, network function virtualization, also known as NFV, is another promising technique that implements a variety of network service functions as pieces of software within cloudlets (servers or clusters of servers). When virtualized network services are made available to mobile users, the user service experience can be improved, network service deployment can be made simpler, and network resource management can become simpler. Mobile users, on the other hand, move in and out of networks at will, and different users typically ask for different services, each of which has unique resource requirements and delay requirements. It is therefore a very difficult task to provide mobile users in a MEC network with reliable and seamless virtualized network services while also meeting their individual delay requirements. This is all subject to the capacities of the resources available on the network. In this paper, our primary focus is on the provisioning of virtualized network function services for mobile users in MEC in a way that takes into account the requirements pertaining to user mobility as well as delay in service. First, we develop two brand-new optimization problems for user service request admissions. The goals of these problems are to maximize the accumulative network utility and the accumulative network throughput for a specific amount of time, respectively. After that, we come up with an algorithm for the utility maximization problem that uses constant approximation. In addition to this, we devise an online algorithm to solve the problem of maximizing the accumulative throughput. In conclusion, we use experimental simulations to assess how well the proposed algorithms perform. The outcomes of the experiments show that the proposed algorithms have a lot of potential.