An Efficient Dynamic Multilevel Queuing and Multi Request Handling Model for Mobile Cloud Network Using Machine Learning Technique

The rapid increase in the usage of mobile devices in the day-to-day applications generates more requests and a suitable technology to handle the requests in short range of distance is Mobile Cloud Network (MCN). The research about the multiple requests handling problem is carried in this work by considering the performance, bandwidth utilization, Quality of Service (QoS) and cost issues. In this paper, a Dynamic Multilevel Queuing and Multi Request Handling (DMQMRH) model is proposed to address the above mentioned issues for efficiently handling multiple requests in dynamic environment. In the proposed model, first the multilevel queuing technique is used to schedule the requests in mobile cloud which are based on the prioritization without request dropping. Next, machine learning techniques such as prediction and classification are applied in this proposed model where the prediction technique predicts the request completion time of DMQMRH model to effectively utilize the bandwidth and the QoS classification technique classifies the type of service with minimal cost. Finally, the ascending priority queue scheduling technique is implemented in the proposed model to optimize the processing of multiple requests. The implementation and evaluation of these show that the proposed DMQMRH model is capable of handling the multiple requests with agreed performance, effective bandwidth utilization, minimal cost and also achieves QoS by agreed user prioritization without request drop in a better way when compared with the Propositional Deadline Constrained (PDC) and bi-criteria approximation algorithms without computing capacity (Approx_noCP) methods.