Optimization of Computation and Communication Driven Resource Allocation in Mobile Cloud

With the emergence of accessing Smartphones in day-to-day life, Mobile Cloud Computing (MCC) technology has become popular with the advantage of resolving the resource constraints in mobile devices through the offloading method. The existing models have presented the different resource allocation solutions to ensure the seamless execution of the applications for the resource-constrained mobile devices with the Quality of Service (QoS). The optimization of resource allocation is the process of potentially allocating remote resources to mobile users without violating the Service Level Agreements (SLAs). However, resource allocation is still becoming a major constraint in the Mobile Cloud (MC) data centers due to higher consumption of energy and time factors during the execution of mobile requests on the remote cloud. The consumption of the energy and response time of the offloaded tasks or applications heavily relies on the cloud resource allocation for the mobile users. Hence, Resource Allocation Optimization (RAO) emerged as the significant objective to select the appropriate cloud resources for the requested tasks to increase the lifetime of the devices with improved time efficiency. Thus, this work focuses on optimizing MC resource allocation by optimizing the allocation of both the computation and communication resources. The proposed RAO model considers two potential factors, such as the energy and response time while allocating the computational and communicational resources. Initially, the Energy and Response time-driven RAO (EARO) approach prioritizes the request generated from the mobile users based on the estimated execution time. Modeling the Estimated Communication and Execution Time (ECET) algorithm tends to allocate the cloud resources and accomplish the minimal response time of the application requests. The EARO approach intends to minimize the execution time as well as the response time towards the target of alleviating the energy consumption during the resource allocation. Moreover, it selects the resources for the inter-VM communication with the knowledge of the minimal migration time ensuring bandwidth resources. Thus, EARO preserves the device's energy with minimal application completion time. The experimental results illustrate that the time efficiency of the proposed EARO model outperforms the existing resource allocation model in the MC environment.