QoS-aware single service selection mechanism for ad-hoc mobile cloud computing

Mobile technology has made notable progress over the past few years, this is due to the development of new handheld devices, improved wide area cellular coverage and seamless integration of wireless data access into mobile devices. These recent advances in wireless internet technologies have given birth to the Mobile Cloud Computing paradigm. However, intermittent internet disconnection and others lead to the evolution of ad-hoc mobile cloud where mobile devices exposes their computing resources and residence services to other devices. One challenge in the ad-hoc mobile cloud is that of service selection especially in a virtualized ad hoc mobile environment. This is because the best service to be selected might have left after discovery as a result of the dynamic nature of the ad hoc mobile system. To resolve this challenge, we propose a single service selection mechanism where the instances or images pertaining to the information of community members are stored temporarily on centralized virtual node. When a selection is to be done, these images are searched instead of searching individual node. A multi-criteria decision is proposed as our solution approach. Experiment is conducted using Execution time as the QoS parameter. Evaluation and analysis are carried out. We first recorded an increase in execution time as the number of services increases. But at a certain period, the execution time increment is not well noticeable or proportional to service increment. This is attributed to the service images that are already in the virtual leading server thereby reducing the execution time.