Improving Computational Efficiency for Personalized Medical Applications in Mobile Cloud Computing Environment
George Mathew, Zoran Obradović · 2013
Mobile computing and cloud services are two technologies that have gained momentum in recent times. The proliferation of mobile computing devices and network connectivity has made it an attractive platform for delivering personalized services in many business domains including healthcare. Personalized health and wellness mobile applications have computational and data requirements that are necessitated by the localized processing needs of the application. The on-demand provisioning capability and elasticity of cloud services combined with the local processing capability of mobile devices can provide an ecosystem for pervasive access to health information. In this study we explore some of the specifics of these health and wellness applications. We introduce promotion algorithm as a mechanism to efficiently process data points locally by mobile devices. This algorithm can take advantage of the local processing power of smart phones and help reduce communication costs between mobile endpoint and cloud-based long-term data services. Experiments were performed using an Android smart phone for real time data acquisition of more than 10 million data points and a Linux server in a private cloud over 4G network simulating a health service. Results showed that the proposed algorithm could help preserve battery life by a factor of 10 and reduce data communication time by a factor of 20 as a result of utilizing local computation on a mobile device.