A Service and Device Monitoring Service for Smart Items Infrastructures

It is a vision to integrate real-world data seamlessly into workflows and business applications using ubiquitous computing technology. A large number of deployed low-cost devices enhanced with sensors is observing the physical world. That way, real-world data is provided to business and workflow applications fully automated and in real-time without suffering from media-breaks, that means no manual and only few automated transformations of data formats are allowed. Status information like battery status, CPU and memory utilization are to be gathered in order to manage such a device fleet. In this paper a distributed monitoring service architecture to gather the required status data automatically is presented. The architecture is designed with respect to scalability requirements for a heterogeneous landscape mainly consisting of limited and low-powered devices. A prototype has been implemented and evaluated with respect to CPU and memory utilization. The performance analysis showed promising results in terms of the applicability of the chosen design.