Real-time Calculating Over Self-Health Data Using Storm

With the continuous development and popularity of smart wearable devices, more and more people tend to use the devices to record their health indicators and exercise indicators. Thus a larger amount of indicators called self-health Datais generated all the time. Obviously, it is necessary to process the data in real-time. For example, it may lead to serious problems when someone have an emergency, but if we can process the data in real-time, such situation can be avoidable. However the existing treatments have deficiency in real-time processing. This paper proposed a real-time processing scheme for the self-health data from a variety of wearable devices. We designed a framework using Apache Storm, distributed framework for handling stream data, and making decisions without any delay. Apache Storm is chosen over a traditional distributed framework (such as Hadoop, MapReduce and Mahout) that is good for batch processing. We contrasted different methods to verify the effectiveness of the proposed framework, and we also provided real-time analytic functionality over stream data to show and to improve the efficiency greatly. In our framework we have improved the efficiency by 68 percent compared with the old method of using regular task with DB cluster.