Efficient data storage in edge cloud computing for real-time system

In the last two decades, there has been a lot of progress in the information world and associated information technologies. For example, nowadays, we are witnessing the evolution of Internet to Internet of things (IoT). Internet of things (IoT) refers to billions of electronic devices around the world that are connected to each other via wireless networks, such as Bluetooth, ZigBee, Wi-Fi, 2G, 3G, 4G. Example applications are smart infrastructures, smart city, smart mobility, smart technologies, etc. These smart systems will generate huge amount of data which is captured from the physical world using sensors. We will need to store this data in a centralized system (e.g., cloud computing) and process it in real time to get faster response. Nowadays, data analytics on edge clouds is becoming more and more important in order to make real-time decisions. Edge cloud computing is a new cloud computing paradigm which allows to store and process data close to where it is generated instead of sending it to a centralized cloud. This greatly reduces the communicatin latency and enables real-time response for latency sensitive applications. Because of the limited storage on edge clouds, it ischallenging to find a trade-off between the quality and quantity of the data stored on edge cloud for real-time decisions. In this paper, we use three architectural layers for efficient data storage and management at the edge cloud. We use an adaptive algorithm that dynamically finds a trade-off between providing high prediction accuracy necessary to improve real-time decision and decreasing the amount of data stored in limited storage space. We have used smart home data set published by the Smart* project to analyze our efficient data storage mechanism.