Efficient Resource Management Scheme for Storage Processing in Cloud Infrastructure with Internet of Things
Dec. 2016. By Young-Sik Jeong
Keyword: Resource management, Internet of Things (IoT), Cloud computing, Resource cassification, QoS
Recently, research on cloud-integrated Internet of Things where an Internet of Things (IoT) is converged with a cloud environment has been actively pursued. An IoT operates through interaction among many composition elements, such as actuators and sensors. At present, IoTs are used in diverse areas (for example, traffic control and safety, energy savings, process control, communications systems, distributed robots, and other important applications). In daily life, IoTs should provide services of high reliability corresponding with various physical elements. In order to guarantee highly reliable IoT services, optimized modeling, simulation, and resource management technologies integrating physical elements and computing elements are required. For such reasons, many systems are being developed where autonomic computing technologies are applied that sense any internal errors or external environmental changes occurring during system operation and where systems adapt or evolve themselves. In an IoT environment composed of large-scale nodes, autonomic computing requires a high processing amount and efficient storage processing of computing in order to process sensing data efficiently. In addition, due to the heterogeneous composition of IoT environments, separate middleware is required to share collected information.
This research proposed an efficient resource management scheme (ERMS) that efficiently manages IoT resources using cloud infrastructure satisfying the high availability, expansion, and high processing amount requirements. ERMS provides a XML-based standard sensing data storage scheme in order to store and process heterogeneous IoT sensing data in the cloud infrastructure. In addition, ERMS provides classification techniques to efficiently store and process distributed IoT data.