In this abstract, we propose the first framework to support fault-tolerant and energy-efficient remote storage and processing under a dynamic network topology, i.e., mobile cloud. Our framework aims for applications that require energy-efficient and reliable distributed data storage and processing in dynamic network. We integrate the k-out-of-n reliability mechanism into distributed computing in mobile cloud formed by only mobile devices. k-out-of-n, a well-studied topic in reliability control, ensures that a system of n components operates correctly as long as k or more components work. More specifically, we investigate how to store data as well as process the stored data in mobile cloud with k-out-of-n reliability such that: 1) the energy consumption for retrieving distributed data is minimized; 2) the energy consumption for processing the distributed data is minimized; and 3) data and processing are distributed considering dynamic topology changes. In our proposed framework, a data object is encoded and partitioned into n fragments, and then stored on n different nodes. As long as k or more of the n nodes are available, the data object can be successfully recovered. Similarly, another set of n nodes are assigned tasks for processing the stored data and all tasks can be completed as long as k or more of the n processing nodes finish the assigned tasks. The parameters k and n determine the degree of reliability and different ðk; nÞ pairs may be assigned to data storage and data processing. System administrators select these parameters based on their reliability requirements.
You are here: / / K-Out-Of-N Computing Approach For Data Storage And Processing In Mobile Cloud