K-Storage-Node Problem of Distributed Data Storage for Internet of Things

Abstract

Ubiquitous sensing enabled by wireless sensor network results in increasingly large amounts of sensor data. Effective data storage and query are effective means of dealing with this issue, and distributed storage technology is a focus in the current research. This paper focuses on the issue of storage node selection, and discusses how to select k nodes as storage nodes when the data generating speed of nodes are different. This problem is formulated as a k-storage-node problem in this paper and proven to be NP-hard, then three distributed storage schemes are proposed; random strategy based data storage scheme (RDS), reverse greedy strategy based data storage scheme (GDS), and SQGA (small world model based quantum genetic algorithm) based data storage scheme (SDS). Simulation results showed that GDS and SDS had better performance than RDS in the network lifecycle, energy consumption, storage delay and query delay. Furthermore, taking the balance of node energy consumption into account, SDS performed better.

Keywords:

• Data storage
• wireless sensor network
• big data
• k-storage-node problem
• internet of things