Data allocation optimization for sensor information of internet of things

摘要

As an emerging hot technology at home and abroad, the Internet of Things combines data characteristics with the advantages of distributed real-time database information storage management, and data distribution strategy as the key technology of data storage scheme is the focus of research. According to the mass, spatial-temporal correlation, access imbalance and continuous variability of sensor information in the Internet of Things, a time-domain based data allocation model is needed to adapt to it, so as to design a dynamic data allocation strategy based on adaptive time-domain load feedback. According to the data characteristics, the static data distribution problem is reduced to a simple linear programming problem, and the adaptive time domain is used to feedback the load information. Finally, the dynamic load threshold function is set to realize the dynamic data distribution. For the allocation of global scalar data, this paper uses integer linear programming for modeling, and proposes a global data allocation algorithm (GDP) based on RODP algorithm. GDP algorithm can quickly solve the allocation problem of scalar data in the whole program within the polynomial time complexity. Finally, the numerical experiments in the program cycle body show that the proposed strategy has better performance in terms of short time domain load balancing and system data migration than similar algorithms. Simulation experiments are carried out on two sets of benchmark programs respectively. The experimental results show that the global data allocation algorithm and the iterative optimal data allocation algorithm proposed in this paper are superior to the greedy strategy based data allocation algorithm in terms of access delay and energy consumption for all the test programs.