Optimal Design And Implementation Of Distributed Data Storage And Access For Crowd Intelligence

Crwod intelligence treat ordinary users of smart mobile devices(mobile phone,pad)as the basic unit of collaboration awareness,consciously or unconsciously through the mobile Internet,the perceived task distribution and perception of data collection and storage,complete the social perception task of large-scale and complex,the swarm intelligence in intelligent sensing transportation,environmental monitoring,health care and other aspects are widely used.In the mobile intelligent terminal,the diversity of the sensor makes the type and structure of the sensing data varied,which puts forward higher requirements for the data storage and access in the process of crwod intelligence.According to the characteristics of massive,heterogeneous,real-time sensing data have different sensor crowdsourcing process,this paper first analyzes the characteristics of the data crwod intelligent process and classify sensor data of different methods,then analysis the existing distributed data access and storage and access to crwod intelligence process of metadata management.Therefore,in view of the characteristics of perceptual data and the storage and access requirements of different applications,the following research work has been done in this paper:(1)According to the characteristics of multi-source,sensing data of crowdsourcing in the process of heterogeneous,this paper studies the characteristics of various types of data sensing equipment,and according to the perceived characteristics of the classification of sensing data,and the validity of the classification is verified by the experiment of classification and prediction.(2)Study on the characteristics of multi-source metadata sensing data,this paper proposes a distributed metadata management method based on perception features and load balancing algorithm,and through metadata management based on perception features and load balancing experiments prove that this method can improve the performance of metadata management.(3)Study on the sensing data distributed storage and access method of different categories and characteristics based on numerical data,including the activation of data and graphics data,audio data access method,and through the access speed experiment of different types of data to verify that the method can optimize the performance of data storage and access.