Research On Decentralized Load Balancing Method For Distributed Spatiotemporal Database

At present,distributed spatio-temporal database has become the mainstream storage mode of spatio-temporal data,and it is an essential way to store multi-source heterogeneous data and massive spatio-temporal data.However,in the process of using a distributed spatio-temporal database to store and query data,there is a problem of low reading and writing efficiency caused by high concurrent users.Therefore,designing a good storage architecture and achieving load balancing has always been a research hotspot in distributed storage technology.From the current research situation,in the face of large-scale spatiotemporal data,a new cluster architecture solution needs to be designed.The load balancing strategy can be divided into two kinds: centralized control strategy and decentralized control strategy.Centralized control strategy is used in traditional distributed system,that is,a central node is used as the load node of cluster.It will lead to excessive reliance on the central node,which is not conducive to the stability of the cluster.To solve this problem,a spatio-temporal data storage method based on centralization is proposed in this paper.First,determine the three-tier architecture model and use the Postgres-XL cluster as a data storage solution.Then,a global partition table is established in the load balancing part to establish the mapping relationship between the nodes and the data.Finally,the load balancing indicators of each node are sorted and ranked to select the nodes with low loads to store,and the distributed storage of data is completed.During the query process,load balancing can reasonably allocate user access requests.Through various test experiments,it is shown that the architecture scheme and the proposed load balancing method in this paper can dynamically manage spatio-temporal data.In the face of multi-user high concurrency,load balancing of distributed nodes can be achieved.Therefore,compared with the centralized storage mode,the decentralized load balancing method is more suitable for storing spatio-temporal data.