Research on the archiving method of social media information network based on GLOBUS in the context of big data

In order to solve the problem that traditional archiving methods do not support multi-threaded simultaneous data discovery, resulting in low archiving processing efficiency, a GLOBUS-based archiving method for social media information networks in the context of big data is proposed. Under the framework of GLOBUS, the P-MDS discovery model is designed to perform multi-threaded search on social media grid resources. The distributed spanning tree algorithm is introduced to improve the LRC service of GLOBUS framework to locate the resource grid copies in the information network. The SVM algorithm combined with cardinality statistics is used to classify the data within the grid copies and establish the corresponding storage mapping to archive the social media information data. The simulation experimental results show that the processing efficiency of this method is about 3.4 times that of the comparison method, and it puts less pressure on the hardware and has higher archiving processing efficiency.