Analysis And Research On The Background Driven Platform Respect To The Big Data Synchronization Which Based On The China Mobile Ng4.0

With the advent of the era of Big Data, Data has become a major asset for a large of companies.How to effectively manage data together has become an urgent problem.Saving those data in different geographical or different storage space is an effective solution for the problem of data management.Then, Studying of big data synchronization has become a hot research direction.Study of data Synchronization has been a lot of development on System disaster recovery,campus intelligence, wisdom agriculture and so on.This paper background for data synchronization of “China mobile ng4.0 renovation project”,fully analyzed NGBOSS system for current mobile CRM and BILLING.In this system we faced slow sync, high frequency error occurs,the probability of the need to manually handle,for those problems we proposed a new solution,The main work of this paper is following two:1.Current business operation support system in the process of generating the synchronization file specification does not synchronize the file size, resulting in a long time consuming data synchronization occurs during synchronization, the synchronization process or even dead, the need for manual handling,face those problems the paper designed a custom data switch,main thinking is the following:defined a data switch in the form of parameters in a XML file,This switch used for providing data synchronization threshold of each file can hold CRM host fish for synchronizing data according to the switch parameters.Experimental data shows that this method can control the synchronization file size,Solved since the synchronization file is too large to synchronize data failed to affect the whole system operation.2.Synchronize files exist in duplicate data,arranged disorderly,wasted disk space, consume more network resources, impact efficiency synchronous.In order to solve these problems there is a method to serialize the synchronizing files.the theory is that:At first,CRM side generating synchronous temporary files;Then,JAVA program cross-call Linux command,merge the synchronization temporary file,delete duplicate records in the file and synchronize files to rearrange the sequence.This method can effectively remove the redundant data synchronization file,specification synchronous serial data file,help to save the host disk space,reduce the file on a network resource consumption during synchronization,improve the efficiency of data storage.