Design And Implementation Of The Migration Middleware Of Heterogeneous Database

Data migration,refers to extracting data from a data environment(source database),migrating to another data environment(target database)as needed.Currently data migration needs to implement data subscription capabilities and diverse needs based on data migration and data subscription capabilities in addition to move data from a database to another.The data migration middleware currently researched and designed mainly includes two major functional modules: data migration and data subscription.The data migration module is mainly divided into full migration(migrating all data from the source database)and incremental migration(update the data generated by the source database to the target for playback).The full-scale migration can realize high-speed replication of data between heterogeneous databases,which can fully meet the functional requirements of existing data off-site disaster recovery;incremental migration can realize real-time data synchronization,and is mainly applied to markets such as data non-stop migration,but with the development of the Internet industry,higher requirements are imposed on the real-time nature of incremental migration which leads to more complex incremental migration requirements such as remote data sharing.Data subscription function refers to encapsulate data information into messages in a message queue so that consumers can freely consume the messages in the message queue,with this function we can improve the system work efficiency and decouple between the producer and the consumer and mainly applicable to the asynchronous message notification and Cache updates and more.The heterogeneous database migration middleware implemented in this paper also implements these three main functions,including full data migration,incremental data migration,and data subscription so that can achieve these functions such as non-stop data migration,data offsite disaster recovery,and live offsite,cross-border data synchronization,real-time data warehouse,query report offload,cache update,asynchronous message notification and more.