Access process

	The middleware can make a custom mapping of table fields, table names, and schema names, and supports inconsistent default conversion of source and destination end table field data types, and supports field filtering;
	Support for custom business logic by implementing interfaces (already the demand side is doing custom business logic work).

	Source:
		Table structure information:
			schema name
			table name
			When the source targets are inconsistent, a mapping relationship is required.
			Primary key (required)
			Update time field (not required, if there is no source, the number of data on the target side is checked)
		bin-log:
			ROW format
		Database connection information:
			It is not required. If it is not provided, the source and destination data will not be checked.
		Data extraction connection information:
			Mysql account with replication privileges
    Target:
    	Table structure information : same as Source
    	Target database connection information	
    	If it is mysql, it is not case sensitive (temporarily not case sensitive)

	Source:
		Operation and Maintenance Engineer cooperate with OGG configuration.
		Table structure information:
			schema name
            table name
			When the source targets are inconsistent, a mapping relationship is required.
			Primary key (required)
			Update time field (not required, if there is no source, the number of data on the target side is checked)
		Database connection information:
			It is not required. If it is not provided, the source and destination data will not be checked.
    Target:
    	Table structure information : same as Source
    	Target database connection information	
    	If it is mysql, it is not case sensitive (temporarily not case sensitive)
		

	See the plugin development chapter for details.

	Involving custom business logic, it is necessary to advance the docking plug development work (developed by the demand side) to audit the custom logic code in the multi-threaded concurrent environment;
	If the custom business logic is not involved, the deployment period is completed within 1 day. The time is uncontrollable due to parameter synchronization.

	Use the hardware resources of the demand side to assist in deployment.

	In a short period of time, the existing hardware can carry the test available architecture resources, and use up the destruction.

	Unified deployment, separately applying hardware resources to access data synchronization clusters.
	Can be deployed separately if there are special needs.

	Help documentation:
		See the plugin development chapter.