Communication between databases

Do databases and databases need to exchange data?

1.
Different systems need to be integrated, and direct data communication channels must be established

2.
Different business departments need to communicate, and the pipeline from the database to the database must be implemented

3.
Data flow must be realized between two databases in the same system.

Figure: data exchange between databases

What is the cross-database data exchange requirement?

1.
Data synchronization between databases of the same type

2.
Data synchronization between different types of databases

3.
Data conversion between databases and files

4.
Achieve cross-Database Business Operations

I. Data Flow between databases

"Simple data synchronization", "selective data synchronization", and "collaborative jobs" are three levels of data flow across databases.

Figure, Data Synchronization for data objects of the same structure

For "simple data synchronization", the corresponding table structures in the two databases are consistent, but data migration is completed.

Figure, Heterogeneous Data Object Data Synchronization

"Selective Data Synchronization": The corresponding table structures in the two databases are inconsistent, and even the Data Types of the corresponding columns are inconsistent. You need to select the corresponding columns and convert the data types.

Figure, cross-database collaboration

"Collaborative jobs": data migration between two databases can be completed, and business operations are triggered along with data flow. Perform cross-database federation. Collaborative jobs operate on two or more databases, just like operating on multiple data tables in a database, business data and business data are combined to implement business processes.

Ii. Methods of data flow between databases 2.1 Database Replication Server

The Database Replication Server is an effective tool for "simple data synchronization. It can synchronize small data volumes quickly.

Figure, copying data between objects with the same structure

The replication server triggers the replication action by monitoring the "operation log" of the master database, and performs the same action in the slave database. Is a trigger mode, so it is more timely and rapid.

The replication server requires that the master database and slave database copy objects be exactly the same. For example, the table structure, field data type, and even database type must be the same. In terms of performance, the system has been bound to flexible data circulation.

The replication server provides data synchronization tools from the database system.

2.2 Message Middleware

Message-oriented middleware is a data transmission pipeline tool implemented across data nodes.

Figure: message-oriented middleware for data transmission

Message-oriented middleware only implements the MPs queue layer. We need to compile the warehouse picking and warehouse receiving operation procedures. Then, data is transmitted across nodes through message-oriented middleware.

Using message middleware, We need to write database interface programs and business data flow control programs.

Message-oriented middleware provides data transmission pipelines from the network layer.

2.3 Data loop middleware ---- isframe DC

Data loop middleware DC (Data circulattion) is part of the isframe of the Information System Architecture middleware. Data circulation means that data is circulating between various organs like the blood of an organism, and operations of various organs are linked by carrying business information.

Both the replication system and message-oriented middleware provide a low-level interface. Isframe DC starts with business process control to meet the needs of cross-database, cross-data nodes, cross-System "Data Synchronization", and "collaborative jobs.

Figure: DC cross-node resource collaboration

DC connects resources of different nodes and databases to facilitate "Data Synchronization", "Data Conversion", and "joint operations, is a large set of operation objects formed by the associated database.

Iii. Features of Data loop middleware 3.1 eliminate database differences

"Data loop middleware" supports mainstream databases in Oracle, Sybase, sqlserver, and MySQL, and supports data transfer and conversion between application files and databases.

The database types can be different, the corresponding table structure can be different, and the Data Types of corresponding fields can be different.

Without the need for database interface programming, the middleware automatically performs operations such as data warehouse receiving, warehouse receiving, and execution of stored procedures.

3.2
Automatic Control of Data Flow

During cross-database data transmission, "Data loop middleware" can implement a complete dialog logic control. That is, DC Automatically loads data, performs the submit operation on the target end, and then returns the result of the submit operation to the source end for the receipt operation. "Loading", "Submission", and "receipt" constitute a complete conversation process.

3.3
Batch control of business data

In cross-database joint operations, DC can implement incremental step-by-step and rollback control. Cross-node operations are both intuitive and easy to control.

 

 

 

For more information about "Architecture middleware", visit www.tcxzsoft.com.