Method for realizing load balancing cluster of MICROSOFT SQL SERVER database

Abstract

The invention provides a method for realizing a load balancing cluster of an MICROSOFT SQL Server database. The design philosophy of the invention is to enable a plurality of SQL Server database nodes to work at the same time through a database middleware program, and when a certain node has a fault, the rest nodes can provide a service continuously. The method is characterized in that: on the basis of architecture without a shared disk, the database middleware program is deployed on the nodes of an SQL Server database cluster; and when any node data in the cluster changes, the database middleware can synchronize the changed data to other nodes. Simultaneously, the database middleware can sufficiently ensure the consistency of the data all the time. A database query operation can be executed in any node so as to achieve a real load balance.

Description

technical field [0001] The invention relates to the application field of database clusters, in particular to a method for realizing load balancing clusters of Microsoft SQL Server databases. Background technique [0002] Database cluster (Cluster) software still meets the requirements of market competition. Although various manufacturers claim how perfect their architecture is, it still cannot change the fact that Oracle is the leader and everyone is chasing after it. Oracle RAC is still ahead of Microsoft in terms of cluster solutions. Other database vendors, it can meet the needs of customers for high availability, high performance and easy expansion. [0003] Microsoft has occupied a large part of customers with its advantages such as ease of use, but at the same time it has to face such a situation: with the development of enterprises, the amount of data and visits of enterprises will also increase rapidly (such as some e-commerce enterprises or websites ), and at this ...

AI Technical Summary

Problems solved by technology

The system cannot be expanded sustainably, and the problem cannot be fundamentally solved;

[0005] Second, build a database cluster, and the cluster solution provided by Microsoft cannot meet the needs very well. Many users are faced with the problem of transplanting to the Oracle platform and using RAC to solve the problem.

[0006] So is there a solution that does not develop in the direction of large machines and does not need to be upgraded to Oracle RAC? Can a cluster solution with high performance, high availability and easy expansion like Oracle RAC be implemented on the Microsoft SQL Server architecture? This is exactly the problem that the present invention tries to solve

Images