Method for high-risk query SQL based on database middleware

Abstract

The invention discloses a method for high-risk query SQL based on database middleware, and the method specifically comprises the steps: S1, receiving a query SQL sent by a business program through the middleware, and judging whether the query SQL is a slow SQL query or not; s2, counting the slow SQL query, and caching and recording the slow SQL query; s3, requesting to obtain an execution plan of the slow SQL query, and analyzing the condition whether the index exists in the execution plan or not; s4, analyzing a query condition of the slow SQL, and optimizing the slow SQL query; according to the method, manual participation of development or operation and maintenance personnel in optimization is not needed, each step is automatically realized, each optimization has a data basis, and problem SQL playback verification is carried out after optimization is completed. And for some SQL which cannot be optimized by adding indexes according to business needs, the method also performs management and control, on one hand, the concurrent quantity of the SQL with the problem is limited, and on the other hand, the SQL with the problem is forwarded to the slave database of the database, so that the stability of the master database of the database is protected, and the durability of the business system is realized.

Description

Technical field [0001] The invention relates to the technical field of database middleware, and is specifically a method for querying high-risk SQL based on database middleware. Background technique [0002] The main function of the existing database middleware is to realize functions such as separation of reading and writing, sub-database and sub-table, high availability, etc. However, it does not have a good management effect on the database attacks caused by high-risk query SQL generated by the business system. In the current era of 5G, AI, and big data, high concurrency scenarios put great pressure on databases. If database middleware can only manage the business data layer, it is completely insufficient. On the one hand, high-risk query SQL will cause database downtime, causing the business system to be unable to respond to use, and will ultimately have a huge negative impact on these Internet companies. We often see a certain Internet company's website or APP go down ...

AI Technical Summary

Problems solved by technology

[0002] The main function of the existing database middleware is to realize read-write separation, sub-database sub-table, high availability and other functions, but the high-risk query SQL generated by the business system does not have a good management effect on the database.

In the current era of 5G, AI artificial intelligence, and big data, high-concurrency scenarios put great pressure on the database. If the database middleware can only manage the business data layer, it is not enough at all

On the one hand, high-risk query SQL will cause the database to go down, which will cause the business system to be unresponsive, and will eventually have a great negative impact on these Internet companies

Images