Database self-learning optimization method and device based on traffic mirroring

Abstract

The present invention provides a database self-learning optimization method based on traffic mirroring, comprising: obtaining the latest database configuration parameters of the database to be optimized through an adaptive learning method, sending the latest database configuration parameters to the training database; establishing a database model of the database to be optimized Consistent database mirroring model, and send the database mirroring model to the training database; establish a traffic mirroring model consistent with the traffic model received by the database to be optimized, and send the traffic mirroring model to the training database; obtain training results fed back from the training database ; Update the database configuration parameters of the database to be optimized according to the training results of the training database; repeat the above steps R times, obtain the latest database configuration parameters when the TPS throughput of the training database in the R times of training results is the maximum value, as the final configuration parameters of the database Configured for the database to be optimized. For real traffic scenarios, it can effectively improve the performance of the database.

Description

technical field [0001] The invention relates to the field of database configuration, in particular to a traffic mirroring-based database self-learning optimization method and a traffic mirroring-based database self-learning optimization device. Background technique [0002] Multi-databases usually have many parameters. For example, MySQL and PostgreSQL have hundreds of parameters, while Oracle has thousands of parameters. These parameters will affect the running process of the database, thus affecting the overall throughput and response time of the database. big impact. For a database application scenario, proper parameter selection can improve performance several times. In order to improve the performance of the database, the company usually needs to hire a professional DBA to tune the database parameters. However, for different application scenarios, different loads, different hardware and operating systems, the optimal parameters required are often different. DBAs usual...

AI Technical Summary

Problems solved by technology

[0008] (1) In the face of hundreds of performance parameters, the DBA needs to analyze the load of the database, provide parameter configurations based on the type of database access SQL and personal experience, and conduct test observations, which requires a lot of time and labor costs

[0009] (2) For the scenario of expected traffic growth, DBA lacks estimated model verification, and usually needs to use experience to estimate. The effect of parameter setting is not only dependent on DBA's experience, but it is difficult to follow the rules of the attempted configuration. Difficult to achieve the best

[0010] (3) In addition, there are disadvantages such as single scene, immutable traffic, dependence on training data, serialization of training, and no support for online optimization.

Images