Rule-driven and data-driven combined cross-platform SQL query optimization method

Abstract

The invention discloses a rule-driven and data-driven combined cross-platform SQL query optimization method. The method comprises the following steps of: 1, analyzing a cross-platform SQL statement into a logic query plan in a system; 2, enabling an optimizer scheduling module to schedule the most suitable optimizer to perform query optimization according to the characteristics of the logic queryplan; 3, enabling the rule-driven optimizer to perform plan search according to the rule, selecting an execution plan according to the cost model and the cardinal number estimation to obtain an optimal physical execution plan, and importing an optimization result into a sample collection module; and 4, converting the sample imported by the sample acquisition module into a training sample through adata adaptation module, performing reinforcement learning model training by using the training sample through a data-driven optimizer, and inputting query into the trained model to obtain an optimalphysical execution plan. The cross-platform SQL query optimization method solves the problems that an existing cross-platform SQL query optimization method is poor in expansibility, low in flexibility, poor in optimization effect and the like.

Description

technical field

[0001] The invention relates to the technical field of cross-platform SQL query optimization, in particular to a cross-platform SQL query optimization method combining rule-driven and data-driven methods. Background technique

[0002] In recent years, with the increasing demand for big data analysis and processing applications in all walks of life, big data query systems have gradually developed in a diversified direction. These data query systems have their own characteristics in terms of query language, computing model, system architecture, and underlying storage technology, and are suitable for different vertical application scenarios. Therefore, modern enterprises or organizations usually construct a variety of different data query systems in order to handle diversified businesses. system. However, many comprehensive businesses need to be able to conduct convenient and efficient cross-platform data query. For example, the unified data analysis business b...

Images