Resource optimization method for structured query and database query system

Abstract

The invention relates to the technical field of embedded databases, in particular to a resource optimization method for structured query and a database query system. The resource optimization method for the structured query includes the steps of acquiring multiple query statements, and configuring hardware modules used for query processing according to the query statements. The database query system includes an instruction acquisition module for acquiring the query statements and a query processing module for configuring the hardware modules used for query processing according to the query statements. According to the resource optimization method for the structured query and the system, the hardware modules used for completing query processing can be dynamically adjusted according to the queuing condition of current query statements to be processed, so that the hardware module resources are reasonably arranged, and the query efficiency is improved.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of embedded databases, in particular to a resource optimization method for structured query and a database query system. Background technique [0002] The artificial intelligence computing platform is a service platform that can provide users with training, inference tools, and data storage and processing services required for artificial intelligence training. The artificial intelligence computing platform generally provides support through one or more data centers. Each data center can be composed of multiple servers, and each server provides external services through the network. [0003] The database query system of the artificial intelligence computing platform is a service system that provides users who use the artificial intelligence computing platform to query data in the database of the data center. At present, in the database environment, the most commonly used database query ...

AI Technical Summary

Problems solved by technology

However, even if the hardware modules can be configured with different amounts of parallelism, the amount of parallelism of the hardware modules will be fixed after configuration. Since it is impossible to predict the composition of multiple pipeline segments of each SQL query statement in the multi-path query, it will still be There is a mismatch between demand and resources, resulting in waste of resources and poor performance

[0005] The existing method to solve the mismatch between the configuration and requirements of hardware modules is to reserve a large enough amount of parallelism for each hardware module, but due to limited resources, it is impossible to reserve the maximum amount of parallelism for each hardware module , and the number of pipeline segments in each stage of the query statement is also unknown, so it is impossible to guarantee that the maximum number of parallelism reserved can meet the demand of processing quantity. Therefore, it is still impossible to effectively solve the waste of hardware module resources and low processing efficiency The problem

Images