Method of resource allocation based on SQL sentence runtime and computer equipment

Abstract

The invention provides a method of resource allocation based on the SQL sentence runtime. The method comprises the steps of collecting executing record IDs of SQL histories, SQL sentences and the runtime; according to the executing record IDs, obtaining resources allocated when the SQL sentences are executed to form resource vectors R, analyzing the SQL sentences to obtain operators, table names and column names, querying metadata information according to the table names and the column names, obtaining data information and corresponding the data information to the operators to form operation vectors OP; regarding the runtime as a target variable and R and OP as predicting variables, using a machine learning algorithm to build a predicting module, and training a predicting model; submittingthe SQL sentences and the resource vector values R, and predicting the needed executing time through the predicting model; according to the executing time, selecting resource allocation modules. Theinvention further provides computer equipment. By means of the method of the resource allocation based on the SQL sentence runtime and the computer equipment, the system expense is effectively lowered, the burden of a user is lowered, and the user experience is improved.

Description

technical field [0001] The invention relates to a resource allocation method and computer equipment based on the running time of SQL statements. Background technique [0002] There are two types of distributed computing resource allocation modes: batch mode - computing resource (CPU, MEM, etc.) allocation and environment initialization are performed when the task is submitted, resources are recovered and the environment is canceled when the task ends, and resources need to be reallocated when the task is submitted again and initialize the environment. Interactive mode - computing resource allocation and environment initialization are performed when the first task is executed, and subsequent tasks will continue to reuse the allocated resources and environment configuration. The disadvantage of the batch mode is that each time a task is submitted, there is an overhead of allocating resources and initializing the environment. The advantage is that resources are recovered and v...

AI Technical Summary

Problems solved by technology

The usual practice is to leave it to the user to make their own judgment. The consequence is that the user has a heavy burden (need to understand a lot of knowledge about the execution engine) and inaccurate judgment (many determinants and complex relationships), resulting in high system overhead or waste of resources.

Images