Resource allocation method and device, computer device and storage medium

Abstract

The invention provides a resource allocation method and device, a computer device and a storage medium. The method comprises the steps of obtaining a to-be-executed target task; determining a total demand resource corresponding to the target task; when the residual resources in the first resource pool to which the target task belongs are less than the total demand resources, determining demand gap resources corresponding to the target task; and on the basis of the demand gap resources, allocating available resources in a second resource pool are allocated to obtain outer-region borrowing resources, wherein the outer-region borrowing resources and the remaining resources are used for executing the target task, and the second resource pool and the first resource pool have the same role attribute and different region attributes. By adopting the method, the resource utilization rate can be improved, and the task execution efficiency can also be improved.

Description

technical field [0001] The present application relates to the technical field of resource scheduling, and in particular to a resource allocation method, device, computer equipment, and storage medium. Background technique [0002] Resource scheduling is a resource balancing strategy in the field of big data computing. Its main purpose is to uniformly control and allocate resources for different computing tasks, so that the resources of different tasks can be isolated without interfering with each other, thereby promoting resource utilization. Maximize and perform tasks more efficiently. [0003] However, the existing resource scheduling methods generally adopt the first-in-first-out strategy, and it is very easy for someone to submit multiple SQL (Structured Query Language) within a short period of time, resulting in the entire cluster resources being monopolized by one person. At the same time, since it is impossible to judge which of the two SQLs, that is, A and B resourc...

AI Technical Summary

Problems solved by technology

[0003] However, the existing resource scheduling methods generally adopt the first-in-first-out strategy, and it is very easy for someone to submit multiple SQL (Structured Query Language) in a short period of time, resulting in the entire cluster resources being monopolized by one person

At the same time, since it is impossible to judge which of the two SQLs, that is, A and B resource requirements, is more suitable for the current cluster resources, the tasks after resource allocation can only be executed in sequence based on the SQL submission time, so the following situations are also very likely to occur: Tasks submitted earlier cannot be executed due to resource requirements that do not match the current cluster, and other tasks that are submitted later and whose resource requirements match the current cluster are blocked, reducing task processing efficiency

[0004] Therefore, the existing resource scheduling method has the technical problem of low resource utilization due to unreasonable resource scheduling.

Images