Resource scheduling method and device, electronic equipment and storage medium

Abstract

The invention relates to a resource scheduling method and device, electronic equipment and a storage medium. The method comprises the steps of obtaining a target scheduling task; according to a resource condition required for running the target scheduling task, screening out a first candidate node from the computing nodes, then screening out a second candidate node from the first candidate node according to the expected power consumption required by the operation of the target scheduling task and the power consumption information of the cabinet where the first candidate node is located, and processing the target scheduling task through the determined second candidate node. On the basis of the mode, power consumption information of a cabinet and expected power consumption required by operation of a target scheduling task are jointly used as conditions for screening nodes for processing the target scheduling task; the operation power consumption of the cabinet can be controlled to meet the rated power consumption limitation of the cabinet, so that the server downtime caused by the overload of the cabinet power consumption is avoided, the placement of the server in the cabinet does not need to be adjusted, and the maximization of the space utilization rate of the cabinet is realized while a large amount of labor cost is saved.

Description

Technical field [0001] This application relates to the field of computer technology, and in particular to a resource scheduling method, device, electronic equipment and storage medium. Background technique [0002] Kubernetes, K8s for short, is an open source, used to manage containerized applications on multiple hosts in the cloud platform. The goal is to make the deployment of containerized applications simple and efficient. Since the launch of K8s, it is based on the underlying architecture of the K8s cluster design. It is more popular and easier to deploy software, which truly realizes "one time deployment, direct operation". [0003] Currently, when scheduling containers in a K8s cluster, idle computing nodes are usually selected to process scheduling tasks. In order to avoid overloading the power consumption of the cabinet, the load situation of the cabinet is usually artificially estimated and the server placement plan is made in advance. When the load changes, the placemen...

AI Technical Summary

Problems solved by technology

The existing container scheduling methods have the following deficiencies: (1) only computing resource factors are considered during scheduling, and the power consumption factor of the cabinet is ignored, which leads to overloading of the power consumption of the cabinet and server downtime; (2) in the When the existing computing nodes cannot meet the scheduling conditions, the scheduling will be stopped, and the processing method for the tasks that fail to be scheduled is not reasonable enough; (3) Power consumption estimation requires a large labor cost, and it is difficult to take into account the maximization of cabinet space utilization and server Minimize the risk of downtime. On the one hand, it is hoped to store as many servers as possible in the cabinet. Once the artificially estimated power consumption is wrong, it may cause server downtime. On the other hand, in order to reduce the risk of server downtime, it is often considered to reduce a cabinet The number of servers that can be stored, so that the cabinet cannot be fully utilized

Images