Cluster node processing method and system, equipment and medium

Abstract

The embodiment of the invention provides a cluster node processing method and system, equipment and a medium. The method comprises the steps: generating a first scheduling task based on a first calculation request; waiting to execute the first scheduling task according to the sequence of the first scheduling task in a scheduling queue; if the first scheduling task waits for timeout due to insufficient node resources, triggering a node capacity expansion operation; and executing the first scheduling task so as to execute calculation operation on the expanded node. According to the scheme, after the calculation operation fails to be executed through the first scheduling task, whether node resources in the current cluster are sufficient or not is further determined; and if the node resources are insufficient, the nodes in the cluster are expanded, and the node expansion can be performed before the problem of insufficient nodes occurs, so that the influence on normal execution of related calculation work due to insufficient node resources is avoided or alleviated.

Description

technical field [0001] The present application relates to the computer field, and in particular to a cluster node processing method, system, device and medium. Background technique [0002] Cloud native technology was born. It mainly represents the theory and technology of container technology, kubernetes technology, micro-service, service network and declarative API. It can be used in new dynamic environments such as public cloud, private cloud and hybrid cloud. Run elastically scalable applications. [0003] With the continuous change of user needs, the online load of node resources in the cluster will also be in a dynamic change, so we often face the dilemma that the capacity planning of node resources cannot meet the online load changes. When applications obtain node resources on demand, if they are allocated according to the upper limit of node resource requirements, a large number of node resources will be left idle and wasted. If you allocate according to the lower ...

AI Technical Summary

Problems solved by technology

When an application obtains node resources on demand, if it is allocated according to the upper limit of node resource requirements, a large number of node resources will be idle and wasted

If you allocate according to the lower limit of node resource requirements, although the cost will be reduced, when the access peak arrives, the application will often be severely blocked, especially when the number of nodes is seriously insufficient and the node resources cannot be effectively expanded in time, which may lead to a system crash.

Images