Virtualization cluster resource scheduling method based on resource rationalization application

Abstract

The invention belongs to the technical field of virtualization and cloud computing, and particularly relates to a virtualization cluster resource scheduling method based on resource rationalization application. Compared with the prior art, the comprehensive heuristic efficient resource scheduling method based on the virtualization cluster is adopted. Firstly, in the resource scheduling triggeringstage, on the basis of considering CPU, memory and network multi-threshold triggering conditions, dynamic comprehensive load conditions are added, and the resource utilization condition of a server can be more effectively balanced. Then in a virtual phase selection, for the scheduling problem triggered by the comprehensive load condition, a computing method of virtual machine resource matching isintroduced, a reasonable virtual machine is selected for scheduling, resources of a physical server can be more fully utilized in the scheduling process, and the utilization rate of the resources of the physical server is increased. Meanwhile, frequent migration of the virtual machine caused by oscillation of a load peak value is reduced, and the purpose of further improving the performance of a processor is achieved.

Description

technical field [0001] The invention belongs to the technical field of virtualization and cloud computing, and in particular relates to a virtualized cluster resource scheduling method based on resource rationalization application. Background technique [0002] Virtualization technology is a technology that can break the structural barriers of traditional physical machines, abstract hardware resources, and run multiple virtual machines on one physical machine without interfering with each other, thereby improving system availability and saving hardware costs. . Virtualization cluster technology is a technology that combines multiple physical servers into a virtual resource pool and provides resources externally for use by multiple virtual machines, which can effectively improve the resource utilization rate of physical servers. In a virtualized cluster, providing high performance and high availability to the system is mainly achieved through the scheduling of cluster resour...

AI Technical Summary

Problems solved by technology

[0006] Static resource scheduling is implemented by manpower, and the operation is cumbersome, requiring a lot of manpower and material resources, and the stability of the system is poor, and the resource utilization rate is low

And the static resource scheduling has hysteresis, which seriously reduces the performance of the whole system

[0007] Therefore, although the current mainstream dynamic scheduling method considers the resource load from multiple individual thresholds of CPU, memory, and network, it does not consider the comprehensive load of the entire server resource, resulting in the resource utilization of the server not exceeding the scheduling rate. Threshold, but the overall load of the entire server resource is too high, affecting the performance of the virtual machine

At the same time, the mainstream scheduling algorithm is not accurate enough to evaluate the peak load of the host, and the peak load fluctuation of the host leads to frequent migration of virtual machines, which also increases the waste of resources.

Images