Virtual machine scheduling method based on coexisting coefficient balance

Abstract

The invention discloses a virtual machine scheduling method based on coexisting coefficient balance. The virtual machine scheduling method comprises the following steps: based on pressure values and tolerance degrees of VCPUs (Virtual Central Processing Units) running on a multi-core physical server, calculating to obtain the pressure coefficients and tolerance coefficients of the VCPUs, calculating the existing coefficients of the VPCUs on shared caches and the coexisting coefficients of the shared caches according to the pressure coefficients and the tolerance coefficients, judging whether the coexisting coefficients of the shared cache in a virtual machine system are in balance or not according to the coexisting coefficients and set thresholds; when the coexisting coefficients are not in balance, performing VCPU scheduling according to the coexisting coefficients of the VCPUs, so as to keep the coexisting coefficients of the shared caches in balance. By adopting the virtual machine scheduling method, the problem that virtual machines compete with one another to obtain shared cache resource, a memory controller and memory bandwidth resource in a multi-core processor is effectively solved, the situation that the virtual machines compete for and use the shared resource is identified, the operation environment of the virtual machines is optimized, and the service quality of the virtual machines is improved.

Description

technical field [0001] The invention relates to the technical field of virtual machines, and more specifically, to a virtual machine scheduling method based on co-occurrence coefficient balance. Background technique [0002] System virtualization technology is one of the basic core technologies in cloud computing applications. In the current cloud computing data center, most applications are deployed in virtual machines, and the virtual machines run on physical servers with commercial multi-core architecture. The use of system virtualization technology enables different applications to run in their own virtual machines, which greatly facilitates the deployment and management of application services, ensures the isolation and security between applications, and improves the The overall operational efficiency of the data center. However, the existing virtualization technology cannot well solve the problem of resource competition when multiple virtual machines share the underl...

AI Technical Summary

Problems solved by technology

In the second method, the hardware logic of the shared cache needs to be modified, and the implementation cost is too high

Images