Dispatching method of virtual machine under multi-core environment

Abstract

The invention provides a dispatching method of a virtual machine under a multi-core environment. The dispatching method divides zones for CUP resources according to dispatching strategy types when a system is started, monitors the CPU loading condition of all divided zones in real time when the system operates, and dynamically adjusts the size of the CPU resources in the divided zones. The dispatching method carries out dispatching in the same divided zone by using the same dispatching strategy, thus improving dispatching efficiency, achieving the purpose of resource loading balance by dynamically adjusting the size of the CPU resources of the divided zone, realizing the full utilization of resources and reducing waste of resources.

Description

technical field [0001] The invention relates to a virtual machine scheduling method, in particular to a virtual machine scheduling method in a multi-core environment, and belongs to the technical field of computer operating systems and virtualization. Background technique [0002] With the advancement of multi-core processor technology, the number of processor cores integrated on a single processor is increasing year by year, and the computing power of servers is getting stronger and stronger. At present, general servers have at least 4-16 cores. The increase in processor cores makes a single server More applications can be gathered on it. At the same time, based on server virtualization technology, it has gained unprecedented attention. The basic idea of ​​virtualization technology is to manage the underlying hardware resources through virtual machine monitor (VMM) software, and provide multiple isolated virtual hardware execution environments ( VM), so that multiple diffe...

AI Technical Summary

Problems solved by technology

They have their own advantages and disadvantages, but they fail to consider the scheduling of different application types

For example, the BVT (Borrowed Virtual Time) scheduling algorithm is a fairness-first scheduling algorithm, so that the VCPU (virtual processor) with a large load and the VCPU with a small load can obtain almost the same amount of CPU time. The real CPU time of the VCPUs is not enough, and the real CPU time allocated to the VCPUs with small loads is surplus, resulting in a waste of CPU resources. The CPU resources are not utilized to the greatest extent, and the different needs of different applications are not considered; while SEDF It solves the problem of BVT, but it can only perform SEDF scheduling for each CPU separately. In the case of multi-core, this algorithm cannot achieve load balancing well, and it also ignores the different requirements of different applications for resources; finally, the Credit scheduling algorithm, CPU time can be allocated to each virtual CPU in a fair and efficient manner; each physical CPU can also be allocated to each virtual CPU by SMP to achieve load balancing; however, the same Credit scheduling algorithm does not consider the type of application as a scheduling consideration aspect.

[0008] Therefore, with the current development of server multi-core technology and the increasing number of application services on the server, a single virtual machine scheduling algorithm can no longer meet the needs of effective resource utilization and efficient operation of applications.

A scheduling algorithm may be effective for one type of application, but when multiple types of applications are scheduled together, resource allocation will be unreasonable. For example: BVT algorithm is more suitable for real-time applications, but when real-time applications and batch applications are in Together, there will be an unreasonable allocation of resources

Images