Multi-tenant network optimization method and system, computer device and memory medium

Abstract

The invention provides a multi-tenant network optimization method. The method comprises the steps of establishing a plurality of virtual machines and a plurality of tenants, wherein the plurality of tenants at least comprise a first tenant and a second tenant, and the plurality of virtual machines at least comprises a first virtual machine and a second virtual machine; configuring the first virtual machine as the virtual machine of the first tenant, configuring the second virtual machine as the virtual machine of the second tenant; configuring a network of the first tenant as a self-service network and configuring the network of the tenant as a shared external supplier network; and deploying a virtual router and carrying out connection on the self-service network and the shared external supplier network through the virtual router. Personalized setting demands of the tenants for a private cloud are satisfied and the demands of the tenants for the network performance are improved. The invention also discloses a multi-tenant network optimization system, a computer device and a memory medium.

Description

technical field [0001] The present invention relates to the technical field of cloud computing, in particular to a multi-tenant network optimization method and system, computing equipment, and computer storage media. Background technique [0002] With the development of cloud computing, data center virtualization has become a trend. The private cloud service project is mainly used to solve multi-data center resource virtualization and improve resource utilization. The project is developed based on the open source cloud platform OpenStack. At present, the common multi-tenant network architecture of OpenStack is the self-service network (Self-service network). [0003] Please refer to figure 1 , which is the existing self-service network architecture. Although the self-service network has rich functions and can customize the virtual private cloud (Virtual Private Cloud, VPC), but because the network ingress and egress traffic of all tenants’ virtual machines (Virtual Machine,...

AI Technical Summary

Problems solved by technology

[0003] Please refer to figure 1 , which is the existing self-service network architecture. Although the self-service network has rich functions and can customize the virtual private cloud (Virtual Private Cloud, VPC), but because the network ingress and egress traffic of all tenants’ virtual machines (Virtual Machine, VM) needs to Through the virtual router (Virtual Router), the virtual router is concentrated on the network node, and the virtual router finally sends or receives the data packet through the physical network card of the network node, so the network performance is limited by the network card bottleneck of the network node, and the self-service network exists network Performance bottleneck, not suitable for tenants with high network performance requirements

[0004] Please refer to figure 2 , for the existing provider network architecture, the inbound and outbound traffic does not need to pass through the virtual router on the network node, but the provider network itself does not support tenants to create a virtual private cloud, that is, all virtual machines under the same tenant cannot be formed into a private cloud

[0005] To sum up, the existing network cannot combine the advantages of self-service network and supplier network in a multi-tenant environment, so as to achieve both high network performance requirements and customizable private cloud creation

Images