A multi-tenant cloud platform task scheduling method supporting bandwidth guarantee

Abstract

The invention relates to a task scheduling method for a multi-tenant cloud platform supporting bandwidth guarantee, and belongs to the field of network virtualization. The tenant makes a request for virtual resources required by the task according to its own business needs, and sends the request to the cloud platform; after receiving the task request from the tenant, the cloud platform retrieves the number of virtual machines for the existing task and the allocation location, and according to the new task According to the virtual resource requirements and the virtual resource allocation of existing associated tasks, a resource allocation network is constructed. Calculate the number of virtual resources actually required for scheduling the task, select the optimal solution, and finally determine whether to allocate virtual resources. According to the virtual resource request of the tenant task, the present invention can calculate the task scheduling scheme that occupies the least resources according to the resource usage of the existing cloud platform, so as to improve the resource utilization rate of the cloud platform.

Description

technical field [0001] The invention relates to a multi-tenant cloud platform task scheduling method supporting bandwidth guarantee, belonging to the field of network virtualization. Background technique [0002] With the emergence of public data center cloud platforms represented by Amazon EC2, more and more enterprises migrate their business from local data centers to cloud platforms. In order to support a large number of users to share hardware (such as CPU, storage, bandwidth, etc.) resources, cloud platform operators usually use multi-tenant technology to abstract physical hardware resources into virtual resources (such as virtual machines). Each user (also known as a tenant) can rent virtual resources of different sizes according to its own business needs, and pay the operator according to the amount of resource usage. Therefore, the multi-tenant cloud platform has two advantages: first, it can serve a large number of tenants, improve the utilization rate of physical ...

AI Technical Summary

Problems solved by technology

[0003] However, the main challenge faced by multi-tenant platforms is resource isolation

Existing commercial cloud platforms can guarantee various computing and storage resources (such as CPU, memory, and storage space) required by tenants, but do not provide any form of network bandwidth guarantee

As a result, many tenants compete for limited network bandwidth resources, resulting in unpredictable completion time of tasks running on the virtual network, seriously affecting the experience of end users using their services

In the long run, this will weaken users' enthusiasm for migrating their business to the cloud platform, and hinder the large-scale promotion and commercial use of the cloud platform to a certain extent.

Images