Multi-tenant container cloud platform virtual computing resource adaptive scheduling method and system

Abstract

The invention belongs to the field of network technology, and discloses a multi-tenant container cloud platform virtual computing resource self-adaptive scheduling method and system. Through a unified cloud computing service portal, user container resource requests are unified to the cloud computing service portal and delivered to The resource allocator; the resource allocator interacts with the D-Agent of each data center, and uses the method based on the combined auction to decide the deployment plan of the application, which is then implemented by each data center; during operation, the D-Agent collects the information of each container cluster Relevant performance information and make dynamic adjustments; make multiple data centers based on container technology cooperate with each other. During the operation of the present invention, the D-Agent collects the relevant performance information of each container cluster and makes dynamic adjustments to ensure the performance of the application and the quality of service of the system; enables multiple data centers based on container technology to cooperate with each other, deploying virtual machines of users more simple, convenient and efficient.

Description

technical field [0001] The invention belongs to the field of network technology, and in particular relates to a method and system for self-adaptive scheduling of virtual computing resources on a multi-tenant container cloud platform. Background technique [0002] At present, the existing technologies commonly used in the industry are as follows: [0003] The current research on static scheduling of computing resources in data centers mainly focuses on the placement and migration of virtualized resources, and the collaborative adaptive configuration with changes in user needs and application system parameters, hoping to reduce resource fragmentation and improve computing resource utilization. [0004] In a complex and transient cloud computing environment, when user applications are deployed to the data center, if only static resource allocation methods are used, it is difficult to guarantee the performance of multi-tenant applications. Therefore, when the tenant application...

AI Technical Summary

Problems solved by technology

[0006] (1) Poor management of computing resources in multiple data centers and poor management of computing resources in servers

[0007] (2) In a complex and transient cloud computing environment, when user applications are deployed to the data center, if only static resource allocation methods are used, it is difficult to guarantee the performance of multi-tenant applications

Therefore, when the tenant application load changes suddenly, there is no dynamic container resource scheduling algorithm to ensure that the service quality and user service level agreement are met

[0008] (3) Although most data center dynamic scheduling algorithms can accept indeterminate and indeterminate resource constraints, they lack the verification of the computational feasibility of the algorithm

[0009] (4) Most data center computing resource scheduling is usually based on the assumption that the status of the data center and network connections are fully or partially known and that the data center has sufficient servers and can be switched at will, but the resources of the data center in the real cloud computing environment The state is changing rapidly, unpredictable, and there is a lack of consideration of the situation when the data center is overloaded

Combined with functions and application scenarios, the commands contained in the container can be divided into types related to environment information, system operation and maintenance, log information, and Hub services. The inconsistency of front and back commands and parameters can easily lead to system fault tolerance complexity improvement and a substantial decrease in system performance

[0013] Difficult to measure sudden traffic and redundant resources: new tenants cannot accurately predict the amount of user service requests. Once there is a large fluctuation in user demand, the traditional processing mode is to quickly expand the capacity of such unplanned service requests. And verified online, but this processing mode is not only time-consuming, but also needs to be prepared in advance to deal with sudden traffic, which can easily cause a lot of resource redundancy

Images