WO2018001004A1

WO2018001004A1 - Docker based cloud platform control method and apparatus

Info

Publication number: WO2018001004A1
Application number: PCT/CN2017/085626
Authority: WO
Inventors: 童遥; 彭勇; 孙自广; 蒋天超; 申光
Original assignee: 中兴通讯股份有限公司
Priority date: 2016-06-27
Filing date: 2017-05-24
Publication date: 2018-01-04
Also published as: CN107547596A; CN107547596B

Abstract

Disclosed are a Docker based cloud platform control method and apparatus. The method comprises: receiving a resource deployment request input by a user (S402); when the resource deployment request is a dynamic scheduling request, according to the resource deployment request, acquiring load information about each host on the cloud platform and application information needing to be scheduled (S404); and according to a pre-set scheduling policy, selecting a host with the minimum load information, and deploying a copy of the application information needing to be scheduled on the host (S406). By means of re-deploying application information according to load information on each host, and adjusting the application information from a host with a high load to a host with a low load, the extension of a host application and the reasonable allocation of cloud platform resources are realized, thereby improving the availability of the cloud platform, realizing the running of the application information in different running environments, and solving the problem of poor compatibility caused by the fact that application information can only run in a single running environment in the prior art.

Description

Docker-based cloud platform control method and device

Technical field

The present invention relates to the field of cloud platform technologies in cloud computing data, and in particular, to a Docker-based cloud platform control method and apparatus.

Background technique

As Internet technology continues to deepen people's lives, Internet companies are competing to launch new products to meet the needs of existing users and attract new users. Cloud services have become one of the key points in the practice of cloud computing, and it will be used for application operations. IT resources and infrastructure are provided to users in a service-oriented manner, including middleware services, information services, connectivity services, integrated services, and messaging services. In order to realize platform services, the industry has proposed Platform as a Services (hereinafter referred to as PaaS). Through the PaaS platform, application developers can easily host applications and services on the platform without concern for infrastructure settings such as their underlying hardware configuration and operating environment. For developers, PaaS greatly reduces the overhead and pain of IT deployment, providing resources to applications on demand to make them more scalable. Driven by the open PaaS cloud computing platform, developers can focus on application innovation without worrying about the cumbersome tasks of the operating environment, so that applications on the Internet will become increasingly rich.

At present, Google and Sina Internet have launched their own PaaS cloud platform. Google has launched the Google App Engine (GAE) cloud service, providing application developers with a web application running and management platform that is simple to develop, easy to deploy, and scalable. Sina's Sina App Engine (SAE) is a distributed Web service development and operation platform. As the first public PaaS cloud platform in China, SAE chose PHP, the most popular web development language in China, as its preferred support language. Provides a range of basic capabilities such as distributed storage, distributed caching, etc. for developers to use. However, the above-mentioned Sina and Google PaaS cloud platform still has shortcomings, that is, the application environment is single, SAE only supports PHP applications in the initial stage of launch, GAE only supports Java and Python applications, which leads users to expand. The issue of compatibility is not conducive to the rapid expansion and versatility of the PaaS platform, and can not meet the diversified needs of application providers. Therefore, it is imperative to improve the usability and scalability of the platform.

Summary of the invention

The Docker-based cloud platform control method and apparatus provided by the embodiments of the present invention are used to solve the problem that the PaaS cloud platform application in the prior art can only operate in a specific running environment, and the PaaS platform cannot achieve rapid expansion and versatility. technical problem.

To solve the above technical problem, an embodiment of the present invention provides a Docker-based platform control method, including:

Receiving a resource deployment request input by a user;

When the resource deployment request is a dynamic scheduling request, the load information of each host on the cloud platform and the application information that needs to be scheduled are acquired according to the resource deployment request, where the application information is application information on a host with the highest load information. ;

Selecting a host with the lowest load information according to a preset scheduling policy, and deploying a copy of the application information that needs to be scheduled to the host.

In one aspect, an embodiment of the present invention further provides a Docker-based platform control apparatus, including:

a receiving module, configured to receive a resource deployment request input by a user;

And obtaining, by the acquiring module, the load information of each host on the cloud platform and the application information that needs to be scheduled according to the resource deployment request, where the application information is the highest load information. Application information on the host;

Selecting a module, and setting a host with the lowest load information according to a preset scheduling policy;

The resource scheduling module is configured to deploy the copy of the application information that needs to be scheduled to the host.

In another aspect, an embodiment of the present invention further provides a Docker-based platform control apparatus, including: a resource scheduling subunit and a platform branch unit;

The platform support unit is configured to receive a resource deployment request input by a user;

The resource scheduling unit is configured to: when the resource deployment request is a dynamic scheduling request, select a host with the lowest load information according to load information of each host on the cloud platform, and deploy a copy of the application information that needs to be scheduled on the host. The application information is application information on a host with the highest load information.

The embodiment of the invention further provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the foregoing Docker-based platform control method.

The beneficial effects of the invention are:

A Docker-based cloud platform control method, apparatus, and computer storage medium according to an embodiment of the present invention, the method includes receiving a resource deployment request input by a user, and when the resource deployment request is a dynamic scheduling request, according to the resource deployment request Obtaining load information of each host on the cloud platform and application function information to be scheduled, selecting a host with the lowest load information according to a preset scheduling policy, and deploying a copy of the application information on the host; The load information redeploys the application information and adjusts the application information from the high-load host to the low-load host, thereby realizing the expansion of the host application and the reasonable allocation of the cloud platform resources, thereby improving the availability of the cloud platform. Further determining whether the host is in a low-load state according to the load information of each host, and if so, adjusting the application information that needs to be scheduled on the host in the high-load state to the host with low load by deploying the copy, thereby Implemented application information in different operating environments OK, it solves the prior art application of information can only be run under a single operating environment, resulting in poor compatibility issues.

DRAWINGS

1 is a schematic structural diagram of a Docker-based cloud platform control apparatus according to a first embodiment of the present invention;

2 is another schematic structural diagram of a Docker-based cloud platform control apparatus according to a first embodiment of the present invention;

FIG. 3 is still another schematic structural diagram of a Docker-based cloud platform control apparatus according to a first embodiment of the present invention; FIG.

FIG. 4 is a flowchart of a Docker-based cloud platform control method according to a second embodiment of the present invention.

detailed description

The embodiments of the present invention are further described in detail below with reference to the accompanying drawings.

First embodiment:

In order to realize that the application information can be run in different operating environments corresponding to different containers of the PaaS cloud platform, thereby improving the expansion performance and availability of the PaaS cloud platform, the embodiment of the present invention provides a cloud platform control device based on Docker, the device Based on Docker's open source application container engine, developers can package their applications and dependencies into a portable container, then publish them to any popular Linux machine, or virtualize them to implement PaaS cloud. For the expansion of the platform and the improvement of compatibility, please refer to FIG. 1. FIG. 1 is a schematic structural diagram of a Docker-based cloud platform control apparatus according to an embodiment.

The Docker-based cloud platform control apparatus 1 provided by the present implementation includes a receiving module 11, an obtaining module 12, a selecting module 13, and a resource scheduling module 14, wherein:

The receiving module 11 is configured to receive a resource deployment request input by a user, where the resource deployment request includes a first scheduling request and a dynamic scheduling request;

The obtaining module 12 is configured to acquire load information of each host on the cloud platform and application information that needs to be scheduled according to the resource deployment request when the resource deployment request is a dynamic scheduling request;

The selecting module 13 is configured to select a host with the lowest load information according to a preset scheduling policy;

The resource scheduling module 14 is configured to deploy a copy of the application information that needs to be scheduled to the host.

In this embodiment, when the user develops the application through the PaaS cloud platform, the infrastructure creation of the application is first required, and the infrastructure for creating the application can directly create or configure the infrastructure by inputting the resource deployment request. When the resource deployment request received by the receiving module 11 is the initial scheduling request, the resource scheduling module 14 acquires the physical server on the host according to the resource deployment request, and filters the physical server according to the request policy, and rejects the host. The physical server is used, and then the resource is allocated for the host according to a preset scheduling policy.

In this embodiment, when the resource deployment request is a dynamic scheduling request, the obtaining module 12 acquires load information of each host on the cloud platform according to the resource deployment request, and the load information may be an available physical server of the host, and It may be application information of a container set on the physical server, and the selection module 13 selects the lowest and highest load information according to the load information, and then deploys the application in the host with the highest load information to the host with the lowest load information. Specifically, the application is deployed in a container on the host with the lowest load information, and a copy of the application is executed through the container, thereby implementing scheduling of the application.

Because the running environment of each host on the cloud platform is different, the application of the application information to other hosts may cause incompatibility. For the problem, the cloud platform control device 1 provided in this embodiment further sets a general container. 15. The universal container 15 is a Docker container, which can automatically create a corresponding application running infrastructure and IT resources according to the deployment request, without requiring the user to manually redeploy. Specifically, the universal container 15 performs the container according to different requests. Encapsulate and package a variety of different web containers, such as Docker-Jetty, Docker-Tomcat, Docker-PHP5, and more.

In order to realize the operation of the scheduled application correctly on the scheduled host, in addition to the provision of the container, the routing of the application information needs to be adjusted from the original pointing to the container after the scheduling, specifically, In this embodiment, a routing unit 16 is provided on the cloud platform control device 1, and the routing unit 16 is configured to perform all the received scheduling requests. Routing, re-adjusting the IP correspondence between the application and the container.

In this embodiment, as shown in FIG. 2, another schematic structural diagram of the Docker-based cloud platform control apparatus 1 provided in this embodiment includes a platform support unit 21 and a resource scheduling unit 22, and the platform supports The unit 21 is configured to receive a resource deployment request input by the user, and the resource scheduling unit 22 is configured to: when the resource deployment request is a dynamic scheduling request, select a host with the lowest load information according to load information of each host on the cloud platform, and A copy of the application information that needs to be scheduled is deployed on the host.

In this embodiment, the load information includes load information of the container, and the resource scheduling unit 22 is configured to select a container with the lowest load information from the host according to the preset scheduling policy, and on the container. Deploy a copy of the application information.

As shown in Figure 2, the cloud platform control apparatus 1 provided by this embodiment further includes a routing control unit 23 and a general container setting unit 24;

The universal container setting unit 24 is configured to set, on the host, a general container for running a copy of the application information, where the universal container is a container including at least one programming language running environment;

The routing control unit 23 is configured to adjust the route of the application information to a container corresponding to the copy of the application information after the resource scheduling unit deploys the copy of the application information.

In this embodiment, the scheduling of the application is performed according to the load information of each host on the cloud platform, and the resource scheduling unit 22 further includes determining, according to the acquired load information of the host on the cloud platform, whether the load is in a high load state, and if so, The alarm information is sent; if not, the application information on the host with lower load information is deployed to other hosts, and the application information on the host is deleted.

In this embodiment, each module in FIG. 2 can also communicate with each other by setting a message bus, and the specific connection relationship is as shown in FIG. 3.

As shown in FIG. 3, the platform support unit 21 provides some basic modules for the cloud platform in addition to receiving the request, and provides some basic capabilities for the cloud platform, such as setting the API Server 211 (Application Programming Interface Server, server application program interface). The API Server 211 is used as an external interface of the entire device, receives the request through the interface, and sends the received request to the resource scheduling unit 22, and the server application interface 211 exposes it by abstracting the restful-style HTTP interface. On the device, the user is convenient to call the interface to implement the request receiving and reasonable resource allocation; in addition, the platform supporting unit 21 is also provided with Memcached, which provides session sharing between different containers, and the Cache provides a high performance distributed. Object caching service, Mysql provides a native relational database, and Freedisk provides a small file storage service.

In this embodiment, the resource scheduling unit 22 serves as a resource scheduling core of the cloud platform, and is mainly used for reasonably allocating IT resources and infrastructure provided by the cloud platform according to the request, specifically, for running the application. The container is deployed and operated on the host. In this embodiment, the resource deployment request includes a first scheduling request and a dynamic scheduling request, where the initial scheduling request is a scheduling of resources in the cloud platform when the user first creates an application, including The creation, deletion, modification, and other operations of the resource are created to set the corresponding running environment infrastructure and allocate IT resources for the newly added application; deleting refers to shrinking the container, and scheduling the application in the container with lower load information to After the other containers are deleted, the original container is deleted; the modification refers to reallocating other resources such as IT resources to the container according to the scheduling request.

The dynamic scheduling request is a scheduling in the running process of the system, and the scheduling is performed on a host or a container that is too high or too low in the running process, mainly by monitoring the state of the application and the state of the container, according to the monitoring. The actual situation expands or shrinks the application.

In an actual application, the resource scheduling unit 22 specifically includes a primary resource scheduler 221, a dynamic resource scheduler 222, and a node agent 223, where

The initial resource scheduler 221 implements the transfer and coordination of all initial resource operation requests. The implementation manner is divided into two steps. The first step is to take out all the physical servers, and use different filters (filters) to filter the physical servers for different request policies. , get a list of available servers, the purpose of this step is to ensure the availability of the host. The second step is to introduce the weight and scoring concept. The initial resource scheduler calculates the score of each host by different weights, then sorts the host according to the score from high to low, and deploys multiple copies of the application (the default is 2) to On a host with a high score.

The dynamic resource scheduler 222 implements dynamic scheduling, dynamically increasing and reducing the number of application backend containers. After the dynamic resource scheduler obtains the load information of the host and the load information of the Docker container, the application copy expansion is completed according to the following policy: the host with the lowest load is selected according to the corresponding scheduling policy, and a copy of the application is deployed on the host, and the route is simultaneously routed. The subsystem sends a request to modify the direction of the route so that the newly added application replica can receive the request, thereby implementing load balancing. When all hosts in the cluster are close to a high load state, the expansion of the application replica is stopped and an alert message is sent to the administrator. Similarly, if the average access volume of all copies of an application is found to be too low, the application copy shrinks according to a certain strategy. In addition, during the process of copy expansion and contraction, regular defragmentation is added to ensure multiple applications of the same application. Under the premise that the replicas are distributed on different hosts, try to distribute the distributed applications to some hosts.

The node agent 223 runs on each computing node as a resident process, and periodically reports heartbeat data to the primary resource scheduler. The heartbeat data includes information such as CPU utilization, memory usage, and remaining space of the hard disk. On the other hand, it adapts the underlying Docker container and obtains the load data of each container on it, including CPU usage, memory usage, hard disk read and write status, and so on.

As shown in FIG. 3, the routing unit 24 provided in this embodiment includes a software router 241 (App Router), an Nginx controller 242 (Ngnix Controller), and a Tengine, where:

The software router 241 is the core control component of the PaaS cloud platform routing. After determining that the application on a host needs to be scheduled, and after setting up the container, the routing direction of the application needs to be adjusted, and the component is responsible for maintaining a copy. A routing table containing all the application routing policies, and controlling the Nginx to complete the routing of the application access request according to a specific task scheduling algorithm. The routing for implementing the above functions is mainly performed in two steps. The first step is to externally domain name to internal according to the routing request. The routing of the temporary domain name, that is, the IP address of the external IP to the host and the IP of the host to the physical server. The second step is the routing of the internal temporary domain name to the container, that is, the IP address from the physical server to the container IP, providing internal temporary The domain name can also be used to facilitate application developers to check and confirm the application after submitting the application, thereby implementing load balancing, health check, session sticking and the like among multiple application copies.

In this embodiment, before performing routing, the software router 241 further includes determining the validity of the request, determining whether the routing table needs to be updated after determining the legality, and notifying the Ngnix Controller (controller) to update the routing table if necessary. operating.

The Nginx controller 242 loads the most from the database after receiving the notification of the updated routing information of the software router 241. New routing information, and the corresponding configuration file is generated by Java's Velocity technology, replacing the current Nginx configuration file and restarting Nginx, so that the routing information takes effect in time.

Based on Nginx, Tengine adds advanced features and features to the needs of large-volume websites, such as consistent hashes, session persistence, health checks, and automatic offline and offline based on server status.

As shown in FIG. 3, in the embodiment, the universal container 23 is configured to set a running environment container for an application, and the general container 23 performs different deployment allocation according to different requests, according to a corresponding writing language and a compatible language of the application. The Web container is re-encapsulated. For example, the Java application can only be run on the Java container. The general container 23 is used to implement the fast package to provide the corresponding container, which improves the compatibility of the PaaS platform.

The general container 23 provided in this embodiment can specifically abstract two layers, which are an adaptation layer and a container layer respectively. The adaptation layer can collect information such as the host load and the like as a resident node (that is, a node agent). Aspects can be adapted to the underlying container. The container layer encapsulates different web containers based on Docker Api, such as Docker-Jetty, Docker-Tomcat, Docker-PHP5, and so on. In order to facilitate the platform administrator to quickly release and delete the container, the embodiment of the present invention also introduces a centralized image management module of the Image Server, which facilitates the release, replacement, and deletion of the image.

The Docker-based cloud platform control device provided by the embodiment of the present invention receives the resource deployment request input by the user through the receiving module. When the resource deployment request is a dynamic scheduling request, the acquiring module acquires load information of each host of the cloud platform according to the resource deployment request. The selection module selects the host with the lowest load information according to the preset scheduling policy, and deploys a copy of the application information on the host, and deploys the application information on the host with the highest load information to the host with the lowest load information, so that the application information is obtained. It can be scheduled between multiple hosts, which realizes the expansion and availability of the cloud platform, and also improves the utilization of cloud platform resources.

Further, the embodiment of the present invention further provides a corresponding container for the scheduled application information, so that the application information can be normally operated after being scheduled, and the PaaS cloud platform application in the prior art can only be run on a specific host. Running under, resulting in low compatibility, the universal container can implement container packaging in a variety of programming languages, providing corresponding containers for different application information, thereby improving the compatibility and scalability of the PaaS cloud platform.

Second embodiment:

Referring to FIG. 4, FIG. 4 is a Docker-based cloud platform control method according to an embodiment. The Docker is an open source application container engine, which allows developers to package their applications and dependencies into a portable container. And then released to any popular LINUX machine, can also achieve virtualization, the control process specifically includes the following steps:

S402. Receive a resource deployment request input by a user.

In this step, the resource deployment request received by the cloud platform control device includes a primary scheduling request or a dynamic scheduling request. When the user first creates the application, the received request should be the initial scheduling request, and the cloud platform control device should Allocating IT resources and setting up the infrastructure according to the initial scheduling request is implemented in two steps. The first step is to take out all the physical servers and use different filters (filters) to filter the physics for different request policies. The server gets a list of available servers. The purpose of this step is to ensure the availability of the host. The second step is to introduce weights and scoring concepts. Resource Schedule calculates the score of each host by different weights, then sorts the host's score from high to low, and deploys multiple copies of the application (default is 2 ) to the host with high scores.

S404, when the resource deployment request is a dynamic scheduling request, acquiring, according to the resource deployment request, load information of each host on the cloud platform and application information that needs to be scheduled, where the application information is on a host with the highest load information. Application information

When receiving a dynamic scheduling request, the cloud platform control device should dynamically increase the number of applications and reduce the number of application backend containers according to the request. After obtaining the load information of the host and the load information of the Docker application container, complete the application copy expansion according to the following policy: select the host with the lowest load according to the preset scheduling policy, deploy a copy of the application on the host, and send the copy to the routing unit at the same time. Request, request to modify the direction of the route, so that the newly added application copy can receive the request, thereby achieving load balancing. When all hosts in the cluster are close to a high load state, the expansion of the application replica is stopped and an alert message is sent to the administrator. Similarly, if the average access volume of all copies of an application is found to be too low, the application copy shrinks according to a certain strategy. In addition, during the process of copy expansion and contraction, regular defragmentation is added to ensure multiple applications of the same application. Under the premise that the replicas are distributed on different hosts, try to distribute the distributed applications to some hosts.

S406. Select a host with the lowest load information according to a preset scheduling policy, and deploy a copy of the application information that needs to be scheduled to the host.

In this embodiment, the load information acquired by the cloud platform control device may specifically be load information of the container, where the load information may be understood as application information; and the load is selected from the host according to the preset scheduling policy. The container with the lowest information and a copy of the application information deployed on the container.

The load information obtained by the cloud platform control device may specifically be the load information of the physical server on the host, where the load information may be understood as an application container; and the load information is selected from the host according to the preset scheduling policy. The lowest physical server and a copy of the application information is deployed on the physical server.

In this embodiment, after the replica of the application information that needs to be scheduled is deployed on the host, the method further includes: setting, on the host, a general-purpose container for running a copy of the application information, and sending a routing request Orienting the application information to a container corresponding to a copy of the application information, where the universal container is a container including at least one programming language running environment, so that the scheduled application can run normally, thereby implementing the PaaS cloud. Compatibility between different applications of the platform and the container.

In summary, the Docker-based cloud platform control method and apparatus provided by the embodiment of the present invention includes receiving a resource deployment request input by a user, and when the resource deployment request is a dynamic scheduling request, according to the resource deployment request. Obtaining load information of each host on the cloud platform and application function information to be scheduled, selecting a host with the lowest load information according to a preset scheduling policy, and deploying a copy of the application information on the host; The load information redeploys the application information and adjusts the application information from the high-load host to the low-load host, thereby realizing the expansion of the host application and the reasonable allocation of the cloud platform resources, thereby improving the availability of the cloud platform. The application information is run in different operating environments, which solves the problem that the application information in the prior art can only be operated in a single operating environment, resulting in poor compatibility.

Obviously, those skilled in the art should understand that the modules or steps of the above embodiments of the present invention can be implemented by a general computing device, which can be concentrated on a single computing device or distributed among multiple computing devices. On the network, optionally, they may be implemented by program code executable by the computing device, such that they may be stored in a computer storage medium (ROM/RAM, disk, optical disk) by a computing device, and at some In some cases, you can The steps shown or described are performed in a different order than that herein, or they are separately fabricated into individual integrated circuit modules, or a plurality of modules or steps thereof are fabricated as a single integrated circuit module. Therefore, the invention is not limited to any particular combination of hardware and software.

The above is a detailed description of the embodiments of the present invention in conjunction with the specific embodiments, and the specific embodiments of the present invention are not limited to the description. It will be apparent to those skilled in the art that the present invention may be made without departing from the spirit and scope of the invention.

Industrial applicability

The present disclosure is applicable to the cloud platform technical field in cloud computing data, to implement extension of the host application and reasonable allocation of the cloud platform resources, improve the availability of the cloud platform, and realize the application information running in different operating environments, and solve the present problem. In the technology, application information can only be run in a single operating environment, resulting in poor compatibility.

Claims

A Docker-based cloud platform control method, including:

Receiving a resource deployment request input by a user;

When the resource deployment request is a dynamic scheduling request, the load information of each host on the cloud platform and the application information that needs to be scheduled are acquired according to the resource deployment request, where the application information is application information on a host with the highest load information. ;

Selecting a host with the lowest load information according to a preset scheduling policy, and deploying a copy of the application information that needs to be scheduled to the host.
The Docker-based cloud platform control method according to claim 1, wherein the load information includes load information of the container, and the method further comprises: selecting the lowest load information from the host according to the preset scheduling policy. a container and a copy of the application information is deployed on the container.
The Docker-based cloud platform control method according to claim 2, wherein after the copy of the application information that needs to be scheduled is deployed on the host, the method further includes: setting, on the host, the application for running the application A generic container of copies of information, and a routing request to route the application information to a container corresponding to a copy of the application information, the universal container being a container including at least one programming language runtime environment.
The Docker-based cloud platform control method according to any one of claims 1 to 3, wherein after receiving the resource deployment request input by the user, the method further includes: determining that the resource deployment request is an initial scheduling request, according to the resource The deployment server obtains a physical server of each host of the cloud platform, performs weight setting on the physical server according to a preset resource deployment policy, obtains a priority order of each host, and applies the host according to the priority ranking. deploy.
A Docker-based cloud platform control device, including:

a receiving module, configured to receive a resource deployment request input by a user;

And obtaining, by the acquiring module, the load information of each host on the cloud platform and the application information that needs to be scheduled according to the resource deployment request, where the application information is the highest load information. Application information on the host;

Selecting a module, and setting a host with the lowest load information according to a preset scheduling policy;

The resource scheduling module is configured to deploy the copy of the application information that needs to be scheduled to the host.
A Docker-based cloud platform control device includes: a resource scheduling subunit and a platform branch unit;

The platform support unit is configured to receive a resource deployment request input by a user;

The resource scheduling unit is configured to: when the resource deployment request is a dynamic scheduling request, select a host with the lowest load information according to load information of each host on the cloud platform, and deploy a copy of the application information that needs to be scheduled on the host. The application information is application information on a host with the highest load information.
The Docker-based cloud platform control apparatus according to claim 6, wherein the load information includes load information of a container; the resource scheduling unit is configured to select load information from the host according to the preset scheduling policy. The lowest container and a copy of the application information is deployed on the container.
The Docker-based cloud platform control apparatus according to claim 6, further comprising a routing control unit and a general container setting unit;

The universal container setting unit is configured to set, on the host, a general container for running a copy of the application information, the universal container being a container including at least one programming language operating environment;

The routing control unit is configured to adjust the route of the application information to a container corresponding to the copy of the application information after the resource scheduling unit deploys the copy of the application information.
The Docker-based cloud platform control device according to claim 1, wherein the resource scheduling unit is further configured to determine whether the load status of the host on the cloud platform is in a high load state, and if yes, send an alarm. information;

If not, the application information on the host with lower load information is deployed to other hosts, and the application information on the host is deleted.
The Docker-based cloud platform control apparatus according to any one of claims 6 to 9, wherein the resource scheduling unit is further configured to acquire the according to the resource deployment request when the resource deployment request is an initial scheduling request The physical server of each host of the cloud platform performs weight setting on the physical server according to a preset resource deployment policy, and obtains priority ranking of each host, and the resource scheduling module applies the host according to the priority ranking. deploy.
A computer storage medium storing execution instructions for performing the method of any one of claims 1 to 4.