CN114968273A - Multi-cloud-scene-oriented application software continuous delivery method and system - Google Patents

Abstract

The invention discloses a method and a system for continuously delivering application software facing to a multi-cloud scene, which comprise a server side and a client side; the server is deployed at one end of an application delivery provider and used for realizing instantiation of application containing component classes based on the provided component classes, generating a component deployment list through configuration information and transmitting the component deployment list to the client after safety inspection; the method is also used for testing the application containing component class examples after the client application deployment is finished; the client is installed in a production environment designated by a user and used for carrying out environment configuration on the received component deployment list to form an execution list, and after resource initialization is carried out on the execution list, component execution is carried out on the execution list to carry out application deployment.

Description

Multi-cloud-scene-oriented application software continuous delivery method and system

Technical Field

The invention belongs to the technical field of application continuous delivery, and particularly relates to a method and a system for continuously delivering application software for a multi-cloud scene.

Background

With the development of information technology, the trend of enterprise informatization is increasingly obvious, and the development of a plurality of enterprises gradually tends to intelligent operation, namely the enterprises are in cloud. The enterprise cloud means that the enterprise connects social resources, a shared platform, working contents and the like through an internet technology and a cloud computing technology mode, so that the whole process of applying information management infrastructure construction, a management method, a business process and the like is developed. As more and more enterprises are increasingly cloud services, the difficulty of privatization delivery of enterprise applications increases exponentially. On one hand, the cloud environments used by each enterprise are different, including public cloud, private cloud, hybrid cloud and the like, so that each deployment environment needs to be configured during delivery; on the other hand, different enterprises use different applications, and there are also different versions using the same application. For example, user A needs to deploy version 1.0 of application 1 and version 1.0 of application 2, while user B needs to deploy version 2.0 of application 1. In order to ensure that the enterprise user can obtain good service experience, the enterprise application delivery platform needs to efficiently complete different environment configurations and multi-version management, otherwise, the user's requirements for multi-cloud deployment and version upgrading cannot be met.

In large-scale cloud computing and cloud storage scenarios, it is impractical to manually manage and deploy tens of millions of virtual machines. In order to implement efficient deployment of applications, many automated deployment tools for cloud resources have appeared in the last decade, and their main implementation manners are to use a Domain Specific Language (DSL) to describe cloud resources required for deployment of applications, parse the DSL into a scripting language for deployment of cloud resources by a parsing engine, and use the parsed language to call an API of a cloud service provider to implement allocation of cloud resources and deployment of applications. In the last decade, many organizations and companies have released delivery/deployment tools for cloud environments that abstract the application layer, reducing the differences in deployment among different cloud service environments by setting the environment variables to configurable inputs, thereby enabling cross-cloud deployment.

However, when an enterprise performs actual deployment, a link of configuring service middleware is often involved, for example, initiating deployment of an initial database script, releasing configuration center data, or registering open platform service data. If the enterprise chooses to deploy using existing delivery tools, then significant human intervention, such as database initialization, is required during the deployment process. In these manual operations, there is a high probability that command line operations are involved, increasing the probability of error and also raising the user threshold. Therefore, there is a need to develop a delivery tool that achieves zero manual intervention in the delivery process to reduce the user's threshold.

Currently, mainstream Cloud service providers all launch Deployment platforms for their Cloud services, such as Cloud format launched by AWS, Azure's Cloud Recourses and Google Cloud Deployment Manager launched by Azure. The deployment platforms provided by the cloud service providers have the problem of provider locking and cannot be applied to multi-cloud deployment. To address the vendor lock issue, several open organizations have proposed open standards for cloud application deployment, such as OVF, TOSCA, CAMP, and OCCI. Students also expand on the basis of the open standards to support flexible capacity expansion, fine-grained deployment and PaaS and SaaS services.

Disclosure of Invention

Aiming at the defects in the prior art, the invention aims to provide a method and a system for continuously delivering application software oriented to a cloudy scene, which can quickly and accurately realize the continuous delivery of application function software oriented to the cloudy scene.

In order to achieve the above purpose of the present invention, an embodiment provides a system for continuously delivering application software for a cloudy scene, including a server and a client.

The server is deployed at one end of an application delivery provider and used for realizing instantiation of application containing component classes based on the provided component classes, generating a component deployment list through configuration information and transmitting the component deployment list to the client after safety inspection; the method is also used for testing the application containing component class examples after the client application deployment is finished;

the client is installed in a production environment designated by a user and used for carrying out environment configuration on the received component deployment list to form an execution list, and carrying out component execution on the execution list to carry out application deployment after carrying out resource initialization on the execution list.

In one embodiment, the component class includes environment variables and operation templates of the components; the environment variables are used for improving the universality of the components in different deployment environments, including the environment configuration of a user;

the operation template comprises metadata, a deployment mode and capability of the component, wherein the metadata of the component is used for describing basic information of the component; the deployment mode is used for specifying a mode of selecting the deployment component by a user; the capability is used to support finer grained deployment operations for describing deployment rules for the component.

In one embodiment, the capability comprises metadata of the capability, a base template and a front-end template, wherein the metadata of the capability is used for describing the base information of the capability; the basic template and the front-end template are used for multiplexing codes with the same function in different components, the basic template is used for describing component execution details, and the front-end template is used for describing front-end components used in the processes of generating a component deployment list and an execution list.

In one embodiment, the server includes a component/application repository, a configuration rendering engine, a security gateway, and a testing engine;

the component/application repository is used for storing component classes and applications, wherein each application comprises a plurality of instances of the component classes;

the configuration rendering engine is used for realizing instantiation of the component class through configuration information of the component class;

the security gateway is used for establishing communication with the client and carrying out security detection;

the test engine is used for testing the instantiation of the application containing component class after the application client deployment is finished.

In one embodiment, the configuration rendering engine instantiates the component class and converts the component class into a manifest template, renders the manifest template into a dynamic form that can be displayed on a web page, and after receiving configuration information filled by a user through the dynamic form, replaces a placeholder in the manifest template with the configuration information filled by the user, and generates a component deployment manifest.

In one embodiment, after the deployment of the client is finished, the server receives an automatic test flow chart drawn by a user through a webpage to schedule an automatic test; the test engine sends the automatic test flow chart to an engine factory at the rear end in a JSON file format for processing and converting into executable engine examples, and the engine factory executes all the engine examples and records the operation results of the examples to finish the test.

In one embodiment, the client supports various deployment environments including a physical machine, a public cloud, a private cloud, a hybrid cloud, including an environment rendering engine, an execution engine;

the environment rendering engine is used for acquiring the component deployment list through the security gateway, rendering the component deployment list into a dynamic form on a web page, receiving the environment variables filled by the user through the dynamic form, and then modifying the component deployment list by using the filled environment variables to generate an execution list;

the execution engine is used for converting the execution list into an executable code and executing components through the connector after receiving the execution list and initializing the resources, so that the deployment of the application is realized.

In one embodiment, the component class further comprises some built-in components, the built-in components comprise an encapsulation for an existing deployment tool and a support for middleware, and the execution engine performs resource initialization by using the existing deployment tool provided by the component class, wherein the middleware refers to a class of software between application and system software.

In one embodiment, the process of component execution in the execution engine includes:

(1) generating a directed acyclic graph for a plurality of component classes contained in the application according to the component dependency relationship, and carrying out topological sorting on the component classes in the directed acyclic graph;

(2) executing a plurality of independent component classes which are not dependent and are found by topological sorting in parallel;

(3) executing the component capability aiming at each independent component class, searching components depending on the independent component class after the execution is successful, and if all the components are successfully executed, successfully deploying the application; when the execution of the component capability of the independent component class fails, the components depending on the independent component class are searched, the failure states of the components are set, and the failure state of submitting the independent components is displayed.

In order to achieve the above object, an embodiment of the present invention further provides a method for continuously delivering application software for a cloudy scene, where the method applies the above system, and includes the following steps:

step 1, deploying a server at an application delivery provider end, using the server to realize instantiation of an application containing component class based on the provided component class, generating a component deployment list through configuration information, and transmitting the component deployment list to a client after safety inspection;

step 2, installing the client in a production environment designated by a user, performing environment configuration on the received component deployment list by using the client to form an execution list, and performing component execution on the execution list to perform application deployment after performing resource initialization on the execution list;

and 3, after the client application deployment is finished, testing the application containing component class examples.

Compared with the prior art, the invention has the beneficial effects that at least:

the method is applicable to continuous interaction of various cloud environment applications and can also quickly realize multiplexing of deployment definitions (configuration information and environment configuration).

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.

Fig. 1 is a schematic structural diagram of a system for persistent delivery of application software oriented to a multi-cloud scenario, provided by an embodiment;

FIG. 2 is a flow diagram that illustrates the execution of components provided by the embodiments;

FIG. 3 illustrates an instantiation diagram of an application containing component classes;

fig. 4 is a schematic diagram of the capabilities provided by the embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the detailed description and specific examples, while indicating the scope of the invention, are intended for purposes of illustration only and are not intended to limit the scope of the invention.

Fig. 1 is a schematic structural diagram of a system for persistent delivery of application software oriented to a cloudy scene according to an embodiment. As shown in fig. 1, the system for continuously delivering application software provided by the embodiment includes a server and a client. The server is deployed at one end of an application delivery provider and used for realizing instantiation of application containing component classes based on the provided component classes, generating a component deployment list through configuration information and transmitting the component deployment list to the client after safety inspection; and the method is also used for testing the application containing component class instances after the client application deployment is finished. In an embodiment, an application refers to a service class application provided for a user, and an application contains multiple instances of component classes, each component class containing multiple capabilities.

The component class provided by the embodiment is defined as two classes contained in the DSL of the component, namely, the environment variable of the component and the operation template of the component. The system extracts environment variables in the component to promote the universality of the component class in different deployment environments. The environment variables include the environment configuration of the user, such as the type of the used resource (public cloud, private cloud, hybrid cloud, physical machine, etc.) and the configuration of the database (such as the address of the database). The operation template comprises metadata, deployment modes and capabilities of the components, namely the metadata, deployment modes and capabilities of the components are packaged into the operation template of the components by the system. The metadata of the component describes basic information of the component, and the deployment mode specifies the mode of the component selected by the user. Of course, the system also supports deployment in a variety of ways, including component deployment using existing deployment tools such as kubernets and Helm.

The capabilities in the operation template define the capabilities that the component has for supporting finer grained deployment operations. A set of capability DSLs is defined to describe the deployment rules of the component, which contains metadata of the capabilities, basic templates of the capabilities and front-end templates of the capabilities. The metadata of the capability is similar to the metadata of the component, and is used to describe the basic information of the capability. To maximize code reuse, a base template and a front-end template are used to reuse the same functional code in different components. The base template describes the execution details of the execution engine, while the front end template is used to describe the front end components in configuring the rendering engine and the environment rendering engine to generate dynamic forms.

Specifically, the server includes a component/application repository, a configuration rendering engine, a security gateway, and a testing engine. The component/application repository is used for storing component classes and applications, wherein each application comprises a plurality of instances of the component classes. Embodiments also provide some built-in components, including encapsulation for existing deployment tools (hellm, ansile, kubernets) and support for common middleware (databases, message queues, etc.). The common middleware refers to a type of software between application and system software, and includes databases (MySQL, Redis, MongoDB, PostgreSQL, policdb-X), message queues (rockmq, RabbitMQ), load balancing, distributed applications (schedule X, enterprise-level distributed application service EDAS), object storage, full-text search engines (Elasticsearch), and the like.

When an application is deployed, component classes corresponding to components included in the application need to be instantiated. The configuration rendering engine instantiates the component class and converts the component class into a manifest template, renders the manifest template into a dynamic form which can be displayed on a webpage, receives configuration information (such as a database address and an address of a cloud resource) filled by a user through the dynamic form, replaces a placeholder in the manifest template with the configuration information filled by the user, and generates a component deployment manifest. When a client side initiates a request to a server side, the request can carry out a series of security checks through a security gateway, and if the request is judged to be a security request, the server side can return a corresponding component deployment list. The security gateway includes a white list filter, a generic gateway, and a Web Application Firewall (WAF). After the application deployment operation of the client is completed, the server receives an automatic test flow chart drawn by a user through a webpage so as to schedule an automatic test; the test engine sends the automatic test flow chart to the engine factory at the back end in the JSON file format for processing and converting into an executable engine instance. The final engine factory will execute all the engine instances and record the results of the instances' runs to complete the test.

In the embodiment, the client is installed in a production environment specified by a user, and is used for performing environment configuration on the received component deployment list to form an execution list, and performing component execution on the execution list to perform application deployment after resource initialization is performed on the execution list. In particular, the system supports a variety of deployment environments including physical machines, public clouds, private clouds, hybrid clouds, and the like. The client includes an environment rendering engine and an execution engine. Firstly, the environment rendering engine acquires a component deployment list through a security gateway of a server, and renders the component deployment list into a dynamic form on a webpage, so that a user can conveniently fill in environment variable information. And after the input of the user is finished, receiving the environment variables filled by the user through the dynamic form, modifying the component deployment list by using the environment variables filled by the user, and generating an execution list. At this point, the execution engine receives the execution manifest and initializes the resources (middleware, etc.) using the initialization tools provided by the components. After the initialization is completed, the execution engine converts the execution manifest into executable code and performs component execution through a connector (operating on a production environment of a user, such as a cloud network or a physical machine), so as to realize final deployment operation of the application.

FIG. 2 is a flow diagram of component execution provided by an embodiment. As shown in fig. 2, the execution engine of the client, when executing component deployment, includes the following steps:

(1) generating a directed acyclic graph for a plurality of component classes contained in the application according to the component dependency relationship, and carrying out topological ordering on the component classes in the directed acyclic graph;

(2) executing a plurality of independent component classes which are not dependent and found by topological sorting in parallel;

(3) executing the component capability aiming at each independent component class, searching components depending on the independent component class after the execution is successful, and if all the components are successfully executed, successfully deploying the application; when the execution of the component capability of the independent component class fails, the components depending on the independent component class are searched, the failure states of the components are set, and the failure state of submitting the independent components is displayed.

Based on the system provided above, the embodiment further provides a method for continuously delivering application software oriented to a multi-cloud scenario, which includes the following steps:

step 1, deploying a server at an application delivery provider end, using the server to realize instantiation of an application containing component class based on the provided component class, generating a component deployment list through configuration information, and transmitting the component deployment list to a client after safety inspection;

step 2, installing the client in a production environment designated by a user, performing environment configuration on the received component deployment list by using the client to form an execution list, and performing component execution on the execution list to perform application deployment after performing resource initialization on the execution list;

and 3, after the client application deployment is finished, testing the application containing component class examples.

As shown in fig. 3, the embodiment further provides that an application includes two instances of component classes, App component class and MySQL component class. For an instance of the App component class, this component instance will be deployed using kubernets (an open source container orchestration engine), while it owns two dockers (an open source application container engine) mirroring the relevant environment variables imageHost (mirrored host address) and imageTag. The component has a DeployImage capability to deploy the Docker image into the user environment. For a MySQL (relational database management System) component class instance, this component instance will be deployed through the MySQL5.7 installation package. Three database-related environment variables, namely, host (host address), port (port) and dbname (database name), are extracted for database access. The component class has the capabilities of both CreateDB and InitSQL, and provides the capabilities of newly building a database and initializing the database SQL.

Shown in fig. 4 is a DSL describing the capabilities of creating table operations in a relational database, with the content in the content.script as a front-end template as input to configure the rendering engine. In this example, configuring the rendering engine renders a code box for entering code to create the form. After the user input is complete, the configuration rendering engine updates the particular region in the manifest with the user input. The basic template includes content corresponding to the execuutecode and other keys, which describes information related to execution.

The above-mentioned embodiments are intended to illustrate the technical solutions and advantages of the present invention, and it should be understood that the above-mentioned embodiments are only the most preferred embodiments of the present invention, and are not intended to limit the present invention, and any modifications, additions, equivalents, etc. made within the scope of the principles of the present invention should be included in the scope of the present invention.

Claims (10)

1. A system for continuously delivering application software facing to a multi-cloud scene is characterized by comprising a server side and a client side;

the server is deployed at one end of an application delivery provider and used for realizing instantiation of application containing component classes based on the provided component classes, generating a component deployment list through configuration information and transmitting the component deployment list to the client after safety inspection; the method is also used for testing the application containing component class examples after the client application deployment is finished;

the client is installed in a production environment designated by a user and used for carrying out environment configuration on the received component deployment list to form an execution list, and carrying out component execution on the execution list to carry out application deployment after resource initialization is carried out on the execution list.

2. The system for the continuous delivery of the multi-cloud-scenario-oriented application software according to claim 1, wherein the component class comprises environment variables and operation templates of components; the environment variables are used for improving the universality of the components in different deployment environments, including the environment configuration of a user;

the operation template comprises metadata, a deployment mode and capability of the component, wherein the metadata of the component is used for describing basic information of the component; the deployment mode is used for specifying a mode of selecting the deployment component by a user; the capability is used to support finer grained deployment operations for describing deployment rules for the component.

3. The system for the continuous delivery of the multi-cloud-scenario-oriented application software according to claim 2, wherein the capability comprises metadata of the capability, a basic template and a front-end template, wherein the metadata of the capability is used for describing basic information of the capability; the basic template and the front-end template are used for multiplexing codes with the same function in different components, the basic template is used for describing component execution details, and the front-end template is used for describing front-end components used in the processes of generating a component deployment list and an execution list.

4. The system for the continuous delivery of the multi-cloud-scenario-oriented application software according to claim 1 or 3, wherein the server comprises a component/application repository, a configuration rendering engine, a security gateway and a testing engine;

the component/application repository is used for storing component classes and applications, wherein each application comprises a plurality of instances of the component classes;

the configuration rendering engine is used for realizing instantiation of the component class by configuring information on the component class;

the security gateway is used for establishing communication with the client and carrying out security detection;

the test engine is used for testing the instantiation of the application containing component class after the application client deployment is finished.

5. The system for continuous delivery of application software for multi-cloud scenes as claimed in claim 4, wherein the configuration rendering engine instantiates and converts the component class into a manifest template, renders the manifest template into a dynamic form that can be displayed on a web page, and after receiving configuration information filled by a user through the dynamic form, replaces placeholders in the manifest template with the configuration information filled by the user to generate a component deployment manifest.

6. The system for continuously delivering application software to the cloudy scene according to claim 4, wherein after the deployment of the application software to the client is finished, the server receives an automatic test flow chart drawn by a user through a webpage so as to schedule an automatic test; the test engine sends the automatic test flow chart to an engine factory at the rear end in a JSON file format for processing and converting into executable engine examples, and the engine factory executes all the engine examples and records the operation results of the examples to finish the test.

7. The system for continuously delivering the application software facing the multi-cloud scene according to claim 1 or 3, wherein the client supports various deployment environments comprising a physical machine, a public cloud, a private cloud and a mixed cloud, and comprises an environment rendering engine and an execution engine;

the environment rendering engine is used for acquiring the component deployment list through the security gateway, rendering the component deployment list into a dynamic form on a web page, receiving the environment variables filled by the user through the dynamic form, and then modifying the component deployment list by using the filled environment variables to generate an execution list;

the execution engine is used for converting the execution list into executable codes and executing components through the connector after receiving the execution list and initializing resources, so that the deployment of the application is realized.

8. The system for multi-cloud scenario-oriented application software continuous delivery according to claim 7, wherein the component class further includes some built-in components, the built-in components include an encapsulation for existing deployment tools and support for middleware, and the execution engine performs resource initialization using the existing deployment tools provided by the component class, wherein the middleware refers to a class of software between application and system software.

9. The system for multi-cloud-scenario-oriented application software continuous delivery according to claim 7, wherein the process of component execution in the execution engine comprises:

(1) generating a directed acyclic graph for a plurality of component classes contained in the application according to the component dependency relationship, and carrying out topological ordering on the component classes in the directed acyclic graph;

(2) executing a plurality of independent component classes which are not dependent and are found by topological sorting in parallel;

(3) executing the component capability aiming at each independent component class, searching components depending on the independent component class after the execution is successful, and if all the components are successfully executed, successfully deploying the application; when the execution of the component capability of the independent component class fails, the components depending on the independent component class are searched, the failure states of the components are set, and the failure state of submitting the independent components is displayed.

10. A method for continuously delivering application software facing a cloudy scene, wherein the method applies the system of any one of claims 1 to 9, and comprises the following steps:

step 1, deploying a server at an application delivery provider end, using the server to realize instantiation of an application containing component class based on the provided component class, generating a component deployment list through configuration information, and transmitting the component deployment list to a client after safety inspection;

step 2, installing the client in a production environment designated by a user, performing environment configuration on the received component deployment list by using the client to form an execution list, and performing component execution on the execution list to perform application deployment after performing resource initialization on the execution list;

and 3, after the client application deployment is finished, testing the application containing component class examples.

Images