KR20220074367A - Serverless development support platform - Google Patents

Abstract

The serverless development support platform of the present invention includes: a user portal through which a developer can register a function for a function-based application; A development pipe having a source code repository for storing codes for functions registered by the developer, a build tool for building a code stored in the source code repository to generate a Docker image, and an image repository for storing the generated Docker image line; and a function execution unit that executes the Docker image in container (POD) units in the function execution environment.

Description

Serverless development support platform and method {SERVERLESS DEVELOPMENT SUPPORT PLATFORM}

The present invention relates to a serverless development support platform and method that can increase development convenience and related system operation efficiency of popular developers in the power field.

In the electric power field, the increase in renewable power generation and the increase in microgrids that can be individually managed have brought about a situation in which not only electric power companies but also various users collect/search/process power-related information for various purposes.

Power management issues due to the increase in electricity demand that the country is facing recently, expansion of construction of new and renewable energy power generation facilities following the full-scale implementation of the Renewable Portfolio Standard (RPS), and support and investment policies of the government for energy storage devices, such as active laws/systems This is leading to the development of various applications in the power/energy field.

On the other hand, in the power/energy field, relatively little information/data is used, but for the unified operation of the entire system, as a follow-up support for the information/data used, corrections and the like need to be performed collectively. In addition, even if the amount of information/data used is relatively small, if there is an application having a communication channel due to unnecessarily frequent data collection, etc., the operational efficiency of the power system management system and other applications is also reduced. In addition, in the power/energy field, most of the information/data used is derived from the power system, and almost nothing is generated by application developers or users.

However, in the power/energy field, there is no development platform that can support application development of third-party developers and support follow-up measures, so there is no choice but to use general application development tools, which reduces development efficiency and power system operation has become difficult due to the sprawl of applications.

Republic of Korea Publication No. 10-2017-0133055

An object of the present invention is to provide a serverless development support platform and method that can increase developer convenience in developing application programs in the power field and increase support efficiency for applications in a power management system.

Serverless development support platform according to an aspect of the present invention, a user portal through which a developer can register a function for a function-based application; A development pipe having a source code repository for storing codes for functions registered by the developer, a build tool for building a code stored in the source code repository to generate a Docker image, and an image repository for storing the generated Docker image line; and a function execution unit that executes the Docker image in container (POD) units in the function execution environment.

Here, the user portal, a code editor for inputting information about the function and the execution environment; and a function input/register for registering the input code as a function.

Here, it may further include an event processing module for processing function-related events that occur internally and externally.

Here, the event processing module, an event detection unit for detecting the occurrence of an event; a trigger that operates a trigger according to a message type for an event that has occurred; and an event processing unit for processing a corresponding event by executing a function according to the actuated trigger.

Here, it may further include an operator portal for monitoring the function usage status of the function stored in the image storage and the resource used by the function.

Here, the function execution unit may provide an auto-scaling function for automatically expanding resources such as a server and storage according to an increase in traffic.

Serverless development support method according to another aspect of the present invention, the step of receiving a function for a function-based application written by a developer as input; storing codes for functions registered by the developer; building the code for the stored function to create a Docker image; storing the generated Docker image in an image storage for execution; and executing the function in a container (POD) unit using a Docker image in the function execution environment of the application.

Here, if traffic increase by the function is expected in the function execution environment, the method may further include increasing the number of containers (PODs) of the function.

When the serverless development support platform and/or method according to the spirit of the present invention of the above configuration is implemented, the developer's development convenience in developing applications in the power field is improved, and the support efficiency for applications in the power management system is improved. There are advantages to be increased.

The serverless development support platform and/or method of the present invention has the advantage of reducing the time required for power application development and distribution through an automated development pipeline, and concentrating on providing services.

The serverless development support platform and/or method of the present invention can shorten the service improvement cycle by enabling rapid development and distribution due to distribution automation and function unit service distribution through a pipeline, and development/ It has the advantage of reducing costs by minimizing operating personnel.

The serverless development support platform and/or method of the present invention has the advantage of being able to provide services stably by preventing overall service failure due to functional unit development and distribution, thereby reducing the number of occurrences of system failures and risk factors.

The serverless development support platform and/or method of the present invention has the advantage of improving the flexibility of the power application service by utilizing the auto-scaling architecture and enabling efficient resource use.

1 is a conceptual diagram illustrating a change trend of a computing environment.
2 is a conceptual diagram illustrating a change trend of application logic.
3 is a conceptual diagram illustrating an overview of serverless computing.
Figure 4 is a block diagram showing the entire system of the serverless development support platform according to the spirit of the present invention.
Fig. 5 is a diagram illustrating the operational sequence of the platform for a function-as-a-service (FaaS) created according to the teachings of the present invention;
6 is a code editor example screen of the power serverless service platform.
7A to 7C are example interface screens for developers such as code templates.
8 is a diagram illustrating an event generation and trigger processing process.
9 is a diagram illustrating an auto-scaling process.
10 is a schematic diagram illustrating the concept of POD replication/removal during the auto-scaling process.
11 is a block diagram illustrating a monitoring service provided by an operator portal of a serverless development support platform according to the spirit of the present invention.
12 is a flowchart illustrating a serverless development support method according to the spirit of the present invention.

In describing the present invention, terms such as first, second, etc. may be used to describe various components, but the components may not be limited by the terms. The terms are only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, a first component may be referred to as a second component, and similarly, a second component may also be referred to as a first component.

When a component is referred to as being connected or connected to another component, it may be directly connected or connected to the other component, but it can be understood that other components may exist in between. .

The terminology used herein is used only to describe specific embodiments, and is not intended to limit the present invention. The singular expression may include the plural expression unless the context clearly dictates otherwise.

In this specification, the terms include or include are intended to designate that a feature, number, step, operation, component, part, or combination thereof described in the specification exists, and includes one or more other features or numbers, It may be understood that the existence or addition of steps, operations, components, parts, or combinations thereof is not precluded in advance.

In addition, shapes and sizes of elements in the drawings may be exaggerated for clearer description.

1 is a conceptual diagram illustrating a change trend of a computing environment.

As the corporate IT environment has changed from on-premises to cloud-based, scalability and efficiency have increased. As shown in Figure 1, the cloud service provides infrastructure as a service (IaaS) and a platform as a service (PaaS), which is a virtualized development/operation environment, and application to customers. It has evolved to Software as a Service (SaaS), which is a method of providing. Recently, as a serverless architecture that enables more efficient application construction and operation has appeared, many services are moving to a serverless architecture environment.

Serverless architecture does not generate idle resources because it is called and processed only when code is executed in an event-based, triggering manner, unlike the infrastructure-as-a-service and platform-as-a-service where the server is always operated. Serverless architecture is divided into Backend as a Service (BaaS) and Function as a Service (FaaS).

2 is a conceptual diagram illustrating a change trend of application logic.

As shown on the left of FIG. 2 , in the past, a monolithic architecture method in which the entire application is configured as one logic was used. As the system grew and the complexity increased, the microservice architecture method emerged, which is a method of breaking up an application into a combination of small applications. In addition, functions as a service, a method of breaking down into smaller units (functions), have emerged and are being utilized together with microservice architectures.

First, let's look at serverless computing introduced in the application development process in the power field in the present invention.

3 is a conceptual diagram illustrating an overview of serverless computing.

Serverless is a compound word of ‘Server + Less’, and as shown in FIG. 3, only when there is a request or a specific event, a server in the cloud is used or an application to be serviced is operated. The absence of a server is a symbolic expression, and there is a server that executes the work, but the management of the server is transferred to the cloud service provider. Therefore, developers can focus on actual implementation without worrying about server specifications, number, and network configuration.

Instead of setting up a server on a computer or virtual machine to perform a specific task, the task is processed by a backend as a service or a function as a service. Typical backends as services include Google Firebase, Azure Fabric, and Kinvey, and functions as services include AWS Lambda, Azure Functions, and Google Cloud Functions.

Function as a Service (FaaS) used in serverless computing is a form of dividing and implementing functions to be used by application development users into function units and providing services. A service that runs a function whenever a specific event (HTTP Request, API call, specific condition, etc.) occurs by eliminating a series of processes such as uploading a function to the server, configuring and distributing the code, and executing it, and registering the desired logic as a function. to be.

In the case of a platform as a service, since the entire application is distributed, the server runs 24 hours a day, but the function as a service distributes the application divided into function units and executes when a specific event occurs, and ends when the task is finished. Efficient operation is possible because the user pays the cloud provider for the time the function is executed and the number of times it is called, and the server automatically expands and shrinks according to the request amount.

The technology proposed in the present invention is a platform that provides a function-based serverless service as a service for power application development.

4 is a block diagram illustrating an entire system of a serverless development support platform according to the spirit of the present invention.

The illustrated serverless development support platform includes, as components for implementing the spirit of the present invention, a user portal 100 through which a developer can register a function for a function-based application; A source code repository 210 that stores codes for a function registered by a developer, build tools 220 and 230 that build a code (for a function) stored in the source code repository to generate a Docker image, and the generated Docker a development pipeline 200 having an image repository 240 for storing images; and a function execution unit 400 that executes the Docker image in container (POD) units in the function execution environment.

The entire system of the serverless development support platform is a user portal 100 for developers to register functions, a development pipeline module 200 for storing, managing, and building registered functions, and processes internal and external events. It consists of an event processing module 700, 800, 900 for executing a function, a function execution module 400 and 500 for executing a function, and an operator portal 600 for monitoring the function usage status and resource.

The illustrated event processing module 700 , 800 , 900 includes an event detection unit 700 for detecting the occurrence of an event; a trigger 800 for operating a trigger according to a message type for the generated event; and an event processing unit 900 that processes a corresponding event according to the operated trigger.

The function to be used for the power field application can be used by registering it in the user portal 100 . When information on a function and execution environment is input through the code editor provided by the user portal and the function is registered, it is transmitted to the development pipeline 200 . In the development pipeline 200 , the registered function is stored in the source code storage 210 , and an image is generated by building it and stored in the image storage 240 . When a function request occurs, the built image is executed in the container (POD#1 ~#3).

To this end, the user portal 100 includes a code editor 120 for inputting information on functions and execution environments; and function input/registries 140 and 160 for registering the input code as a function.

Next, the sequence of the development process using the serverless development support platform according to the spirit of the present invention will be described.

5 is a diagram illustrating an operation sequence of a platform for a function-as-a-service (FaaS) created according to the teachings of the present invention. That is, an operation sequence for developing and executing an application using the platform proposed by the present invention is shown.

In the diagram shown, the function is executed through the user portal 100 , the source code repository 210 , the CI Tool 230 , and the image repository 240 .

With respect to the user portal 100 , a user inputs and registers a function through a code editor provided by the user portal to implement a function-based application. When the user executes function registration, the user portal 100 stores the code information of the function in the source code storage 210 .

As a build tool, when a user registers a function, the CI Tool 230 receives an API request, executes a CI/CD command, and builds the code stored in the source code storage 210 to generate a Docker image.

The generated Docker image is stored in the Docker registry in the image storage 240 , and the function execution environment 400 requests image execution to execute the application.

In the function execution environment 400, the Docker image is executed in POD (container) units, and the service is controlled. When there is an execution request for the function from the inside or the outside, the function execution environment 400 executes the POD (container) unit.

6 is a code editor example screen of the power serverless service platform.

The illustrated code editor and code template can provide syntax highlighting and real-time syntax checking of a development language such as Python.

7A to 7C are example interface screens for developers such as code templates.

Figure 7a is a code template example screen of the power serverless service platform, 7b is a dashboard example screen of the user portal, 7c is a function management example screen of the user portal. In order to prevent human error, a code template that minimizes the user's code input can be provided, and event processing and code can be packaged so that non-developers can easily develop and apply the service.

8 is a diagram illustrating an event generation and trigger processing process.

When an event occurs, triggers (810 to 840) are activated according to the message type. Triggers filter events based on event attributes. For example, the types of triggers include an HTTP trigger 810 , a scheduler trigger 820 , a database trigger 830 , and a storage trigger 840 . When an event type is added, a trigger is also added to enable event message filtering.

The event data classified in this way is transmitted to the event processing module 900 for asynchronous message processing. The event processing module 900 registers with the event sources 910 to 940 according to the event attribute and stores it in the channel. The load balancer 990 checks the contents stored in the channel, and generates a service call that executes a function according to the traffic condition.

Next, auto-scaling to prevent traffic increase due to a function will be described.

Depending on implementation, the function execution environment 400 of the serverless development support platform of the present invention may provide an auto-scaling function for automatically expanding resources such as servers and storage according to an increase in traffic. As traffic increases, the number of PODs can be increased to provide services without delay and failure.

9 is a diagram illustrating an auto-scaling process.

10 is a schematic diagram illustrating the concept of POD duplication/removal during the auto-scaling process.

When the service is not activated, the route 1700 requests activation from the activator 1600 . When the service is activated, the function is executed in the PODs (1300, 1400). POD resource usage is monitored based on the metric, and the result is delivered to the auto scaler 1800 to resize (scale up, scale down) the POD and distribute it to the replica set 1200 .

At this time, the replica set 1200 auto-scaling by duplicating/removing the POD is shown in FIG. 10 . The replica set 1200 guarantees availability of the number of PODs to be executed and manages the number of PODs to be executed as many as the specified number of PODs. The replica set 1200 resizes the POD by duplicating or removing it based on the resource monitoring result metric.

The serverless development support platform according to the spirit of the present invention is used in the power field, as input data of a supported function, meter reading data, measurement data, distributed power generation data, and environmental data (weather, load/usage) in the power system. ) can be applied, and as events, simple application user's instruction/order/inquiry, sudden power demand, power sale/purchase order, sudden change in weather, distribution system accident, power generation facility accident, microgrid change, contract change etc. may exist. In addition, as specific parameters adjusted in the above-described scaling, for example, input data acquisition interval adjustment and communication resource allocation adjustment may be reflected.

Accordingly, when the transaction and load of the application in the power field increase, it is possible to automatically expand the service to provide an efficient service. In addition, since resources are automatically allocated when an event occurs, and resources are recovered when the service is terminated, efficient resource use is possible and resource shortages can be reduced.

Next, an event-driven service of an application developed according to the spirit of the present invention will be described.

The platform according to the present invention is an event-based service, and can allocate and use resources only when an event occurs. It provides various actions processed according to input events and can be used for various services including interactive services as shown in Table 2 below (representing event-based services).

11 illustrates a monitoring service provided by the operator portal 600 of the serverless development support platform according to the spirit of the present invention. The operator portal 600 may monitor the function usage status of the function stored in the image storage 240 and the resources used by the function.

When an event occurs and the corresponding function is executed, the service mesh platform collects traffic information for monitoring. In the dashboard for monitoring the power serverless platform, information is brought to the monitoring target using the Rest API. Function monitoring targets include log by function, number of function calls, and Web Request success rate as shown in Table 2 below (indicating monitoring functions and targets).

A dashboard for monitoring is provided in the operator portal 600 using the collected data. The dashboard provides function registration status, trigger registration status, function usage status, etc. using visualization tools.

12 is a flowchart illustrating a serverless development support method according to the spirit of the present invention.

The illustrated serverless development support method includes: receiving a function for a function-based application written by a developer (S120); Storing codes for functions registered by the developer (S140); Building a code for the stored function to generate a Docker image (S160); storing the generated Docker image in an image storage for execution (S180); and executing the function in a container (POD) unit of a Docker image in the function execution environment of the application ( S200 ).

Here, before and/or after executing the function, if traffic increase by the function in the function execution environment is expected, increasing the number of containers (PODs) of the function according to autoscaling may be further included. can

The flowchart shown in FIG. 12 shows the serverless development support method performed in the serverless development support platform according to the spirit of the present invention described above in a sequential flow, and overlapping detailed descriptions will be omitted.

The above description is merely illustrative of the technical spirit of the present invention, and various modifications and variations will be possible without departing from the essential characteristics of the present invention by those skilled in the art to which the present invention pertains. Accordingly, the embodiments disclosed in the present invention are not intended to limit the technical spirit of the present invention, but to explain, and the scope of the technical spirit of the present invention is not limited by these embodiments. The protection scope of the present invention should be construed by the following claims, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of the present invention.

100: user portal 200: development pipeline
210: source code repository 220, 230: build tool
240: image storage 400: function execution unit
600: operator portal 700: event detection unit
800: trigger 900: event processing unit

Claims (8)

a user portal where developers can register functions for function-based applications;
a source code storage for storing codes for functions registered by the developer;
A build tool that builds the code stored in the source code repository to create a Docker image;
Image storage to store the generated docker image
A development pipeline comprising; and
The function execution unit that executes the Docker image in container (POD) units in the function execution environment.
A serverless development support platform that includes
According to claim 1,
The user portal is
Code editor to input information about functions and execution environment; and
Function input/register to register the input code as a function
A serverless development support platform that includes
According to claim 1,
Event processing module to handle function-related events that occur internally and externally
Serverless development support platform that further includes.
4. The method of claim 3,
The event processing module,
an event detection unit for detecting the occurrence of an event;
a trigger that operates a trigger according to a message type for an event that has occurred; and
Event processing unit that processes the event by executing a function according to the triggered trigger
A serverless development support platform that includes
According to claim 1,
An operator portal that monitors the function usage status of the function stored in the image storage and the resources used by the function
Serverless development support platform that further includes.
According to claim 1,
The function execution unit,
A serverless development support platform that provides an auto-scaling function to automatically expand resources such as servers and storage according to traffic growth.
receiving a function input for a function-based application written by a developer;
storing codes for functions registered by the developer;
building the code for the stored function to create a Docker image;
storing the generated Docker image in an image storage for execution; and
executing the function in a container (POD) unit using a Docker image in the function execution environment of the application
How to support serverless development, including.
8. The method of claim 7,
Increasing the number of containers (PODs) of the function when traffic increase by the function is expected in the function execution environment
Serverless development support method further comprising.

Images