KR20190069655A - Method and apparatus of providing deep learning development platform based on cloud computing - Google Patents

Abstract

Disclosed are a method and an apparatus of providing deep learning development platform based on cloud computing. According to one embodiment of the present invention, the method comprises the following steps: receiving a service request related to deep learning in a frontend module; and performing a requested service related to deep learning using a backend module. The frontend module and backend module are implemented in a cloud computing server.

Description

[0001] METHOD AND APPARATUS FOR PROVIDING DEEP LEARNING DEVELOPMENT [0002] BASED ON CLOUD COMPUTING [

The following embodiments relate to a method and apparatus for providing a deep learning development platform based on cloud computing.

Platform as a Service (PaaS) is a model for constructing and providing a usable environment without having to build a separate platform for development. With PaaS, users can develop and manage applications without the complexities of managing related infrastructures (middleware, software, etc.).

Cloud Foundry is an open PaaS platform led by VMWare. Using cloud foundries reduces development time and simplifies operations and management.

Open PaaS is a service that provides an environment for testing and running software in a cloud-based virtual environment. Open PaaS provides several programming languages and frameworks for development based on the cloud.

Users can leverage these cloud foundries to create their own PaaS environments

Embodiments can provide a technology that utilizes a cloud computing server to provide a deep learning development platform.

A method for providing a deep learning development platform according to an embodiment includes receiving a deep learning related service request in a front end module and performing a requested deep learning related service using a backend module Wherein the front end module and the back end module are implemented in a cloud computing server.

The receiving may include receiving a code from an object requesting the deep learning related service.

The code may include python code and HyperText Markup Language (HTML) code.

The performing may include communicating with a deep learning server, and executing a command through a cloud foundry command line interface.

The step of performing the communication may include communicating with the deep learning server using a REST (REpresentational State Transfer) Ful API (Application Programming Interface).

The front-end module may be implemented based on at least one of HTML and Javascript.

The back-end module may be implemented as a Python Flask-based web interface.

An apparatus for providing a deep learning development platform according to an embodiment includes a front end module for receiving a deep learning related service request and a backend module for performing a requested deep learning related service, The front-end module and the back-end module are implemented in a cloud computing server.

The front end module may receive a code from an object requesting the deep learning related service.

The code may include python code and HyperText Markup Language (HTML) code.

The back end module may communicate with a deep running server and may execute commands through a cloud foundry command line interface.

The back end module can perform communication with the deep learning server using a REST (REpresentational State Transfer) Ful API (Application Programming Interface).

The front-end module may be implemented based on at least one of HTML and Javascript.

The back-end module may be implemented as a Python Flask-based web interface.

1 shows a schematic block diagram of a system for providing a deep learning development platform in accordance with one embodiment.
2 shows a schematic block diagram of the cloud computing server shown in FIG.
FIG. 3 shows an example of the operation of the deep learning development platform providing system shown in FIG.
Figure 4 shows an example of a web interface using a Python flask.
5 shows an example of a deep learning application using the deep learning platform development apparatus shown in FIG.

It is to be understood that the specific structural or functional descriptions of embodiments of the present invention disclosed herein are presented for the purpose of describing embodiments only in accordance with the concepts of the present invention, May be embodied in various forms and are not limited to the embodiments described herein.

Embodiments in accordance with the concepts of the present invention are capable of various modifications and may take various forms, so that the embodiments are illustrated in the drawings and described in detail herein. However, it is not intended to limit the embodiments according to the concepts of the present invention to the specific disclosure forms, but includes changes, equivalents, or alternatives falling within the spirit and scope of the present invention.

The terms first, second, or the like may be used to describe various elements, but the elements should not be limited by the terms. The terms may be named for the purpose of distinguishing one element from another, for example without departing from the scope of the right according to the concept of the present invention, the first element being referred to as the second element, Similarly, the second component may also be referred to as the first component.

It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, . On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between. Expressions that describe the relationship between components, for example, "between" and "immediately" or "directly adjacent to" should be interpreted as well.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, the terms " comprises ", or " having ", and the like, are used to specify one or more of the features, numbers, steps, operations, elements, But do not preclude the presence or addition of steps, operations, elements, parts, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in commonly used dictionaries are to be interpreted as having a meaning consistent with the meaning of the context in the relevant art and, unless explicitly defined herein, are to be interpreted as ideal or overly formal Do not.

A module in this specification may mean hardware capable of performing the functions and operations according to the respective names described in this specification and may mean computer program codes capable of performing specific functions and operations , Or an electronic recording medium, e.g., a processor or a microprocessor, equipped with computer program code capable of performing certain functions and operations.

In other words, a module may mean a functional and / or structural combination of hardware for carrying out the technical idea of the present invention and / or software for driving the hardware.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, the scope of the patent application is not limited or limited by these embodiments. Like reference symbols in the drawings denote like elements.

1 shows a schematic block diagram of a system for providing a deep learning development platform in accordance with one embodiment.

Referring to FIG. 1, a deep learning development platform providing system 10 includes an object 100 and a deep learning development platform providing apparatus 200. The deep learning development platform providing system 10 can provide a deep learning development platform using the cloud computing to the object 100. [ The object 100 may include developers and applications that wish to utilize a deep learning platform using cloud computing.

The object 100 can develop an application using deep learning by simply connecting to the cloud computing server 210 without setting a separate development environment in order to create an application utilizing deep learning.

The cloud computing server 210 can provide a Python flask based web interface. For example, the object 100 may generate an application using a web interface based on a python flask, without inputting a command through a command window according to a separate programming language or program, And the like.

The object 100 may output a code to the cloud computing server 210. The code may include code related to deep learning development. For example, the code may include python code and HyperText Markup Language (HTML) code.

The deep learning development platform providing apparatus 200 may include a cloud computing server 210 and a deep learning server 230. The object 100 may access the deep learning server 230 through the cloud computing server 210.

The cloud computing server 210 may mediate the deep running server 230 suitable for the object 100 to the object 100 in response to a service request of the object 100. [ For example, the cloud computing server 210 may be implemented as a Platform as a Service (PaaS) server.

The cloud computing server 210 can bind the object 100 and the deep learning server 230 by searching the deep learning server 230 corresponding to the service requested by the object. The deep running server 230 may include a plurality of servers.

The deep learning development platform providing apparatus 200 can provide a deep learning development platform using a platform as a service (PaaS). The deep learning development platform providing apparatus 200 can provide a deep learning development environment using a Python Flask-based web interface.

The object 100 can create its own application in a PaaS environment by writing code using a desired programming language. Programming languages can include BASIC, C, C #, C ++, D, F #, Python, Ruby, Java, Pascal, Prolog, Fortran, COBOL, Lisp, Perl, R, Groovy, Scala, occam, Swift and HTML .

2 shows a schematic block diagram of the cloud computing server shown in FIG.

Referring to FIG. 2, the cloud computing server 210 may include a front-end module 211 and a back-end module 213.

The front end module 211 can receive a deep running related service request. The front end module 211 may receive a code from an object 100 requesting a deep running related service. The front end module 211 may be implemented based on at least one of HTML and Javascript.

The front-end module 211 may control the back-end module 213 in response to a request of the object 100. For example, the front-end module 211 may output a control signal to the back-end module 213.

The back-end module 213 may perform the requested deep-running related service. The back-end module 213 may manage the environment of the cloud computing server 210 in response to the request of the object 100. [

The back end module 213 communicates with the deep learning server 230 and can execute commands through a cloud foundry command line interface (CF CLI). For example, the back-end module 213 may perform a row cloud establishment (CF) command through the CF CLI.

The back-end module 213 loads the deep learning application created by the object 100 into a plurality of containers and communicates with the deep learning server 230 to process the requested service in the application created by the object 100 have.

Container may refer to a virtual operating system where an application can be executed.

The back-end module 213 can communicate with the deep learning server 230 using REST (REpresentational State Transfer) Ful API (Application Programming Interface). The back-end module 213 may be implemented as a Python Flask-based web interface.

A specific code is called through the web interface of the cloud computing server 210 to communicate with the deep learning server 230 in the form of a RESTful API in a service application previously implemented in the back end module 213, Platform can be provided.

Communication between the cloud computing server 210 and the deep learning server 230 may be performed remotely.

The front end module 211 and the back end module 213 may be implemented inside or outside the cloud computing server 210.

FIG. 3 shows an example of the operation of the deep learning development platform providing system shown in FIG.

Referring to FIG. 3, the object 100 may be a user. The deep learning development platform providing apparatus 200 may include a cloud computing server 210 and a deep learning server 230. For example, the cloud computing server 210 may be implemented as a PaaS server.

The cloud computing server 210 may include a front end module 211 and a back end module 213. The front end module 211 may include a web interface, and the back end module 213 may include a back end service.

The user can communicate with the cloud computing server via the web interface. The web interface can control the back-end service in response to a service request received from the user. The back-end service can process the requested service from the user.

FIG. 4 shows an example of a web interface using a Python flask, and FIG. 5 shows an example of a deep running application using the apparatus for developing a deep running platform shown in FIG.

4 and 5, the cloud computing server 210 may distribute and operate an application. The cloud computing server 210 can provide a wide range of runtime and freedom to select frameworks and services.

The cloud computing server 210 is not subject to the characteristic cloud environment and can guarantee freedom of choice and diversity. In addition, the cloud computing server 210 may be extended to newly added technologies and services. The cloud computing server 210 may be used to build a public cloud and a private cloud. The cloud computing server 210 can easily transfer the code created by the object 100 to various environments.

The cloud computing server 210 may bind the deep learning server 230 suitable for the object 100 according to the service request of the object 100. [ The cloud computing server 210 may create and delete containers. In addition, the cloud computing server 210 can manage the container and communicate with the container.

The cloud computing server 210 can establish a network of the created container and communicate with the container.

The object 100 may request a service from the cloud computing server 210. For example, a service requested by an object may include a Graphics Processing Unit (GPU) service and a storage service.

The cloud computing server 210 may search the deep learning server 230 that can provide the service requested by the object 100 to create a container (or instance). The cloud computing server 210 can provide a service requested by the object 100 by binding the object 100 with the container.

The cloud computing server 210 may include a cloud controller. The cloud controller can access the service broker through a unique URL (Uniform Resource Locator) of bosh included in the deep running server 230 providing the service.

The cloud controller complies with the Service Broker API conventions and can request service instances (or containers). The Bosch of the cloud computing server 210 may use the generated service instance to perform a request received by the service broker.

The cloud controller can bind to the development environment of the object 100 included in the cloud computing server 210. When the binding is completed, the instance of the object 100 can directly communicate with the development environment of the object 100 without going through the service broker and the cloud controller.

The cloud computing server 210 may upload the code for creating a service broker to Bosch. The cloud computing server 210 can define and catalog a catalog of the service broker and open the service broker to access the URL.

The cloud computing server 210 can perform routing to an appropriate component (authentication or service layer) when a request for service of the object 100 or a request for authentication of the object 100 occurs. The cloud computing server 210 may perform authentication of the object 100 connected to the cloud computing server 210. The cloud computing server 210 can make a request for service or authentication to a plurality of Bosch through the cloud controller and receive a result of the request.

The cloud computing server 210 can create, store, and manage an application development environment. A Diego cell included in the cloud computing server 210 may execute an application of Bosch that provides a service.

The cloud computing server 210 can connect and manage service instances using a service broker. The cloud computing server 210 can control the internal communication of the cloud computing server 210 using a NATs message bus. The cloud computing server 210 may log operations within the cloud computing server 210.

A service broker can manage service instances. The service broker may create or delete instances at the request of the cloud computing server 210. The service broker can bind the instance with the object 100 development environment of the cloud computing server 210 or can release the existing binding. The service broker can communicate in the form of a HyperText Transfer Protocol (HTTP) request in accordance with the protocols defined in the Service Broker API.

The service broker API can obtain the URL of the service broker from the cloud computing server 210 and write the code in the HTTP request format. The service broker API may request to create a service instance in the deep learning server 230 in response to a service request received via the cloud controller. The service broker API can bind the service instance with the object 100 development environment.

The apparatus described above may be implemented as a hardware component, a software component, and / or a combination of hardware components and software components. For example, the apparatus and components described in the embodiments may be implemented within a computer system, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA) , A programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For ease of understanding, the processing apparatus may be described as being used singly, but those skilled in the art will recognize that the processing apparatus may have a plurality of processing elements and / As shown in FIG. For example, the processing unit may comprise a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as a parallel processor.

The software may include a computer program, code, instructions, or a combination of one or more of the foregoing, and may be configured to configure the processing device to operate as desired or to process it collectively or collectively Device can be commanded. The software and / or data may be in the form of any type of machine, component, physical device, virtual equipment, computer storage media, or device , Or may be permanently or temporarily embodied in a transmitted signal wave. The software may be distributed over a networked computer system and stored or executed in a distributed manner. The software and data may be stored on one or more computer readable recording media.

The method according to an embodiment may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

Claims (14)

Receiving a deep running related service request from a front end module; And
Performing a requested deep learning related service using a backend module
Lt; / RTI >
The front-end module and the back-end module are implemented in a cloud computing server
How to provide a deep learning development platform.
The method according to claim 1,
Wherein the receiving comprises:
Receiving a code from an object requesting the deep learning related service
The method comprising the steps of:
3. The method of claim 2,
The code includes:
Python code and HyperText Markup Language (HTML) code
The method comprising the steps of:
The method according to claim 1,
Wherein the performing comprises:
Performing communication with a deep learning server; And
Steps to perform an instruction through the Cloud Foundry Command Line Interface
The method comprising the steps of:
5. The method of claim 4,
The step of performing the communication includes:
Performing communication with the deep learning server using REST (REpresentational State Transfer) Ful API (Application Programming Interface)
The method comprising the steps of:
The method according to claim 1,
The front end module includes:
It is implemented based on at least one of HTML and JavaScript (Javascript).
How to provide a deep learning development platform.
The method according to claim 1,
The back-
It is implemented as a web interface based on the Python Flask.
How to provide a deep learning development platform.
A front end module for receiving a deep running related service request; And
A backend module that performs the requested deep-running related services
/ RTI >
The front-end module and the back-end module are implemented in a cloud computing server
Deep learning development platform providing device.
9. The method of claim 8,
The front end module includes:
Receiving a code from an object requesting the deep learning related service;
Deep learning development platform providing device.
10. The method of claim 9,
The code includes:
Python code and HyperText Markup Language (HTML) code
And a device for providing a deep learning development platform.
9. The method of claim 8,
The back-
To communicate with the Deep Learning server, and to execute commands through the Cloud Foundry Command Line Interface
Deep learning development platform providing device.
12. The method of claim 11,
The back-
And performs communication with the deep learning server using REST (REpresentational State Transfer) Ful API (Application Programming Interface)
Deep learning development platform providing device.
9. The method of claim 8,
The front end module includes:
It is implemented based on at least one of HTML and JavaScript (Javascript).
Deep learning development platform providing device.
9. The method of claim 8,
The back-
It is implemented as a web interface based on the Python Flask.
Deep learning development platform providing device.

Images