KR20150062198A - System and method for job execution in conjunction with cloud storage - Google Patents

Abstract

The present invention relates to a cloud storage based system and a method to perform a work, comprising: a plurality of cloud storages; at least one computing node which performs a work; a client which transmits a work file request signal including an authentication key and work file designation information to a mediation server, and transmits a work execution command to a mediation server when receiving a work execution command; and a mediation server which identifies a cloud storage of the client through the authentication key according to a reception of the work file request signal from the client, obtains a cloud storage directory and a work file URL corresponding to the work file designation information from the cloud storage, transmits a work file to a corresponding computing node by downloading the work file from the work file URL of the cloud storage directory, transmits the work execution command transmitted from the client to the computing node, and uploads a work execution result file of the computing node to a cloud storage of the client.

Description

[0001] SYSTEM AND METHOD FOR PERFORMING CLOUD STORAGE [0002]

The present invention relates to a cloud storage-based work execution system and method, and more particularly, to a cloud storage-based work execution system and method for identifying a cloud storage assigned to a client using an authentication key transmitted from a client, When the signal including at least one of the execution result upload command is received, the received signal is converted into a job specification according to the type of the job scheduler, and the corresponding computing node downloads the job file, And uploading a result of performing a work, according to a cloud-storage-based work execution system and method.

The existing computing environment, which relies on the independent hardware performance of each terminal according to the technological development of the computer network, is a form of cloud computing in which the service is provided according to the request of the terminal utilizing all the computing resources on the network It is evolving.

Cloud computing services can be defined as 'on-demand outsourcing services of computing resources' through information networks such as the Internet. In a cloud computing environment, service providers consolidate data centers distributed across multiple locations into virtualization technologies to provide services that users need. Service users are not required to install and use necessary computing resources such as application, storage, operating system, OS, security, etc. in each user's own terminal, The service in the virtual space can be selected and used as desired. The user does not pay for the purchase of each computing resource, but instead pays a fee based on usage.

According to the cloud computing service, users can access a cloud network through a terminal that performs network connection and basic computation functions at any place, perform tasks requiring a large-capacity storage device and a high-performance computing resource, and provide advanced services There is an advantage to receive.

However, task managers for existing high-performance computing clusters run tasks on DAS (Direct Attached Storage), so if you are working with data stored in rapidly growing cloud storage, You need to move the files in the storage to the computing resources and then do the work. In addition, there was an inconvenience that the user had to transfer files directly to the cloud storage every time in order to store the data generated as a result of the work in the user cloud storage.

Prior Art 1: Korean Patent No. 10-1081489 (published on May 10, 2010)

SUMMARY OF THE INVENTION The present invention provides a cloud storage-based work execution system and method capable of performing work using a high-performance computing cluster without directly uploading / downloading a file with only a cloud storage authentication key submission. There is a purpose.

Another object of the present invention is to provide a system and method for performing a cloud-based work that enables a user to access a cloud storage at any time by storing a work execution result file in a designated user cloud storage.

It is still another object of the present invention to provide a cloud storage-based work execution system and method capable of file backup and selective storage in a multi-cloud storage service.

It is still another object of the present invention to provide a cloud storage-based task execution system and method capable of adding a command to download a file stored in a cloud storage to a resource accessible by a task manager when a high performance cluster task execution command is generated.

It is still another object of the present invention to provide a cloud storage based work execution system and method capable of adding a command for uploading a result file generated as a result of an operation to a user cloud storage.

According to an aspect of the present invention, there is provided a method for transmitting a work file request signal including a plurality of cloud storages, at least one computing node performing an operation, an authentication key, and job file designation information to an intermediary server, A client for transmitting the job execution command to the mediating server when the job execution command is input; a server for checking the cloud storage of the client through the authentication key upon receipt of a job file request signal from the client, Acquiring a cloud storage directory and a work file URL corresponding to the work file designation information, downloading a work file from the work file URL of the cloud storage directory, transmitting the work file to the corresponding computing node, remind And a mediation server for transmitting a job execution result file at the computing node to the cloud storage of the client.

The mediating server can generate the job execution command transmitted from the client as a job specification according to the type of the job scheduler.

According to another aspect of the present invention, there is provided a communication system including a communication unit for communicating with a communication network, a storage unit for storing an application for performing a cloud storage-based operation, an input unit for inputting authentication keys and job file designation information, And a cloud service unit transmitting the job execution request command to the mediating server when the job execution command is input through the input unit and transmitting the job execution command to the mediating server through the communication unit.

According to another aspect of the present invention, there is provided a method for managing a cloud storage system, the method comprising: identifying a cloud storage allocated to the client using an authentication key transmitted from a client; receiving at least one of a work file download command, A job script generator for converting the received signal into a job specification according to the type of the job scheduler and transmitting the job specification to the head node when the signal including the job signal is received, A mediation server is provided that includes a head node that controls to perform at least one of downloading a file, performing a task, and uploading a task execution result.

Wherein the job script generator checks a cloud storage of the client through the authentication key when a job file request signal including an authentication key and job file designation information is received from the client, And an upload unit for receiving a signal including at least one of a work file download command, a work execution command, and a work execution result upload command from the client, the cloud storage directory and the work file URL corresponding to the cloud storage directory and the work file URL, An API call unit for calling an API necessary for executing the corresponding command in the task scheduler, and a task specification generator for converting the received signal into a task specification according to the type of the task scheduler.

The API calling unit may include at least one of an API call for downloading a work file required for performing an operation in the cloud storage of the client, and an API call for uploading operation result files to the client's cloud storage directory.

The head node may include at least one or more task schedulers.

According to another aspect of the present invention, there is provided a method for transmitting a work file to a mediation server, the method comprising the steps of: (a) transmitting a work file request signal including an authentication key and job file designation information to a mediation server; A cloud storage directory, and a work file URL corresponding to the work file designation information in the cloud storage, (c) the mediation server accesses a work file URL of the cloud storage directory, (D) transmitting the job execution command to the intermediary server when the job execution command is input, (e) transmitting the job execution command to the intermediary server, The server sends the job execution command to the computing node, and the computing node The cloud-based storage operations carried out how to perform the tasks resulting files comprise upload to cloud storage, the client is provided.

In the step (c), the mediating server converts the cloud storage directory and the work file URL into a work specification according to the kind of the job scheduler, and according to the converted work specification, the job file URL of the cloud storage directory And downloading the work file and transmitting the downloaded work file to the corresponding computing node.

Wherein the step (e) includes the steps of: converting the job execution command into a job specification according to a type of the job scheduler; transmitting the converted job specification to the computing node; To the cloud storage of the server.

The computing node may perform an operation corresponding to the job execution command.

According to another aspect of the present invention, there is provided a method for a mediation server to perform a cloud storage based operation, the method comprising the steps of: (a) receiving a work file request signal including an authentication key and work file designation information from a client, A cloud storage directory and a work file URL corresponding to the work file designation information in the cloud storage; (b) accessing a work file URL of the cloud storage directory (C) transmitting the job execution command to the computing node when the job execution command is received from the client, and transmitting the job execution command to the computing node, The work execution result file is stored in the client It is performed comprising uploaded to cloud storage Wood storage-based work methods is provided.

Wherein the step (c) includes the steps of: converting the job execution command into a job specification according to a type of the job scheduler when the job execution command is received from the client, transmitting the converted job specification to the computing node, And uploading an operation result file in the computing node to the client's cloud storage.

According to another aspect of the present invention, there is provided an information processing apparatus, comprising: (a) when a work file request signal including an authentication key and job file designation information is received from a client, Acquiring a cloud storage directory and a work file URL corresponding to the work file designation information in the cloud storage; (b) downloading a work file by accessing a work file URL of the cloud storage directory; (C) transmitting the job execution command to the computing node when a job execution command is received from the client, and transmitting a job execution result file to the computing node, To the cloud storage of A cloud storage-based job performing method is provided which is capable of being read by an electronic device recorded by a program.

According to the present invention, a user can perform work using a high performance computing cluster without directly uploading / downloading a file only by submitting a cloud storage authentication key.

In addition, the result file of the operation performed by the computing node can be stored in the designated user cloud storage, and the cloud storage can be accessed at any time / anywhere.

In addition, when a high performance cluster task execution command is generated, a command to download a file stored in the cloud storage to a resource accessible to the task manager can be added.

You can also add commands to upload the resulting files generated by your work to your cloud storage.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram illustrating a cloud storage based work execution system according to an embodiment of the present invention. FIG.
2 is a block diagram schematically showing the configuration of a client according to an embodiment of the present invention;
3 is a block diagram schematically showing a configuration of an intermediary server according to an embodiment of the present invention;
FIG. 4 is a block diagram specifically showing the configuration of the job script generator shown in FIG. 3; FIG.
5 is a diagram illustrating a cloud storage-based job performing method according to an embodiment of the present invention.

The foregoing and other objects, features, and advantages of the present invention will become more apparent from the following detailed description taken in conjunction with the accompanying drawings, in which: FIG.

1 is a diagram illustrating a cloud storage based work execution system according to an embodiment of the present invention.

Referring to FIG. 1, a cloud storage-based work execution system includes a client 100, a mediation server 200, a plurality of computing nodes 300, and a plurality of cloud storages 400.

The client 100 transmits a job file request signal including an authentication key and job file designation information to the mediation server 200. When the job execution command is input, the client 100 transmits a job execution command to the mediation server 200. [

The client 100 accesses the mediation server 200 through a communication network in order to receive the cloud computing service. The mediation server 200 manages various kinds of cloud computing services, and selects and connects an appropriate cloud computing service according to a request of the client 100. The mediation server 200 accesses the appropriate cloud storage 400 through the communication network in order to provide the storage requested by the client 100. [ The mediation server 200 is connected to the client 100 and the cloud storage 400 through a communication network and supports the client 100 to use the cloud storage 400.

The client 100 may be, for example, a navigation device, a notebook, a mobile communication terminal, a smart phone, a portable media player (PMP), a personal digital assistant (PDA), a tablet PC, Set-top box), smart TV, and the like.

A detailed description of the client 100 will be given with reference to FIG.

The mediation server 200 checks the cloud storage 400 of the client 100 through the authentication key when the work file request signal including the authentication key and the work file designation information is received from the client 100, The work file is downloaded from the work file URL of the cloud storage directory and transmitted to the corresponding computing node 300 after acquiring the cloud storage directory and the work file URL corresponding to the work file designation information in the cloud storage directory.

When the job execution command is received from the client 100, the mediation server 200 transmits a job execution command to the corresponding computing node 300, and transmits a job execution result file to the client 100 Upload to the assigned cloud storage. At this time, the mediation server 200 may generate the job execution command transmitted from the client 100 as a job specification according to the type of the job scheduler, and may operate according to the job specification. Here, the job scheduler is a task scheduler using a high performance computing cluster, and may include, for example, PBS, LL, SGE, and the like.

The mediation server 200 determines at least one cloud storage 400 suitable for a file access request of the client 100 and performs mediation between the client 100 and the cloud storage 400 suitable for the file access request . The mediation server 200 may exist on the Internet as a node independent of the client 100, the computing node 300 and the cloud storage 400, but some or all of the functions may be integrated as a part of the client 100 .

The mediation server 200 will be described in detail with reference to FIG.

The computing node 300 performs an operation according to a job execution command transmitted from the mediation server 200. [

The cloud storage 400 may include a web service network interface and a data storage medium for transmitting / receiving data to / from the Internet. The cloud storage 400 may be a heterogeneous storage. For example, the cloud storage 400 may be provided by different cloud storage providers, and may be configured to store data so that they are accessible according to different file systems, respectively.

Although the cloud storage 400 and the computing node 300 are described separately in FIG. 1, the cloud storage 400 and the computing node 300 may be included in the cloud infrastructure.

2 is a block diagram schematically showing a configuration of a client according to an embodiment of the present invention.

Referring to FIG. 2, the client 100 includes a communication unit 110, an input unit 120, an output unit 130, a storage unit 140, a cloud service unit 150, and a control unit 160.

The communication unit 110 is a communication unit that interconnects the client 100 and the intermediary server through a communication network. The communication unit 110 may be a wireless communication module such as mobile communication or satellite communication, a wired communication module such as the Internet, Modules, and the like.

The input unit 120 is a means for receiving a user request for controlling the operation of the client 100, and converts the user's request into an electric signal in response to a user's operation. The input unit 120 may be a keyboard, a keypad, a touch screen, a visual sensing unit, a tactile sensing unit, a motion sensing unit, , Voice input means, and the like.

The user can input the authentication key and work file designation information through the input unit 120. [ Here, the authentication key may include an ID / password, a private key, and the like.

The output unit 130 may be implemented as a display device for displaying screen information according to application driving, for example, a small flat display device such as an LCD (Liquid Crystal Display) or OLED (Organic Light Emitting Diodes).

The storage unit 140 stores a program necessary for controlling the operation of the client 100 and data generated during the execution of the program. The storage unit 140 stores an application for performing a cloud storage-based work.

The cloud service unit 150 transmits a job file request signal including the authentication key and job file designation information to the mediation server through the communication unit 110. When the job execution command is inputted through the input unit 120, Send the command to the mediation server.

The control unit 170 controls the operation of each unit such as the communication unit 110, the input unit 120, the output unit 130, the storage unit 140, and the cloud service unit 150.

FIG. 3 is a block diagram schematically showing the configuration of an intermediary server according to an embodiment of the present invention, and FIG. 4 is a block diagram specifically showing the configuration of the job script generator shown in FIG.

Referring to FIG. 3, the mediation server 200 includes a job script generator 210, a head node 220, and the like.

The job script generator 210 checks the cloud storage allocated to the client using the authentication key transmitted from the client, and receives a signal including at least one of a work file download command, a work execution command, and a work execution result upload command from the client If it is received, the received signal is converted into a job specification according to the type of the job scheduler and transmitted to the head node 220.

The job script generator 210 may be a command script generator performed by a Job Scheduler of the head node 220.

Referring to FIG. 4, the job script generator 210 includes an authentication unit 212, an API call unit 214, and a work specification generator 216. The job script generator 210 includes an authentication unit 212, an API call unit 214,

When the work file request signal including the authentication key and job file designation information is received from the client, the authentication unit 212 checks the client's cloud storage through the authentication key, Obtain the cloud storage directory and work file URL.

When a signal including at least one of a work file download command, a work execution command, and a work execution result upload command is received from the client, the API calling unit 214 calls an API necessary for executing the corresponding command from the cloud storage allocated to the client do. That is, the API calling unit 214 may call an API for downloading a work file necessary for performing an operation in the client's cloud storage, or may call an API for uploading operation result files to the client's cloud storage directory .

The job statement generating unit 216 converts a signal including at least one of a job file download command, a job execution command, and a job execution result upload command into a job specification according to the type of the job scheduler.

The head node 220 controls the corresponding computing node to perform at least one of downloading a job file, performing a job, and uploading a result of performing a job according to the job specification transmitted from the job script generator 210. The head node 220 may include at least one job scheduler 225 and the job scheduler 225 may include, for example, PBS, LL, SGE, and the like.

5 is a diagram illustrating a cloud storage-based work execution method according to an embodiment of the present invention.

Referring to FIG. 5, when an authentication key and job file designation information are input from a user (S502), the client transmits a work file request signal including the input authentication key and job file designation information to the mediation server (S504).

The mediation server checks the cloud storage of the client through the authentication key (S506), and acquires the cloud storage directory and the work file URL corresponding to the work file specification information in the cloud storage (S508).

Then, the mediating server converts the obtained cloud storage directory and the work file URL into a job specification according to the type of the job scheduler (S510), operates the job scheduler, and creates a job file of the cloud storage directory And downloads the corresponding work file from the URL to the corresponding computing node (S512).

The intermediary server transmits a work file download completion signal to the client (S514).

When a job execution command is inputted from the user (S516), the client transmits the inputted job execution command to the mediation server (S518).

The mediation server generates a job execution command as a job specification according to the job scheduler type and transmits the job execution command to the computing node in operation S520 and uploads the operation result file in the computing node to the client's cloud storage in operation S522.

According to another aspect of the present invention, there is provided an information processing method for a mediation server, the method comprising: (a) when a work file request signal including an authentication key and work file designation information is received from a client, (B) acquiring a cloud storage directory and a work file URL corresponding to the work file designation information from the cloud storage, (b) accessing a work file URL of the cloud storage directory to download a work file, (C) transmitting a job execution command to the computing node when a job execution command is received from the client, and transmitting a job execution result file from the computing node to the computing node, Uploading to the client's cloud storage; A cloud storage-based job execution method including the recording medium readable by the electronic apparatus recorded by the program is provided.

Thus, those skilled in the art will appreciate that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. It is therefore to be understood that the embodiments described above are to be considered in all respects only as illustrative and not restrictive. The scope of the present invention is defined by the appended claims rather than the detailed description and all changes or modifications derived from the meaning and scope of the claims and their equivalents are to be construed as being included within the scope of the present invention do.

100: client 110:
120: input unit 130: output unit
140: Storage unit 150: Cloud service unit
160: control unit 200:
210: job script generator 220: Head node
225: job scheduler 212: authentication unit
214: API calling unit 216: work specification generating unit
300: computing node 400: cloud storage

Claims (14)

Multiple cloud storage;
At least one computing node performing an operation;
A client for transmitting a job file request signal including an authentication key and job file designation information to an intermediary server and transmitting the job execution command to the intermediary server when a job execution command is input; And
A cloud storage directory and a work file URL corresponding to the work file designation information are obtained from the cloud storage, And transmitting a job execution command transmitted from the client to the computing node, and transmitting a work execution result file at the computing node to the cloud of the client, A mediation server that uploads to storage;
A cloud storage-based task execution system.
The method according to claim 1,
Wherein the mediating server generates a job execution command transmitted from the client as a work specification according to a type of the job scheduler.
A communication unit for communicating with a communication network;
A storage unit for storing an application for performing a cloud storage-based operation;
An input unit for receiving an authentication key and job file designation information;
A job file request signal including the authentication key and job file designation information is transmitted to the mediation server through the communication unit, and when a job execution command is input through the input unit, the job execution command is transmitted to the mediation server Service department;
Lt; / RTI >
When a signal including at least one of a work file download command, a work execution command, and a work execution result upload command is received from the client by using the authentication key transmitted from the client to identify the cloud storage allocated to the client, A job script generator for converting the received signal into a job specification according to a type of the job scheduler and transmitting the job specification to the head node; And
A head node for controlling the corresponding computing node to perform at least one of downloading a job file, performing a job, and uploading a result of performing a job according to a job specification transmitted from the job script generator;
≪ / RTI >
5. The method of claim 4,
Wherein the job script generator checks a cloud storage of the client through the authentication key when a job file request signal including an authentication key and job file designation information is received from the client, An authentication unit for obtaining a cloud storage directory and a work file URL corresponding to the URL;
An API call unit for calling an API necessary for performing a corresponding command in the cloud storage of the client when a signal including at least one of a work file download command, a work execution command, and a work execution result upload command is received from the client; And
And a job specification generator for converting the received signal into a job specification according to the type of the job scheduler.
5. The method of claim 4,
Wherein the API call unit includes at least one of an API call for downloading a work file necessary for performing an operation in the cloud storage of the client and an API call for uploading operation result files to the client's cloud storage directory Mediation Server.
5. The method of claim 4,
Wherein the head node comprises at least one or more task schedulers.
(a) transmitting, by the client, a work file request signal including an authentication key and job file designation information to an intermediary server;
(b) the intermediary server confirms the cloud storage of the client through the authentication key, and obtains the cloud storage directory and the work file URL corresponding to the work file specification information from the cloud storage;
(c) the intermediary server accesses a work file URL of the cloud storage directory, downloads a work file, and transmits the downloaded work file to the corresponding computing node;
(d) when the job execution command is input, the client transmits the job execution command to the mediation server; And
(e) the intermediary server transmits the job execution command to the computing node, and uploading a work execution result file at the computing node to the cloud storage of the client;
Based cloud storage.
9. The method of claim 8,
In the step (c)
Converting the cloud storage directory and the work file URL into a job specification according to a type of the job scheduler; And
And accessing the work file URL of the cloud storage directory according to the converted work specification to download the work file and transmitting the downloaded work file to the corresponding computing node. .
9. The method of claim 8,
The step (e)
Converting the job execution command into a job specification according to a job scheduler type; And
And transmitting the converted work specification to the computing node, and uploading a work execution result file at the computing node to the cloud storage of the client.
9. The method of claim 8,
Wherein the computing node performs a task corresponding to the job execution command
A method for a mediation server to perform a cloud storage based operation,
(a) when a work file request signal including an authentication key and job file designation information is received from a client, the client device checks the cloud storage of the client through the authentication key, Obtaining a cloud storage directory and a work file URL;
(b) accessing a work file URL of the cloud storage directory to download a work file, and transmitting the downloaded work file to a corresponding computing node; And
(c) when the job execution command is received from the client, transmitting the job execution command to the computing node and uploading a job execution result file in the computing node to the client's cloud storage;
Based cloud storage.
13. The method of claim 12,
The step (c)
Converting the job execution command into a job specification according to a type of the job scheduler when the job execution command is received from the client;
Transmitting the converted job specification to the computing node; And
And uploading a resultant file of the task execution result at the computing node to the cloud storage of the client.
When executed by the mediation server,
(a) when a work file request signal including an authentication key and job file designation information is received from a client, the client device checks the cloud storage of the client through the authentication key, Obtaining a cloud storage directory and a work file URL;
(b) accessing a work file URL of the cloud storage directory to download a work file, and transmitting the downloaded work file to a corresponding computing node; And
(c) when the job execution command is received from the client, transmitting the job execution command to the computing node and uploading a job execution result file in the computing node to the client's cloud storage; and A method of performing a storage-based job, the method comprising the steps of:

Images