KR102361217B1 - The method and system for providing cloud service including machine learning service based blockchain - Google Patents

Abstract

Embodiments relate to a cloud service providing method including blockchain-based machine learning service and a system thereof and, more specifically, to a blockchain-based cloud service providing method and a system thereof which provide a cloud service by using computing resources of nodes configuring a blockchain and provide rewards to the nodes, which provide the computing resources.

Description

A method and system for providing a cloud service including a blockchain-based machine learning service

The following embodiments relate to a block chain-based cloud service providing method and system, and specifically, a block providing a cloud service using the computing resources of the nodes constituting the block chain, and providing a reward to the nodes that provided the computing resources It relates to a method and system for providing a cloud service including a chain-based machine learning service.

Blockchain technology, also referred to as distributed ledger technology, is a new technology in which one or more computing devices jointly participate in “accounting” and jointly maintain a fully distributed database. Blockchain technology has the characteristics of decentralization, openness, and transparency, and each computing device can participate in database recording, and the computing devices can quickly perform data synchronization. Methods of establishing a decentralized system using blockchain technology in the distributed database of blockchain and recording various executable programs for automatic execution in the distributed database of blockchain are widely applied in many fields.

The characteristics of big data technology, which are the core of the 4th industrial revolution, can be defined as 3V, which are volume, velocity, and variety. Also, recently, not only 3V but also validity and reliability are mentioned as big data characteristics. Validity is a factor that considers whether data utilization is accurate and not narrow, and reliability is a factor that considers whether utilization data is reliable. This can be viewed as a new characteristic because it does not overlap with the existing one. Therefore, it can be seen that the big data characteristics are technically extended from 3V to 5V.

At this time, blockchain can be considered as one of the means for realizing the validity and reliability required in big data. As a decentralized platform, blockchain is a technology that shares data agreed upon by all system participants. To this end, a peer-to-peer (P2P) network and a consensus algorithm are applied, and these two technologies expand the validity and reliability of big data.

Therefore, the combination of blockchain and artificial intelligence provides validity and reliability of data. Blockchain reveals the data sources underlying AI analytics. It also authenticates that the data is trusted by consensus. Not only that, it is also possible to implement distributed artificial intelligence. The principle of blockchain is that participants own all shared data. This allows for individual use of artificial intelligence. Existing centralized systems have all data in one place, and analysis is done in one place. However, in blockchain, the scope of analysis can be determined among participants. This is because the analysis results can be shared and received. The convergence of these two promising technologies is causing a paradigm shift in a broader direction. Accordingly, various services using blockchain and artificial intelligence are being developed and released.

Matters described in this background section are prepared to promote understanding of the background of the invention, and may include matters not already known to those of ordinary skill in the art to which this technology belongs.

The following embodiments have been devised to solve the above-described problem, and a blockchain-based machine that provides a cloud service by using the computing resources of the nodes constituting the blockchain and provides rewards to the nodes that provided the computing resources. An object of the present invention is to provide a method and system for providing a cloud service including a learning service.

Problems to be solved by one embodiment are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the following description.

One aspect of the present invention for achieving the above object provides a cloud service providing method and system including a block chain-based machine learning service.

The blockchain-based cloud service providing system includes: a client terminal requesting a cloud service; a cloud service providing apparatus that manages a block chain network and provides the cloud service to the client terminal using nodes constituting the block chain network as distributed computing devices; and a user terminal that participates in the block chain network as the node, and is provided with block chain resources according to the block chain network as a reward for participation.

The blockchain network is composed of a hybrid blockchain network including a public blockchain network and a private blockchain network at the same time.

The cloud service providing apparatus may provide the cloud service by selecting one of the public blockchain network and the private blockchain network according to the type, latency, and amount of computing resources of the cloud service.

The cloud service providing apparatus may include: a public block chain management unit for managing computing resources of each of the first nodes constituting the public block chain network, and a coin wallet; a private blockchain management unit that authenticates second nodes constituting the private blockchain network and manages computing resources; a framework management unit for configuring and managing a framework for providing the cloud service; a blockchain API broker that relays the public blockchain network to communicate with an external public blockchain network; a development tool management unit that drives a development tool on the web using a computing resource provided through the block chain network and provides it to the client terminal; a resource management unit that interworks with the public block chain management unit and the private block chain management unit to receive computing resources from nodes constituting the block chain network, and provides the received computing resources to the client terminal; and an application driving unit that operates an application on the web in conjunction with the public block chain management unit, the private block chain management unit, and the resource management unit and provides it to the client terminal.

The private blockchain management unit creates a channel, which is a limited messaging path between the first nodes, and controls to exchange confidential information for a specific subset of the first nodes through the channel. can

The private blockchain management unit may control the first nodes so that each of the first nodes performs transaction execution before transaction ordering.

The public blockchain management unit may configure each of the second nodes as an Inter Planetary File System (IPFS)-based distributed cloud storage having the same file system as each other.

The cloud service providing apparatus may perform a machine learning service that performs deep learning or machine learning-based training using the private block chain network and provides a training result to the client terminal.

According to the embodiments as described above, it is possible to provide a distributed cloud environment combined with a block chain to the client, thereby enhancing security and reducing data processing costs by storing, managing, and processing data in a decentralized environment. have.

In addition, by converging block chain technology and artificial intelligence technology, clients can easily perform deep learning and machine learning tasks through reliable big data artificial intelligence analysis.

In addition, by utilizing existing proven open sources, many clients can access AI more closely and use it for work or business, so that individuals can strengthen their work capabilities, companies can increase their productivity, and many platforms (Platform) Developers can easily develop according to their platform concept

Effects of one embodiment are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

1 is a diagram illustrating a cloud service providing system according to an embodiment.
2 is a conceptual diagram of a cloud service providing system according to an embodiment.
3 is a diagram for explaining a public blockchain network according to an embodiment.
4 is a block diagram of a node according to an embodiment.
5 is a diagram conceptually illustrating a block connected to a block chain according to an embodiment.
6 is a diagram for explaining a private blockchain network according to an embodiment.
7 is a diagram for explaining a distributed file system (IPFS) implemented through a public blockchain network according to an embodiment.
8 is a diagram illustrating an architecture for a cloud service providing system according to an embodiment;
FIG. 9 is a detailed diagram illustrating the architecture according to FIG. 8 .
10 is a flowchart of a cloud service providing method according to an embodiment.
11 is a flowchart of a cloud service providing method according to another embodiment.
12 is a block diagram of a user terminal according to an embodiment.
13 is a diagram exemplarily illustrating a hardware configuration of an apparatus for providing a cloud service according to an embodiment.
14 is a diagram illustrating a screen of a user terminal on which a computing resource provision status is output, according to an embodiment.
15 is a diagram illustrating a screen of a user terminal on which a compensation status is output, according to an embodiment.

Advantages and features of the present invention and methods of achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only the present embodiments allow the disclosure of the present invention to be complete, and those of ordinary skill in the art to which the present invention pertains. It is provided to fully understand the scope of the present invention to those skilled in the art, and the present invention is only defined by the scope of the claims.

The terminology used herein is for the purpose of describing the embodiments and is not intended to limit the present invention. In this specification, the singular also includes the plural unless specifically stated otherwise in the phrase. As used herein, “comprises” and/or “comprising” does not exclude the presence or addition of one or more other components in addition to the stated components. Like reference numerals refer to like elements throughout, and "and/or" includes each and every combination of one or more of the recited elements. Although "first", "second", etc. are used to describe various elements, these elements are not limited by these terms, of course. These terms are only used to distinguish one component from another. Accordingly, it goes without saying that the first component mentioned below may be the second component within the spirit of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used herein will have the meaning commonly understood by those of ordinary skill in the art to which this invention belongs. In addition, terms defined in a commonly used dictionary are not to be interpreted ideally or excessively unless specifically defined explicitly.

Hereinafter, an embodiment will be described in detail with reference to the accompanying drawings.

1 is a diagram illustrating a cloud service providing system 10 according to an embodiment.

Referring to FIG. 1 , a cloud service providing system 10 according to an embodiment may include a cloud service providing apparatus 100 , a client terminal 200 , and a user terminal 300 .

Communication between various entities included in the cloud service providing system 10 may be performed through a wired/wireless network. A wired/wireless network may use standard communication technologies and/or protocols.

The cloud service providing device 100, the client terminal 200 and the user terminal 300 in the cloud service providing system 10 are, for example, a computer, an Ultra Mobile PC (UMPC), a workstation, a net-book. , PDA (Personal Digital Assistants), portable computer, web tablet, wireless phone, mobile phone, smart phone, electronic such as PMP (portable multimedia player) As one of the devices, all electronic devices capable of installing and executing an application related to an embodiment may be included. The electronic device may perform overall service operations such as, for example, configuration of a service screen, data input, data transmission/reception, data storage, and the like, under the control of the application.

The cloud service providing device 100 provides a distributed cloud based on a blockchain, and uses the computing resources of nodes constituting the blockchain network to provide various cloud services. It is a platform (Blockchanin Platform).

The computing resource of an embodiment may include idle resources such as CPU, NETWORK, MEMORY, STORAGE, etc. provided by nodes constituting the blockchain network.

The cloud service providing apparatus 100 may borrow computing resources of nodes constituting the blockchain network and provide blockchain resources to the nodes that have provided the computing resources as a reward in return for using the computing resources.

In this specification, block chain resources are described as tokens or coins implemented using block chains, but one embodiment is not limited thereto, and it will include various digital currencies that can be implemented in block chains.

The cloud service providing apparatus 100 may provide various cloud services requested by the client by using input data received from the client terminal 200 .

The input data according to an embodiment is cloud service related data requested by the client terminal 200 . As an example, input data of a machine learning-related cloud service may be composed of data input to perform machine learning.

The cloud service providing apparatus 100 may generate result data as a result of various cloud services by using the input data. The cloud service providing apparatus 100 may provide the generated result data to the client terminal 200 .

The cloud service of one embodiment is an application service (SaaS, Software as a Service) provided by running an application on the web using the computing resources of nodes constituting the block chain network (at the same time constituting the cloud for the cloud service); Infrastructure as a Service (IaaS) that provides computing resources provided from the nodes constituting the block chain network, and various software development tools using the computing resources provided from the nodes constituting the block chain network. It may include a platform service (PaaS, Platform as a Service).

Specifically, the application service is a bio-healthcare application, an application related to the creation and management of a block chain, an application related to streaming, an application to perform analysis using AI, an application related to battery management, and a file sharing and synchronization application , Internet of Things (IoT) device management applications, and the like may include services provided on the web using the cloud. Platform services include Jupyter Notebook, Visual Studio Code, an Ethereum app development tool, Truffle, which are application development tools, and AutoML, an app development tool for machine learning. It may include services provided on the web using the cloud. Infra service is a file distributed in a distributed file system consisting of nodes constituting a block chain network, training according to Machine Learning (ML) or Deep Learning (DL), construction of big data, and It can provide computing resources related to the combination of pieces of

The cloud service providing device 100 serves as a gateway to manage the block chain network, maintains and manages the block chain network, and provides a framework and various software tools required to provide cloud services. It serves to indirectly provide the distributed computing resources of the nodes constituting the blockchain network to the client terminal 200 .

The cloud service providing device 100 implements and provides a block chain platform and a cloud for this purpose, and the cloud service providing device 100 described below provides not only the cloud service providing device 100 itself, but also a cloud service. It may refer to a blockchain platform implemented by the device 100 and a cloud.

The cloud service providing device 100, when the user terminal 300, which is a node constituting the block chain network 20, provides the computing resources necessary to obtain the result data, the user terminal 300 is rewarded with blockchain resources as a reward. can provide

The client terminal 200 is a subject that uses the cloud service and may request a desired cloud service from the cloud service providing apparatus 100 .

The client terminal 200 may transmit input data to the cloud service providing apparatus 100 to request a cloud service, and receive result data from the cloud service providing apparatus 100 as a result of the cloud service.

As an example, when the client terminal 200 requests a cloud service for performing ML/DL, the client terminal 200 provides ML/DL input data (where the input data is machine learning (ML) or deep learning (DL)). (which may include a training data set required to perform ) is transmitted to the cloud service providing device 100 . The cloud service providing apparatus 100 performs ML/DL on input data using computing resources of nodes constituting the block chain network, generates result data, and transmits the generated result data to the client terminal 200 . is to transmit

As an example, when the cloud service providing device 100 provides the cloud service providing device 100 with the right to use the result data transmitted by the client device 200 , the client device 200 also provides the cloud service. Blockchain resources may be provided as a reward from the device 100 . The cloud service providing apparatus 100 may provide result data of the client terminal 200 to which the authority has been provided to another third client terminal.

The user terminal 300 may function as a node constituting a block chain network and may be a distributed computing device for providing a cloud service to the client terminal 200 at the same time. Accordingly, the user terminal 300 may provide the cloud service providing apparatus 100 with a computing resource required to perform the cloud service. As described above, the computing resource of an embodiment means an idle resource such as CPU, NETWORK, MEMORY, STORAGE, etc. provided by the user terminal 300 .

The user terminal 300 may participate as a node in a blockchain network managed by the cloud service providing device 100 . Hereinafter, the user terminal 300 and the node in the block chain will be used with the same meaning.

The user terminal 300 provides a computing resource to the cloud service providing apparatus 100 through a block chain, and the cloud service providing apparatus 100 uses computing resources provided from a plurality of user terminals 300 to provide various cloud services. service can be performed. The user terminal 300 may receive a block chain financial resource from the cloud service providing apparatus 100 as a reward for providing the computing resource.

2 is a conceptual diagram of a cloud service providing system 10 according to an embodiment.

Referring to FIG. 2 , a cloud service providing system 10 according to an embodiment requests a cloud service and receives a request from a client terminal 200 and a client terminal 200 receiving result data for the request. A cloud service providing apparatus that receives, manages the block chain network 20, and uses the nodes 300 constituting the block chain network 20 as a distributed computing device to generate result data and provide it to the client terminal 200 100 , and a block chain network 20 managed and maintained by the cloud service providing device 100 and composed of a plurality of nodes 300 .

The blockchain network 20 may be configured as a hybrid blockchain network including a public blockchain network 21 and a private blockchain network 22 at the same time. Therefore, throughout this specification, the node 300 is a node constituting one of the public blockchain network 21 and the private blockchain network 22, or a private blockchain network while constituting the public blockchain network 21 It may be a node constituting (22).

In addition, the nodes constituting the block chain network 20 perform distributed cloud computing to provide cloud services, and if necessary, record information necessary for cloud services as transactions in the ledger to create a block chain corresponding to the ledger. can Accordingly, information required for cloud service can be trusted by a consensus algorithm according to the block chain, and some nodes selected or agreed upon among nodes can share or distribute information required for cloud service and store it.

In general, a public blockchain network can ensure data transparency through decentralization between all nodes, but has a slow disadvantage, whereas a private blockchain network aims for fast and high performance, but consists of only authorized nodes, so the number of nodes difficult to expand significantly. The blockchain network 20 according to the present invention may employ a hybrid blockchain network to ensure both fast performance, decentralization, and scalability.

In the present invention, the public blockchain network 21 may mean an internal public blockchain network composed of nodes managed by the cloud service providing device 100 . Here, the internal public blockchain network 21 is a blockchain API broker 104 implemented through a private blockchain network 22, including Bitcoin, Ethereum, and EOS. It is configured to communicate with a typical external public blockchain network, such as

The private blockchain network 22 may be a blockchain network composed only of nodes authorized by the cloud service providing device 100 .

The block chain management unit 10 may create and manage the block chain network 20 . The block chain management unit 10 may include a public block chain management unit 101 and a private block chain management unit 102 .

The public block chain management unit 101 may create the public block chain network 21 and manage computing resources, coin wallets, etc. of each of the nodes constituting the public block chain network 21 .

The private blockchain management unit 102 may create the private blockchain network 22 , authenticate the nodes constituting the private blockchain network 22 , and manage each computing resource. For example, the private blockchain management unit 102 may be implemented using Hyperledger fabric, an open source project for an enterprise blockchain led by the Linux Foundation.

Specifically, the private blockchain management unit 102 manages user identification information of nodes constituting the private blockchain network 22 in order to activate the authorized private blockchain network 22, and the private blockchain network A membership service for authenticating nodes constituting (22) may be provided. The private blockchain management unit 102 may create an access control list and provide an additional layer of authority through approval of a specific network operation using the generated access control list.

In one embodiment, the private blockchain management unit 102 permits and provides access to the private blockchain network 22 to the client terminal 200 or nodes constituting the private blockchain network 22 that require a private transaction. can do. The private blockchain management unit 102 may create a channel that is a limited messaging path between nodes constituting the private blockchain network 22 . A channel is a limited messaging path and can be used to provide personal information or such confidential information to a specific subset of nodes constituting the private blockchain network 22 . The channel is not accessed by nodes to which access is not granted.

In an embodiment, the private blockchain management unit 102 may assign a network role to each type of node 400 constituting the private blockchain network 22 . To provide concurrency and parallelism to the network, transaction execution at nodes 400 proceeds prior to transaction sequencing. If the transactions are executed before ordering, each node 400 can process multiple transactions at the same time. Such concurrent execution increases the processing efficiency of each node 400 and accelerates transaction delivery to the scheduling service. In addition to enabling parallel processing, the node 400 is freed from the consensus workload because the division of work does not burden the node scheduling tasks in transaction execution and maintenance demands. One transaction execution process may be executed independently of the verification of other nodes 400 . Accordingly, the private blockchain management unit 102 performs transaction ordering in each node, and then performs transaction ordering, and commits to reflect the ordered transactions in the private blockchain network 22. 400) can be controlled.

In one embodiment, the private blockchain manager 102 may execute a chaincode, which is a program called when receiving a specific type of transaction in the private blockchain network 22 . Specifically, when receiving a specific type of transaction, the private blockchain management unit 102 may execute a chain code through a verified node among the nodes constituting the private blockchain network 22 . A chaincode creates a ledger through processing corresponding to a particular type of transaction received, where the generated ledger may not be accessed when other chaincodes are executed. Chaincode is a concept corresponding to a smart contract on the Ethereum platform, and the chaincode or smart contract may be a program that defines various terms and processes for creating and managing a ledger.

The framework management unit 103 may configure and manage a framework for providing a cloud service using computing resources provided from the nodes 300 constituting the block chain network 20 . As an example, the framework management unit 103 may configure and manage an artificial intelligence framework (AI Framework) for performing machine learning or deep learning.

The blockchain API broker 104 (blockchain API broker) connects the public blockchain network 21 and/or the private blockchain network 22 to common It can be relayed to communicate with external public blockchain networks. For example, the blockchain API broker 104 supports the Ethereum platform-based smart contract and the Hyperledger Fabric-based chain code at the same time, and supports interoperability and conversion by supporting Hyperledger Fabric. It supports compatibility and communication between the base private blockchain network 22 and external public blockchain networks, and compatibility and communication between the public blockchain network 21 it manages and the public blockchain network 21 can support Accordingly, the client terminal 200 may use an external public blockchain network as well as the internal blockchain network 20 through the cloud service providing device 100 .

The development tool management unit 105 may drive a development tool on the web using a computing resource provided through the block chain network 20 and provide it to the client terminal 200 . To this end, the development tool management unit 105 may operate in conjunction with the blockchain management unit 10 . For example, development tools include Jupyter Notebook, which is an application development tool, Visual Studio Code, Truffle, an Ethereum app development tool, and AutoML (AutoML), an app development tool for machine learning. AutoML) and the like.

The resource management unit 106 manages computing resources provided from nodes constituting the block chain network 20 in cooperation with the block chain management unit 10, and manages computing resources according to the request of the client terminal 200. It may be provided to the client terminal 200 . In addition, the resource management unit 106 may provide the computing resources managed by itself to the development tool management unit 105 and the application driving unit 107 .

The application driving unit 107 may operate an application on the web in conjunction with the block chain management unit 10 and the resource management unit 106 to provide it to the client terminal 200 . For example, the application driven by the application driving unit 107 is a bio-health care application, an application related to the creation and management of a block chain, an application related to streaming, an application performing analysis using AI, and battery management. It may be a related application, a file sharing and synchronization application, an Internet of Things (IoT) device management application, and the like.

In addition, the cloud service providing apparatus 100 may further include a compensation unit, a client management unit, and the like.

The compensation unit may provide compensation to the client terminal 200 and the user terminal 300 . The reward unit can provide a reward to the client terminal 200 that allows the use of result data according to machine learning or deep learning (ML/DL), and configures the blockchain network 20 to function as a node It is possible to provide a reward to the user terminal (300).

The compensation unit may determine the compensation size to be provided to the client terminal 200 according to the number and size of the ML/DL result data, and may determine the compensation size provided to the user terminal 300 according to the type, size, provision period, etc. of the computing resource. can decide

The compensation unit may provide the client terminal 200 and the user terminal 300 with block chain resources as a reward, and the block chain resources in an embodiment are digital currency such as tokens or coins implemented in the block chain. may include

The client manager may manage information on the client terminal 200 or information on a user of the client terminal 200 .

3 is a diagram for explaining the public blockchain network 21 according to an embodiment.

Referring to FIG. 3 , the public block chain network 21 according to an embodiment may be composed of a plurality of nodes 300 capable of communicating with each other through a communication network.

First, the blockchain will be described. Blockchain creates online transaction information as a block, connects the generated blocks to each other, and distributes data to the personal digital equipment of participants (peers) rather than the central management server in a peer-to-peer (P2P) network distributed environment. It is a method of jointly managing data by storing

Specifically, the basic structure of a block chain is a collection of blocks connected one after another, and is based on a peer-to-peer (P2P) method. In addition, for a certain period of time, more than half of the participants exchange transaction details with each other, confirm and approve, and only the transaction details agreed upon by digital signatures are made into one block. Then, the newly created block is connected to the previous block chain, and a copy is made and distributed and stored in the user terminals (ie, node 300) of each participant.

Therefore, when blockchain is used, there is no need to separately manage a database for transaction books, so management costs are reduced, and the safety of transactions is improved because it is difficult to hack through distributed processing. The block chain method can be applied to transactions such as stocks and real estate, and can be used in various fields related to security, such as transfer of land rights or home door keys.

In one embodiment, a block chain is implemented with a plurality of nodes 300 as participants, and information distributed and stored in each of the plurality of nodes 300 may be a hash value of a block or the like.

The plurality of nodes 300 may share the information by always maintaining the latest version of the information distributed and stored in each of the plurality of nodes 300 .

Each of the plurality of nodes 300 is a mobile phone, a smart phone, a personal digital assistant (PDA), a portable multimedia player (PMP), any type of handheld that can be connected to an external server through a wireless communication network such as a tablet PC. It may include a wireless communication device based on it, and in addition, it may include a communication device that can be connected to an external server through a network, such as an IPTV including a desktop PC, a tablet PC, a laptop PC, and a set-top box.

Each of the plurality of nodes 300 may be implemented in the form of a terminal as described above, but is not limited thereto, and may be implemented in the form of a server operated by a specific company.

A node implemented in the form of a terminal among the plurality of nodes 300 may store information shared on the block chain in a secure area of the memory, and a node implemented in the form of a server among the plurality of nodes 300 is shared on the block chain information can be stored in the secure area of the database.

That is, in one embodiment, a plurality of nodes 300 are configured based on various types of block chains, such as a block chain in which the participant consists only of a terminal, a block chain in which the participant consists only of a server, and a block chain in which the participant is a mixture of terminals and servers. can do.

The user terminal 300 according to an embodiment participates as one node on the block chain network to generate a data block and provide it to the public block chain network 21 . In this case, at least some of the user terminals 300 may detect forgery of the data block.

4 is a block diagram of a node 300 according to an embodiment.

Each of the plurality of nodes 300 in the block chain network 20 generates (or mines) a block for recording each block chain transaction on the block chain network 20 according to a predetermined algorithm and / Or you can verify. According to an embodiment, the blockchain transactions distributed on the blockchain network 20 through each node 300 during a predetermined time interval may be stored together in a newly created block. According to an embodiment, each of the nodes 300 may store at least a portion of a block chain (ie, a block chain distributed ledger) for the block chain network 20 according to a predetermined algorithm.

Referring to FIG. 4 , the node 300 includes a transaction processing module 402 , a communication module 404 , a block generation/verification module 406 , and a blockchain distributed ledger storage module 408 .

According to an embodiment, the transaction processing module 402 may receive a predetermined blockchain transaction and distribute the received transaction to other nodes 400 on the blockchain network 20 . According to one embodiment, the transaction processing module 402 may receive a transaction distributed by other nodes 400 on the blockchain network 21 .

According to an embodiment, the transaction processing module 402 may process the received transaction (eg, including execution of each smart contract included in the transaction, but the present disclosure is not limited thereto).

According to an embodiment, the transaction received and processed by the transaction processing module 402 may include various hash values for various data generated for the cloud service by the cloud service providing device 100 or various types of data generated for the cloud service. It may include a tracker indicating a location where data is distributed and stored in the nodes 300 , a data history exchanged between the client terminal 200 and the cloud service providing apparatus 100 for a cloud service, and the like. According to one embodiment, the transactions received and processed by the transaction processing module 402 are other various transactions (eg, distributed by any node 400 to at least some of the nodes on the blockchain network 20 ). Transaction information of digital currency or other smart contract information supported on the corresponding block chain network 21, and the present disclosure is not limited thereto.

According to an embodiment, the communication module 404 may operate so that the node 300 may communicate with other nodes 300 according to a predetermined protocol on the blockchain network 20 . According to one embodiment, the communication module 404 may cause a block chain transaction to be distributed on the block chain network 20 through a communication network according to a predetermined protocol, and also from other nodes 300 through the communication network. It is possible to receive various information on the blockchain network 20 .

According to one embodiment, the block generation/verification module 406 generates a block for the blockchain network 20, collects the transactions that occurred on the blockchain network 20 for a predetermined time interval, and collects the appropriate header information together with the It can be written to the generated block. According to an embodiment, the block generation/verification module 406 may notify the generated block on the blockchain network 20 through the communication module 404 . According to an embodiment, the block generation/verification module 406 may perform verification on a block generated by another node 300 known on the blockchain network 20 .

According to an embodiment, the blockchain distributed ledger storage module 408 may store at least a portion of the blockchain distributed ledger (ie, a distributed ledger including all the history of transactions) according to a predetermined algorithm. According to an embodiment, the blockchain distributed ledger storage module 408 may also store, according to a predetermined algorithm, at least a portion of information about a smart contract composed of a blockchain on the blockchain network 20, but The present disclosure is not limited thereto.

5 is a diagram conceptually illustrating a block 500 connected to a block chain according to an embodiment.

5 is a diagram conceptually illustrating the configuration of a block chain and each block 500 included therein, which can be stored in the blockchain distributed ledger storage module 408 included in the node 300 according to an embodiment. to be. As shown, the block chain includes a plurality of blocks connected to each other, and each block 500 may include a block header 502 and a plurality of transaction information 504a-504n.

According to an embodiment, the block header 502 includes a hash value of a previous block header, a nonce value, a merkle root of transaction information 504a to 504n to be included in the corresponding block 500, a block ( 500) may include a timestamp indicating the creation time, and a value such as the mining difficulty of the corresponding block 500.

According to an embodiment, the plurality of transaction information 504a - 504n included in the block 500 is a plurality of transactions distributed on the blockchain network 20 from the nodes 300 during a predetermined time interval. can

According to one embodiment, the transaction information 504a - 504n of the block 500 is a transaction distributed on the block chain network 20 through the node 300, that is, the cloud service providing device 100 provides a cloud service. It may include transaction information related to various transactions and transmission/reception information generated for the purpose.

The description of the node 300 described with reference to FIG. 4 and the block 500 described with reference to FIG. 5 may be commonly or selectively applied to the public blockchain network 21 and the private blockchain network 22. interpreted

6 is a diagram for explaining the private blockchain network 22 according to an embodiment.

The private blockchain network 22 is also called a private blockchain, a closed blockchain, a permissioned blockchain, an enterprise blockchain, or an enterprise blockchain, and transaction history and various actions other than authorized participants are not shared and tracked. In order to participate as the node 300 in the private blockchain network 22, only a participant authorized by a specific subject can participate as the node 300 and create a block.

In one embodiment, the private blockchain network 22 is managed by the cloud service providing device 100 , and the user terminal 300 needs to obtain permission from the cloud service providing device 100 to access the private blockchain network 22 . You can participate as a node 300 .

The private blockchain network 22 of an embodiment may support server redundancy (HA, High Availability) using Hyperledger Fabric. For example, the cloud service providing device 100 copies and stores information distributed and stored by a part of the nodes 300 constituting the private blockchain network 22 to other nodes 300, and a specific node ( At least some of the roles given to the cloud service to the other node 300 may be controlled to be replaced by the other node 300 .

The shared ledger managed by the blockchain through the private blockchain network 22 is a world state, a database that records the ledger states at a specific point in time, and a transaction log that records information about currently generated transactions. (transaction log) may be included. Information recorded in the transaction log can be updated with the world state. In particular, the shared ledger further includes a data area that can replace the world state, and the data area can record before and after the state change of the world state. When the chaincode is executed, it can only interact with that world state to reference and change the record of the world state.

An individual shared ledger is created and managed for each channel configured in the private blockchain network 22, and the node 300 can keep a copy of the shared ledger for channels that it can access.

The nodes 300 of the private blockchain network 22 can reach a consensus by verifying the order and results of the transactions for constituting the block 500 according to the policy criteria specified in advance. At this time, the nodes 300 create blocks through verification of the validity, order, and results of each transaction, and add the generated blocks to the shared ledger to form a block chain by referring to the world state. You can check the version of the shared ledger agreed upon between them.

On the other hand, at least some of the nodes 300 constituting the private blockchain network 20 are verified by the cloud service providing device 100 (distinguished from verification for participation in the private blockchain network 20), and the node 300 ), the verification can be determined according to the security level, computing resources, etc.) as a node that transmits and receives all events and transactions occurring in the private blockchain network 20 and shares them with each other, and at least the rest of the nodes 300 are in the cloud As a node that has not been verified by the service providing device 100 , limited event processing and management of REST API services may be performed.

7 is a diagram for explaining a distributed file system (IPFS) implemented through the public blockchain network 21 according to an embodiment.

In an embodiment of the present invention, the public block chain management unit 101 is a distributed cloud based on Inter Planetary File System (IPFS) having the same file system as each of the nodes 300 constituting the public block chain network 21 . It can be configured as a repository.

The public blockchain management unit 101 receives a file to be stored from the client terminal 200, splits the provided file into a plurality of file fragments, and divides each of the divided file fragments into public Each of the nodes 300 constituting the block chain network 21 may be distributed and stored. At this time, the public block chain management unit 101 obtains the encryption key stored in the block chain managed by the public block chain network 21, encrypts each of the file fragments using the obtained encryption key, and encrypts the encrypted file fragment Each of these may be distributed and stored in each of the nodes 300 .

In addition, the public blockchain management unit 101 generates a hash value corresponding to a file or fragments thereof distributed and stored in each node 300 , and manages the generated hash value by the public blockchain network 22 . It can be stored on a block chain that is used, and the hash value can be provided to the client terminal 200 . After receiving the hash value from the client terminal 200, the public block chain management unit 101 determines the location of the file fragments corresponding to the hash value, and refers to the location of the determined file fragments are distributed and stored in each node 300. The file may be merged, and the merged file may be returned to the client terminal 200 .

8 is a diagram exemplarily illustrating an architecture for a cloud service providing system according to an embodiment, and FIG. 9 is a diagram illustrating the architecture according to FIG. 8 in detail.

As described above, the blockchain network 20 of one embodiment is divided into a private blockchain network 22 and a public blockchain network 21, and the public blockchain network 21 is an IPFS-based distributed file storage. It can function and interwork with each other through an external public blockchain network and the blockchain API broker 104 .

The cloud service providing device 100 may select and use one of the external public blockchain network, the internal public blockchain network 21, and the private blockchain network 22 according to the type of cloud service and the required latency.

The public blockchain network 21 can be equipped with server redundancy (HA) using the aforementioned Hyperledger Fabric, and an Ethereum Virtual Machine (EVM) provided by Hyperledger Fabric. ) can support Ethereum Smart Contract and System Chain code at the same time. Therefore, the blockchain API broker 104 can communicate by relaying the public blockchain network 21 with an external public blockchain network that uses an Ethereum-based smart contract or a hyperledger fabric-based chain code. can make it

That is, the public blockchain network 21 is able to communicate with an external public blockchain network through the Blockchain API Broker 104, and the cloud service providing device 100 provides external developers and users. According to the type of service or task they require, it is possible to manage and control the public blockchain network 21 by selecting either an Ethereum-based smart contract or a Hyperledger Fabric-based chaincode.

As an example, the private blockchain 22 is applicable even when fast data processing is required, such as payment or asset management, and provides developers with various development tools, while providing an open platform for businesses and individuals. It can be designed so that the cloud service providing platform can be easily used in the form.

As one of the software frameworks managed by the framework management unit 103 , an artificial intelligence framework (AI Framework) may be configured. The artificial intelligence framework can be composed of a three-tiered deep learning framework, 'Keras' on the 1st floor, Google 'Tensorflow' on the 2nd floor, Facebook 'Pytorch', and Amazon ' It is composed of a framework such as 'Mxnet', and the third layer (3rd Floor) may be composed of 'ONNX (Model exchange)'. Accordingly, it may be compatible with most of the currently released clouds.

Accordingly, the cloud service providing device 100 uses 'Keras', Google 'Tensorflow', Facebook 'Pytorch', and Amazon 'Mxnet' configured in the first layer (1st Floor) and the second layer (2nd Floor) to deep By performing deep learning, performing machine learning with 'Scikit learn' on the 2nd floor, configuring data for machine learning with 'Pandas', and visualizing data using 'Bokeh' Statistical analysis and data analysis required for machine learning or deep learning (ML/DL) can be performed at the same time.

In addition, the third layer may further include 'Horovod'. Horovod is a framework that supports distributed training in a multi-GPU environment. Based on the Bandwidth Optimal All-reduce Algorithm Paper, ‘Horovod’ can operate in a Ring-all reduce method with optimized bandwidth usage.

Accordingly, the cloud service providing apparatus 100 may perform training required for machine learning or deep learning based on Horovod by using the nodes 300 constituting the block chain network 20 to be distributed.

Specifically, the cloud service providing apparatus 100 may provide a deep learning or machine learning service to the client terminal 200 using the private blockchain network 22 . The private blockchain network 22 may be suitable for providing deep learning or machine learning services because it guarantees continuous and fast computing speed and has high security.

Specifically, the cloud service providing device 100 separately stores a learning model (artificial neural network) to be trained for each node 300 constituting the private blockchain network 22, and stores each other in the individually stored learning model. Different training data can be individually trained one by one using a forward propagation algorithm and a backpropagation algorithm. Here, the forward propagation algorithm is a learning method that updates the weight while applying a weight while moving to the input layer, the hidden layer, and the output layer constituting the learning model, and the back propagation algorithm is the output result of the output layer of the learning model and the desired output. It may be a learning method for updating weights to minimize an error between them.

The cloud service providing apparatus 100, but each node 300 controls to transmit the weight obtained by training the learning model to the other node 300, each node 300, ring-all-reduce (Ring all) Reduce) algorithm to obtain weights obtained by the other node 300 through training. Here, the ring-all reduce algorithm is, 2D-HRA: Two-Dimensional Hierarchical Ring-Based All-Reduce Algorithm in Large-Scale Distributed Machine Learning, Youhe Jiang, Huaxi Gu, Yunfeng Lu, Xiaoshan Yu, 183488p ~ 183494p, 2020) can refer to Next, each node 300 may update the learning model it stores to have the average or median value of the obtained weights as a weight.

The cloud service providing apparatus 100 according to an embodiment of the present invention repeatedly performs the above-described individual learning and updating using the nodes 300 constituting the private blockchain network 22, thereby providing extensive training of the learning model. The training of data can be distributed and processed.

In addition, when the cloud service providing device 100 provides machine learning or deep learning services through the private blockchain network 22, only the authenticated node 300 can participate, providing continuous and large amount of resources. It can be difficult to guarantee.

In order to solve this problem, the cloud service providing apparatus 100 may designate and manage at least some of the nodes 300 constituting the private blockchain network 22 as master nodes. A node designated as a master node is forced to continuously provide more than the minimum computing resource allocated by the cloud service providing device 100, and the cloud service providing device 100 may change the designation of such a master node as it monitors computing resources. can For example, the cloud service providing device 100 monitors the computing resources provided by each of the nodes 300 constituting the private blockchain network 22, and the amount of computing resources provided for a preset period is preset. It is possible to designate and manage the node 300 that is more than the reference value as the master node, and when the computing resource provided from the master node decreases by more than the reference value, the master node may be changed.

In addition, the cloud service providing device 100 of an embodiment creates and provides a Jupyter Notebook, which is one of the development tools, so that the client terminal 200 can create and share a visualization, and the shared code and data collection Interactive editing is also supported. Jupyter Notebook combines code, comments, multimedia, visualizations, and more in an interactive document called a 'notebook' and provides the ability to share, reuse and rework, and runs through a web browser. Uh, Jupyter Notebook itself can be hosted using the public blockchain network 21 .

Most of the code sharing methods provided by the cloud service are not interactive, but Jupyter Notebook in an embodiment can check, run, and display the code directly in a web browser, and because the code is fixed, the web browser Reflecting the feedback provided directly from , you can make small edits and re-execute in real time, and you can embed Jupyter Notebook in a user control device that can be used as a code input source.

Referring to FIG. 9 , the cloud service providing system 10 according to an embodiment may be configured in an architecture having three layers targeting the above-described cloud service providing apparatus 100 and the node 300 .

The first layer may operate in a Kubernetes Platform environment based on a Linux operating system. Kubernetes may be an open source-based management system that provides automatic deployment and scaling of containerized applications.

Kubernetes in one embodiment supports container configuration and automation of tasks, does not require manual processes to deploy and scale containerized applications, and enables uninterrupted service. Accordingly, the cluster can be managed efficiently and the entire host can be expanded.

In Kubernetes of an embodiment, Docker in the form of an image is supported to serve as a container. Because it is difficult to install all of the Deep Learning Framework on the private blockchain network 22 because there are various versions of Linux, configure Docker, reserve a container in a cluster of virtual machines, and immediately run the desired image. can

In Kubernetes of an embodiment, a Hyperledger Fabric Docket is first installed to perform a task related to a coin and a token. Due to security problems, we sample coins generated externally and link them, and depending on the system, if there is a lot of data embedded in the database (DB), the internalization process is not easy. It is also possible to run Hadoop HDFS for processing large amounts of data, such as big data, within the private blockchain network 22 .

10 is a flowchart of a cloud service providing method according to an embodiment.

Referring to Figure 10, the cloud service providing method according to an embodiment includes a computing resource information receiving step (S100), a computing resource information storage step (S110), an input data receiving step (S120), an input data storing step (S130), It may include a result data generation step (S140), a result data transmission step (S150), and a compensation provision step (S160).

The cloud service providing method may be performed by the above-described cloud service providing apparatus 100 .

First, in the computing resource information receiving step ( S100 ), the controller 160 may receive computing resource information from the user terminal 300 .

Computing resource information according to an embodiment may include identification information of the user terminal 300 , a type of a computing resource, a size of a computing resource that can be provided, a provision period, and the like.

In addition, in the computing resource information storage step ( S110 ), the controller 160 may store a portion of the received computing resource information by transacting it in the blockchain managed by the blockchain network 20 . The control unit 160 may distribute and store another part of the computer resource information in the individual nodes 300 constituting the block chain network 20 .

Then, in the input data receiving step ( S120 ), the controller 160 may receive input data from the client terminal 200 .

The input data according to an embodiment is cloud service related data requested by the client terminal 200 , and in the case of an ML/DL service, may be data for ML/DL analysis.

In addition, in the input data storage step ( S130 ), the control unit 160 may store at least a portion of the received input data by transacting it in a block chain managed by the block chain network 20 . The control unit 160 may also store another part of the input data in the node 300 constituting the block chain network 20 .

Then, in the result data generation step ( S140 ), the controller 160 uses the computing resource information to check the currently available computing resources. In addition, the controller 160 may generate result data using input data through available computing resources.

In an embodiment, the computing resource information may further include a priority of the computing resource. In this case, the controller 160 may schedule the available computing resources by using the priority of the computing resources.

The controller 160 may calculate the computing resources required to generate the result data for each time. In addition, the control unit 160 may generate an hourly usage plan of the available computing resources according to the calculated required computing resources.

The control unit 160 may check the available computing resources of the user terminal 300 according to the priority, and may schedule the computing resources required to generate result data.

The result data of an embodiment is a result of the cloud service, and in the case of an ML/DL service, it may be ML/DL analysis data.

Then, in the result data transmission step ( S150 ), the controller 160 may transmit the generated result data to the client terminal 200 .

And, in the step of providing compensation ( S160 ), the control unit 160 may provide, as a reward, blockchain resources to the user terminal 300 that provided the corresponding computing resource according to the computing resource used when generating the result data. .

11 is a flowchart of a cloud service providing method according to another embodiment.

Referring to FIG. 11 , a method of providing a cloud service according to another embodiment is a method of storing input data, including an input data analysis step (S131), a private block chain condition determination step (S132), a public block chain transaction step (S133), It may include a private blockchain transaction (S134) and a cloud storage step (S135).

First, in the input data analysis step (S131), the control unit 160 analyzes the input data, such as the type of cloud service requested by the client terminal 200, the size of the input data, the processing request speed required to process the input data, etc. can be derived.

Then, in the private block chain condition determination step S132 , the control unit 160 determines whether it is necessary to use the private block chain network 22 for the cloud service requested by the client terminal 200 using the analysis result of the input data. can do.

For example, the control unit 160 may determine whether it is necessary to use the private blockchain network 22 according to a preset criterion. The controller 160 may determine that it is necessary to use the private blockchain network 22 when the input data size is less than or equal to a preset size or when the processing request speed of input data is greater than or equal to a preset speed. That is, when the input data is not large, so many nodes are not required, or when a high speed is required to process the input data, the control unit 160 may determine that it is necessary to use the private blockchain network 22 .

For example, when the input data is training data according to machine learning/deep learning, the controller 160 may determine to use the private blockchain network 22 .

And, when it is determined that it is not necessary to use the private blockchain network 22 in the public blockchain transaction step S133 , the controller 160 may transact and store the input data in the public blockchain 21 .

And, when it is determined that it is necessary to use the private blockchain network 22 for the private blockchain transaction (S134), the control unit 160 transacts at least a portion of the input data to the private blockchain network 22 to create a blockchain And, in the cloud storage step (S135), the control unit 160 in the input data is not configured as a block chain data to the node 300 constituting the block chain network 20 It can be distributed and stored.

12 is a block diagram of a user terminal 300 according to an embodiment.

Referring to FIG. 12 , the wireless communication unit 310 includes one or more components for performing wireless communication between the user terminal 300 and the wireless communication system or wireless communication between the user terminal 300 and the network in which the user terminal 300 is located. It can contain elements. For example, the wireless communication unit 310 may include a broadcast reception module 311 , a mobile communication module 312 , a wireless Internet module 313 , a short-range communication module 314 , and a location information module 315 , etc. .

The broadcast reception module 311 receives a broadcast signal and/or broadcast related information from an external broadcast management server through a broadcast channel. Here, the broadcast channel may include a satellite channel and a terrestrial channel. Meanwhile, broadcast-related information may be provided through a mobile communication network, and in this case, may be received by the mobile communication module 312 .

In addition, the mobile communication module 312 transmits and receives a wireless signal to and from at least one of a base station, an external terminal, and a server on a mobile communication network. Here, the wireless signal may include various types of data according to transmission and reception of a voice call signal, a video call signal, or a text/multimedia message.

The wireless Internet module 313 refers to a module for wireless Internet access, and may be built-in or external to the user terminal 300 .

The short-range communication module 314 refers to a module for short-range communication. As the short-distance communication technology, Bluetooth, Radio Frequency Identification (RFID), Infrared Data Association (IrDA), Ultra Wideband (UWB), ZigBee, etc. may be used.

Also, the location information module 115 is a module for confirming or obtaining the location of the user terminal 300 . An example is a Global Position System (GPS) module. The GPS module receives location information from a plurality of satellites. Here, the location information may include coordinate information indicated by latitude and longitude.

Meanwhile, the A/V (Audio/Video) input unit 320 is for inputting an audio signal or a video signal, and may include a camera 321 , a microphone 322 , and the like. The camera 321 processes an image frame such as a still image or a moving image obtained by an image sensor in a video call mode or a shooting mode. Then, the processed image frame may be displayed on the display unit 351 .

The image frame processed by the camera 321 may be stored in the memory 360 or transmitted to the outside through the wireless communication unit 310 . Two or more cameras 321 may be provided according to the configuration of the user terminal 300 .

The microphone 322 receives an external sound signal by a microphone in a call mode, a recording mode, a voice recognition mode, and the like, and processes it as electrical voice data. In addition, the processed voice data may be converted into a form that can be transmitted to the mobile communication base station through the mobile communication module 312 and output in the call mode. The microphone 322 may implement various noise removal algorithms for removing noise generated in the process of receiving an external sound signal.

The user input unit 330 receives an input operation from the user and generates input data for controlling the operation of the user terminal 300 .

The sensing unit 340 detects the current state of the user terminal 300 , such as the location of the user terminal 300 , the presence of user contact, the orientation of the user terminal 300 , and acceleration/deceleration of the user terminal 300 . A sensing signal for controlling the operation of the terminal 300 is generated.

The interface unit 370 serves as an interface with all external devices connected to the user terminal 300 . For example, wired/wireless headset ports, external charger ports, wired/wireless data ports, memory card ports, ports for connecting devices equipped with identification modules, audio input/output (I/O) ports, It may include a video input/output (I/O) port, an earphone port, and the like.

The output unit 350 is for outputting an audio signal, a video signal, or an alarm signal, and may include a display unit 351 , a sound output module 352 , an alarm unit 353 , and the like.

The display unit 351 displays and outputs information processed by the user terminal 300 . For example, when the terminal is in a call mode, a user interface (UI) or graphic user interface (GUI) related to a call is displayed. And, when the user terminal 300 is in the video call mode or the shooting mode, the captured and/or received image or UI and GUI are displayed.

Meanwhile, as described above, when the display unit 351 and the touch pad form a layer structure to form a touch screen, the display unit 351 may be used as an input device in addition to an output device. The display unit 351 includes a liquid crystal display, a thin film transistor-liquid crystal display, an organic light-emitting diode, a flexible display, a three-dimensional display ( 3D display). In addition, two or more display units 351 may exist according to an implementation form of the user terminal 300 . For example, the user terminal 300 may be provided with an external display unit (not shown) and an internal display unit (not shown) at the same time.

The sound output module 352 outputs audio data received from the wireless communication unit 310 or stored in the memory 360 in call signal reception, call mode or recording mode, voice recognition mode, broadcast reception mode, and the like. In addition, the sound output module 352 outputs a sound signal related to a function (eg, a call signal reception sound, a message reception sound, etc.) performed in the user terminal 300 . The sound output module 352 may include a speaker, a buzzer, and the like.

The alarm unit 353 outputs a signal for notifying the occurrence of an event in the user terminal 300 . Examples of events occurring in the terminal include call signal reception, message reception, and key signal input.

The memory 360 may store a program for processing and control of the controller 380, and provides a function for temporary storage of input/output data (eg, phonebook, message, still image, video, etc.). can also be done

The memory 360 may include a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (eg SD or XD memory, etc.), a RAM (RAM, Random Access Memory) SRAM (Static Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory) magnetic memory, magnetic disk, It may include at least one type of storage medium among optical disks.

The controller 380 generally controls the overall operation of the terminal. For example, it performs related control and processing for voice calls, data communications, video calls, and the like. Also, the controller 380 may include a multimedia module 181 for playing multimedia. The multimedia module 381 may be implemented within the control unit 380 or may be implemented separately from the control unit 380 .

The controller 380 may provide a computing resource to the cloud service providing apparatus 100 .

The control unit 380 communicates with the blockchain networks 21 and 22 so that the user terminal 300 participates as a node 400 of the blockchain networks 21 and 22 implemented by the cloud service providing device 100 . can do.

The control unit 380 may provide computing resources through the blockchain networks 21 and 22 , and as a reward for providing the computing resources, may receive blockchain financial resources from the cloud service providing device 100 .

The controller 380 may output the computing resource status provided to the cloud service providing device 100 and the compensation status provided from the cloud service providing device 100 to the user.

The power supply unit 390 receives external power and internal power under the control of the control unit 380 to supply power required for operation of each component.

13 is a diagram exemplarily illustrating a hardware configuration of the cloud service providing apparatus 100 according to an embodiment.

13 , the cloud service providing apparatus 100 may include a communication unit 110 , an input unit 120 , an output unit 130 , a memory 140 , a power supply unit 150 , and a control unit 160 . can The configurations described below should be understood as examples, and may be appropriately added or changed to hardware capable of supercomputing at an appropriate level for configuring and managing the blockchain network 20 .

The communication unit 110 may communicate with an external device through various communication methods. For example, the communication unit 110 may transmit and receive various data by performing communication with the external client terminal 200 and the user terminal 300 .

The input unit 120 may receive various inputs from the user of the cloud service providing apparatus 100 and transmit it to the control unit 160 . In particular, the input unit 120 may include a touch sensor, a (digital) pen sensor, a pressure sensor, a key, or a microphone. The touch sensor may use, for example, at least one of a capacitive type, a pressure sensitive type, an infrared type, and an ultrasonic type. The (digital) pen sensor may be, for example, a part of a touch panel or may include a separate recognition sheet. The key may include, for example, a physical button, an optical key, or a keypad. The microphone is a component for receiving the manager's voice and may be provided inside the cloud service providing device 100, but this is only an example, and it is provided outside the cloud service providing device 100 to provide a cloud service providing device ( 100) may be electrically connected to.

The output unit 130 may provide various screens. For example, the output unit 130 may output the computing resource status, compensation status, client information, user information, and the like.

The memory 140 may store commands or data related to at least one other component of the cloud service providing apparatus 100 . In particular, the memory 140 may be implemented as a non-volatile memory, a volatile memory, a flash-memory, a hard disk drive (HDD), or a solid state drive (SSD). The memory 140 is accessed by the controller 160 , and reading/writing/modification/deletion/update of data by the controller 160 may be performed. In the present invention, the term "memory" refers to a memory 140, a ROM (not shown) in the controller 160, a RAM (not shown), or a memory card (not shown) mounted in the cloud service providing device 100 (eg, micro SD card, memory stick). In addition, programs and data for configuring various screens to be displayed on the display area of the output unit 130 may be stored in the memory 140 .

The power supply unit 150 receives external power and internal power under the control of the control unit 160 to supply power necessary for the operation of each component.

The control unit 160 is electrically connected to the communication unit 110 , the input unit 120 , the output unit 130 , the memory 140 , and the power supply unit 150 to control the overall operation and function of the cloud service providing apparatus 100 . It may consist of a controlling processor. In particular, the controller 160 may provide a cloud service using various modules stored in the memory 140 .

It will be apparent that various operations on the cloud service providing apparatus 100 described in the entire specification (or components 101 to 107 according to FIG. 2 ) may be performed under the control of the controller 160 .

14 is a diagram illustrating a screen of the user terminal 300 on which a computing resource provision status is output, according to an embodiment.

The user terminal 300 may output the current state of the computing resource provided to the cloud service providing apparatus 100 to the user.

As shown in the drawing, the computing resources provided by the user terminal 300 to the cloud service providing apparatus 100 may include idle resources such as CPU, NETWORK, MEMORY, and STORAGE of the user terminal 300 .

As such, the user terminal 300 may receive the current state of computing resources that it provides for the cloud service from the cloud service providing device 100 and display it to the user.

15 is a diagram illustrating a screen of the user terminal 300 on which a compensation status is output according to an embodiment.

The user terminal 300 may output the compensation status provided from the cloud service providing device 100 to the user.

As shown in the drawing, the user terminal 300 may be provided with block chain resources as a reward from the cloud service providing device 100, and the block chain resources in an embodiment are block chain tokens and block chain coins. ) may include digital currencies such as However, the blockchain financing of one embodiment is not limited thereto, and may include various financial resources that can be implemented in the blockchain.

The steps of a method or algorithm described in relation to an embodiment of the present invention may be implemented directly in hardware, as a software module executed by hardware, or by a combination thereof. A software module may contain random access memory (RAM), read only memory (ROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, hard disk, removable disk, CD-ROM, or It may reside in any type of computer-readable recording medium well known in the art to which the present invention pertains.

As mentioned above, although embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art to which the present invention pertains know that the present invention may be embodied in other specific forms without changing the technical spirit or essential features thereof. you will be able to understand Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive.

Claims (1)

It is a cloud service providing system that includes a block chain-based machine learning service,
a client terminal requesting a cloud service;
a cloud service providing apparatus that manages a block chain network and provides the cloud service to the client terminal using nodes constituting the block chain network as distributed computing devices; and
and a user terminal that participates in the blockchain network as the node, and is provided with blockchain resources according to the blockchain network as a reward for participation,
The blockchain network is composed of a hybrid blockchain network including a public blockchain network and a private blockchain network at the same time,
The cloud service providing device,
Selecting one of the public blockchain network and the private blockchain network according to the type and latency of the cloud service, and the amount of computing resources to provide the cloud service,
A machine learning service that performs training based on deep learning or machine learning using the private blockchain network and provides a training result to the client terminal is performed,
The cloud service providing device,
a public block chain management unit that manages computing resources and coin wallets of each of the first nodes constituting the public block chain network;
a private blockchain management unit that authenticates second nodes constituting the private blockchain network and manages computing resources;
a framework management unit for configuring and managing a framework for providing the cloud service;
a blockchain API broker that relays the public blockchain network to communicate with an external public blockchain network;
a development tool management unit that drives a development tool on the web using the computing resource provided through the block chain network and provides it to the client terminal;
a resource management unit that interworks with the public block chain management unit and the private block chain management unit to receive computing resources from nodes constituting the block chain network, and provides the received computing resources to the client terminal; and
Comprising an application driving unit that operates an application on the web in conjunction with the public blockchain management unit, the private blockchain management unit, and the resource management unit and provides it to the client terminal,
The cloud service providing device,
A learning model to be trained is individually stored in each of the second nodes constituting the private block chain network, and different learning data is applied to the learning model stored for each second node. A forward propagation algorithm and a backpropagation algorithm is individually trained using , and the weight obtained by learning in each of the second nodes is controlled to be transmitted to other second nodes except for itself,
Each of the second nodes acquires weights obtained through training of other second nodes using a Ring all Reduce algorithm, and has an average or median value of the obtained weights as a weight. Update the learning model to store,
The framework management unit, but manages an artificial intelligence framework (AI framework) that is one of the software frameworks,
The artificial intelligence framework is at least one of 'Keras' as the first layer (1st Floor), Google's 'Tensorflow' as the second layer (2nd Floor), Facebook's 'Pytorch', and Amazon's 'Mxnet', A cloud service providing system that includes 'ONNX (Model exchange)' as the 3rd floor.

Images