CN114428978A - Industrial Internet block chain trusted computing architecture and method based on computational prediction machine - Google Patents
Industrial Internet block chain trusted computing architecture and method based on computational prediction machine Download PDFInfo
- Publication number
- CN114428978A CN114428978A CN202111471798.4A CN202111471798A CN114428978A CN 114428978 A CN114428978 A CN 114428978A CN 202111471798 A CN202111471798 A CN 202111471798A CN 114428978 A CN114428978 A CN 114428978A
- Authority
- CN
- China
- Prior art keywords
- chain
- machine
- calculation
- prediction
- computing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/64—Protecting data integrity, e.g. using checksums, certificates or signatures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/27—Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/50—Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems
- G06F21/57—Certifying or maintaining trusted computer platforms, e.g. secure boots or power-downs, version controls, system software checks, secure updates or assessing vulnerabilities
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Bioethics (AREA)
- General Health & Medical Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an industrial Internet block chain trusted computing architecture and a method based on a computational prediction machine, wherein the architecture comprises an industrial Internet equipment layer, a block chain layer and a prediction machine layer; the industrial Internet equipment layer comprises various heterogeneous industrial Internet equipment, and industrial computing tasks are generated in the process of managing the equipment; the block chain layer comprises an intelligent contract and an intelligent contract engine is used as a carrier to execute industrial computing tasks; the prediction machine layer comprises a calculation prediction machine and an on-chain prediction machine and is used for improving the computing power of the intelligent contract. The invention designs a block chain calculation system based on a computational prophetic machine, which meets the large calculation requirement in an industrial scene; in an excitation mechanism of the prediction machine network under the chain, a matching algorithm based on the node computing power is provided, the node computing power is considered in the matching process of the prediction machine under the chain and the computing task, and the network total delay of the prediction machine is reduced.
Description
Technical Field
The invention belongs to the technical field of block chains, and particularly relates to an industrial internet block chain trusted computing architecture and method based on a computational prediction machine.
Background
The industrial internet connects and controls different resources in an industrial scene, has poor tolerance to delay, and needs a real-time and efficient network architecture and an industrial application operation mechanism. However, the conventional centralized cloud computing cannot meet the requirement because a large number of nodes are uploaded uniformly to generate network data congestion. Therefore, in order to meet the requirement of 'large connection and low time delay' in an industrial scene, the block chain frame with the decentralized characteristic is a reliable choice for networking the industrial internet.
Under the block chain architecture, the industrial internet carries out decentralized networking, and point-to-point connection among massive nodes is achieved. Meanwhile, the distributed storage of the data solves the problems of single point failure and network congestion of data uploading and downloading. In particular, aiming at the requirements of various industrial applications, the open intelligent contract platform provides a credible solution for the deployment of the industrial applications in the industrial internet based on the block chain. The application developer solidifies the industrial application in the intelligent contract platform in the form of codes, once the trigger condition is met, the industrial application is automatically executed, all relevant data in the whole process is linked up, and any entity cannot modify the contract and the contract running result. However, the ability of current smart contracts has two major drawbacks. Firstly, the intelligent contracts have limited computing power and are difficult to support some complex industrial applications; secondly, Gas costs generated by intelligent contracts on the rented face chain are expensive. Thus, the computational prediction machine network expands the computational power of the intelligent contract by virtue of its abundant computational resources and trusted execution mechanisms.
The computational prophetic network is divided into two types, namely centralization and decentralization. In the centralized computing prediction machine, a single computing prediction machine serves an industrial application task of an intelligent contract, so that task computing failure caused by single-point downtime and unsuccessful solution is easy to occur. In the decentralized prediction machine, for one calculation task, a plurality of calculation prediction machines carry out calculation simultaneously, the credibility of a calculation result is ensured through a consensus mechanism, and meanwhile, the success rate of the calculation task is improved. However, the existing decentralized calculation prediction machine scheme only needs to select a calculation prediction machine group by a random method, neglects the matching of prediction machines with different calculation capabilities and intelligent contract calculation tasks with different complexities, causes huge calculation delay and cannot meet the requirement of low delay of the industrial internet.
Disclosure of Invention
The invention aims to provide an industrial internet block chain trusted computing architecture and method based on a computing prediction machine, so as to meet the large computing requirement in an industrial scene and reduce the computing time delay.
The technical solution for realizing the purpose of the invention is as follows: an industrial internet blockchain trusted computing architecture based on a computational prediction machine comprises an industrial internet device layer, a blockchain layer and a prediction machine layer, wherein:
the industrial Internet equipment layer comprises various heterogeneous industrial Internet equipment, and industrial computing tasks are generated in the process of managing the equipment;
the block chain layer comprises an intelligent contract and uses an intelligent contract engine as a carrier to execute industrial computing tasks;
and the prediction machine layer comprises a calculation prediction machine and an on-chain prediction machine and is used for improving the computing power of the intelligent contract.
A method for operating the computing prophetic-based industrial internet blockchain trusted computing architecture as described, comprising the steps of:
step 1, registering industrial internet intelligent equipment into a legal under-chain calculation predicting machine, issuing a unique account by an under-chain predicting machine system, and paying a set token by the intelligent equipment to serve as a mortgage;
step 6, establishing an excitation mechanism of the prediction machine under the chain, and selecting the calculation prediction machine to execute a calculation task through the excitation mechanism based on the matching algorithm after the network of the calculation prediction machine under the chain receives the request; establishing a consensus mechanism of the prediction machine under the chain, after the calculation task is completed, enabling each node to achieve consensus on the result of the calculation task, and feeding the consensus result back to the agent prediction machine system on the chain;
step 7, the on-chain predicting machine system receives the calculation result of the off-chain calculation predicting machine system by the agent predicting machine, translates the calculation result into a file which can be read by the intelligent contract and then sends the file to the intelligent contract;
step 8, the agent predicting machine sends the service quality of the network of the predicting machine under the chain to the monitoring predicting machine and records the service quality on the chain; meanwhile, the payment predicting machine converts the service fee frozen by the intelligent contract into a state available for extraction, and the service fee is extracted by the lower-chain predicting machine providing the service.
Compared with the prior art, the invention has the following remarkable advantages: (1) considering a calculation scene in the industrial internet, a set of block chain calculation system based on a calculation prediction machine is designed, and the large calculation requirement in the industrial scene is met; (2) in an excitation mechanism of the prediction machine network under the chain, a matching algorithm based on the node computing power is provided, the node computing power is considered in the matching process of the prediction machine under the chain and the computing task, and finally the computing time delay is reduced.
Drawings
FIG. 1 is a design flow diagram of an industrial Internet blockchain trusted computing architecture based on a computational prophetic machine according to the present invention.
Fig. 2 is an industrial internet architecture diagram.
FIG. 3 is a flowchart of the operation of the computer prediction machine mechanism
Fig. 4 is a flow chart of the matching algorithm.
FIG. 5 is a diagram illustrating delay variation curves calculated under different matching algorithms.
Detailed Description
The invention considers the calculation scene in the industrial internet and designs a set of block chain calculation system based on a calculation prediction machine to solve the problem of insufficient calculation capacity of the traditional block chain and meet the large calculation requirement in the industrial scene. Meanwhile, in an excitation mechanism of the prediction machine network under the chain, a matching algorithm based on the node computing power is provided, the node computing power is considered in the matching process of the prediction machine and the computing task under the chain, and finally the computing time delay is reduced.
The invention relates to an industrial Internet block chain trusted computing architecture based on a computational prediction machine, which comprises an industrial Internet equipment layer, a block chain layer and a prediction machine layer, wherein:
the industrial Internet equipment layer comprises various heterogeneous industrial Internet equipment, and industrial computing tasks are generated in the process of managing the equipment;
the block chain layer comprises an intelligent contract and uses an intelligent contract engine as a carrier to execute industrial computing tasks;
and the prediction machine layer comprises a calculation prediction machine and an on-chain prediction machine and is used for improving the computing power of the intelligent contract.
Further, the heterogeneous industrial internet devices are classified into the following three categories:
(1) the conventional device: only original industrial data can be generated, the capacity of transmitting and receiving data is not available, and the computing capacity and the storage capacity of the traditional equipment are lower than those of intelligent equipment;
(2) the intelligent device: the intelligent equipment can become a block chain node or a prediction machine node and has the capability of independently transmitting and receiving data;
(3) the data center comprises: as a centralized, down-link trusted node, a data center is a data hosting facility that is capable of storing multiple types of industrial data.
Further, the block chain layer is composed of uplink nodes, including intelligent contracts, on-chain prediction machines and other on-chain nodes; the block chain layer adopts a private chain, and in a block chain network decentralized by an industrial internet, each intelligent device is regarded as a legal block chain node with an independent account; meanwhile, the industrial application on the platform takes an intelligent contract as a carrier, and all computing logics are realized based on the intelligent contract; the intelligent contract is deployed in advance based on the multiparty agreement, and the calculation result is linked up to realize tamper resistance and traceability.
Further, the predicting machine layer consists of a predicting machine network under the chain and a predicting machine system on the chain;
the prediction machine network under the chain is composed of calculation prediction machines with sufficient calculation resources, intelligent equipment can become a calculation prediction machine node under the chain as long as the intelligent equipment is successfully registered in the prediction machine system on the chain, and the calculation prediction machines under the chain help the intelligent contracts on the chain to execute calculation tasks;
when the computing resources on the block chain cannot meet the requirements, the intelligent contract sends a request to a prediction machine system on the chain, and after receiving the request, the prediction machine system on the chain assists in operation through a computing prediction machine under the chain to jointly complete a computing task; in this process, the on-chain predictive engine system manages the interaction between the on-chain intelligent contracts and the off-chain computing predictive engines.
Further, the system of the online predicting machine comprises a registration predicting machine, a monitoring predicting machine, a payment predicting machine and an agent predicting machine, wherein:
(1) registering a prediction machine: the intelligent equipment under the chain needs to become a prediction machine, besides mortgage and locking a part of token as deposit, the deposit address and the payment address also need to be registered, and the registration process is completed by the registration prediction machine;
(2) monitoring a prediction machine: the monitoring contract of the on-chain predicting machine system is responsible for recording the service quality of the under-chain predicting machine network, including the response delay of the equipment, the response rate and the correct rate of the calculation result, the list of malicious nodes and unprocessed inquiry requests;
(3) payment predicting machine: the intelligent equipment under the chain needs to be registered when the intelligent equipment becomes a predictive speaker node, and a token needs to be mortgage in the registration process; the payment predicting machine serves as a third party in the transaction process of the token, and the intelligent contract on the chain is ensured to pay the payment to the under-chain predicting machine serving the intelligent contract on the chain; paying the service fee paid by the intelligent contract on the temporary conservation chain of the prediction machine until the prediction machine extracts income under the chain of providing the service;
(4) agent prediction machine: the agent prediction machine provides service for the intelligent contracts on the chain, when the intelligent contracts on the chain cannot meet the calculation requirements, the agent prediction machine sends a request to the agent prediction machine, and the agent prediction machine provides a uniform on-chain interface for the intelligent contracts; and after the calculation of the under-chain calculation predicting machine system is finished, the agent predicting machine uses the result in the callback function to asynchronously callback the calculation result to the intelligent contract.
The invention relates to an operation method of an industrial internet block chain trusted computing architecture based on a computational prediction machine, which comprises the following steps:
step 1, registering industrial internet intelligent equipment into a legal under-chain calculation predicting machine, issuing a unique account by an under-chain predicting machine system, and paying a set token by the intelligent equipment to serve as a mortgage;
step 6, establishing an excitation mechanism of the prediction machine under the chain, and selecting the calculation prediction machine to execute calculation tasks through the excitation mechanism based on the matching algorithm after the network of the calculation prediction machine under the chain receives the request; establishing a consensus mechanism of the prediction machine under the chain, after the calculation task is completed, enabling each node to achieve consensus on the result of the calculation task, and feeding the consensus result back to the agent prediction machine system on the chain;
step 7, the on-chain predicting machine system receives the calculation result of the off-chain calculation predicting machine system by the agent predicting machine, translates the calculation result into a file which can be read by the intelligent contract and then sends the file to the intelligent contract;
step 8, the agent predicting machine sends the service quality of the network of the predicting machine under the chain to the monitoring predicting machine and records the service quality on the chain; meanwhile, the payment predicting machine converts the service fee frozen by the intelligent contract into a state available for extraction, and the service fee is extracted by the lower-chain predicting machine providing the service.
Further, the under-link predictive engine consensus mechanism in step 6 refers to a mechanism for the under-link predictive engine to achieve consensus on the calculation result; after the calculation of the chained prediction machine group participating in the calculation task is completed, obtaining multiple results, and obtaining the result agreed by a plurality of nodes through a consensus mechanism, specifically:
when two thirds of the calculation prediction machines in the calculation group complete calculation, selecting the result agreed by most of the nodes as a final result; the final result is sent to the intelligent contract on the chain, and the sent content comprises the result after the consensus and a verification file, and the verification file ensures that the intelligent contract verifies the correctness of the result in only one step.
Further, the matching algorithm in step 6 specifically includes the following steps:
1) the agent prediction machine on the chain records the task related information into an event log of the agent prediction machine, wherein the task related information comprises the calculation task content requested by the intelligent contract, the number of prediction machines required by each calculation task and the reward provided by the intelligent contract; the event logs are subscribed by the computer prediction machine network under the chain, and the relevant information of the tasks is obtained;
2) each unmatched computing speaker sends a request to the smart contract to which the largest reward is provided, wherein the provided reward is inversely proportional to the expected computing latency of the computing speaker;
3) matching G calculation predicting machine nodes with the minimum expected calculation time delay with a calculation task, wherein the G calculation predicting machine nodes jointly form a calculation group of the calculation task;
4) when each task is matched with enough computational prediction machines, the matching process is ended;
5) and calculating the distributed tasks by the calculation prediction machines in the calculation group, and sending a calculation result approved by the members with the set percentage in the group as a final calculation result to the intelligent contract.
The invention is further described in detail below with reference to the drawings and specific embodiments.
Examples
With reference to fig. 1, the design process of the block chain trusted computing architecture of the industrial internet based on the computational prophetic machine of the present invention specifically includes the following steps:
the first stage is as follows: firstly, a block chain computing system based on a computational prediction machine is designed, and a credible under-chain cooperative computing is realized by constructing an under-chain prediction machine network and an on-chain prediction machine system, so that the computing requirement of industrial application in an industrial internet block chain is supported.
The computing architecture includes three layers: the system comprises an industrial internet equipment layer, a block chain layer and a prediction machine layer. The industrial internet device layer includes various heterogeneous industrial internet devices, and a large amount of industrial computing tasks are generated in the process of managing the devices. The computing tasks are executed by using an intelligent contract engine in a blockchain layer as a carrier, and the prediction machine layer is responsible for helping the intelligent contract to improve the computing capability, and the specific architecture is shown in fig. 2.
And a second stage: the industrial internet equipment layer is designed, and consists of various heterogeneous industrial internet equipment, which has very limited computing power and can be roughly divided into three categories:
(1) a conventional device. The conventional device can only generate raw industrial data, has no capability of transmitting and receiving data, and has extremely limited computing power and storage capacity.
(2) And (4) intelligent equipment. The intelligent device can be a blockchain node or a prediction machine node, has the capability of independently transmitting and receiving data, and is abundant in storage and computing resources.
(3) A data center. As a centralized, down-link trusted node, a data center is a data hosting facility that is capable of storing various types of industrial data.
And a third stage: and designing a block chain layer, wherein the block chain layer is formed by nodes of an upper chain and comprises intelligent contracts, prediction machines on the chain and other nodes on the chain. The block chain system designed by the invention adopts a private chain, so that each intelligent device can be regarded as a legal block chain node with an independent account in a decentralized block chain network of the industrial internet. Meanwhile, the industrial application on the platform takes the intelligent contract as a carrier, and all computing logics are realized based on the intelligent contract. The intelligent contract is deployed in advance based on the multiparty agreement, and links the calculation result, so that the anti-tampering and traceability are realized, and the calculation credibility is ensured. However, the limited computing power of the intelligent contract cannot meet the computing requirement of industrial application on the block chain, so the invention provides a distributed computing preplan mechanism, which comprises a preplan network under the chain and a preplan system on the chain, to improve the computing power of the intelligent contract.
A fourth stage: and designing a prophetic machine layer, wherein the prophetic machine layer consists of a prophetic machine network under the chain and a prophetic machine system on the chain. The network of speakers under the chain is made up of a large number of computational speakers with abundant computational resources. It should be noted that the intelligent device can also become a node of the offline computation prediction machine as long as the intelligent device successfully registers in the online prediction machine system. The computation prediction machines under the chains can help the intelligent contracts on the chains to execute computation tasks and make up for the shortage of computation capability on the chains. For example, an intelligent contract for resource scheduling in the industrial internet cannot meet the demand of the computing resources on the blockchain due to high complexity and large computation of the scheduling task. At the moment, the intelligent contract sends a request to the on-chain predicting machine system, and after the on-chain predicting machine system receives the request, the on-chain predicting machine system assists in operation through the on-chain computing predicting machine to jointly complete the computing task. In the process, the on-chain predictive machine system manages the interaction between the on-chain intelligent contract and the off-chain computing predictive machine. In the invention, four types of prediction machines exist in the on-chain prediction machine system: register a predictive machine, monitor a predictive machine, pay a predictive machine, and agent a predictive machine.
(1) Registering a predictive speaker: to become a predictive machine, the intelligent devices under the chain need to register the address of the deposit and the address of the payment in addition to mortgage and lock a portion of the token as the deposit. This registration process is accomplished by registering the predictive engine.
(2) Monitoring a prediction machine: the monitoring contract of the on-chain predicting machine system is responsible for recording the service quality of the under-chain predicting machine system network, such as the response delay of the equipment, the response rate and the correct rate of the calculation result, the list of malicious nodes, unprocessed query requests and the like.
(3) Payment prophetic machine: as mentioned above, the intelligent devices under the chain need to be registered to become the nodes of the prediction machine, and the token needs to be collated in the registration process. The payment prophetic machine serves as a third party in the transaction process of the token, and the intelligent contract on the chain is guaranteed to pay the payment to the under-chain prophetic machine serving the intelligent contract on the chain. And paying the service fee paid by the intelligent contract on the chain by the prediction machine temporarily until the prediction machine under the chain providing the service extracts the income of the prediction machine.
(4) Agent prediction machine: the agent prediction machine mainly provides services for the intelligent contracts on the chain, when the intelligent contracts on the chain cannot meet the computing requirements, the agent prediction machine sends requests to the agent prediction machine, and the agent prediction machine provides a uniform on-chain interface for the intelligent contracts. And after the calculation of the system of the lower-chain calculation predicting machine is finished, the agent predicting machine uses the result in the callback function to asynchronously call back the calculation result to the intelligent contract.
The specific implementation mode of the block chain system of the industrial internet based on the computer prediction machine is as follows:
the method comprises the following steps: a workflow of the designed block chain calculation system of the industrial internet is provided, as shown in fig. 3:
1) first, the intelligent device of the industrial internet registers to become a legal down-link computer for predicting words, the system issues a unique account to the intelligent device, and the intelligent device pays a certain token as a mortgage.
2) The intelligent contract obtains data from the chain and triggers a contract calculation task.
3) When the computing power on the chain can not meet the requirement, the intelligent contract sends a request to the system of the on-chain predicting machine, and the specific implementation method is to call the 'message' of the agent predicting machine. The request includes a computing task, while the intelligent contract freezes a service charge in the account. The agent prediction machine records the calculation task in the event log of the agent prediction machine.
4) The calculation prediction machine network under the chain subscribes the event log of the agent prediction machine on the chain. Therefore, after the agent prediction machine records the calculation task, the network of the down-link calculation prediction machine receives the calculation task request immediately.
5) And the data center feeds back the agent prediction machine according to the task request.
6) After receiving the request, the network of the computational prediction machines under the chain selects the computational prediction machines to execute the computational tasks through the matching algorithm provided by the invention. After the computation task is completed, all nodes achieve consensus on the result of the computation task and feed the consensus back to the agent prediction machine system on the chain.
7) And the on-chain predicting machine system receives the calculation result of the off-chain calculation predicting machine system by the agent predicting machine, translates the calculation result into a file which can be read by the intelligent contract and then sends the file to the intelligent contract.
8) The agent predicting machine sends the service quality of the network of the predicting machine under the chain to the monitoring predicting machine and records the service quality on the chain. Meanwhile, the payment predicting machine converts the service fee frozen by the intelligent contract into a state available for extraction, and the service providing chain predicting machine can extract the service fee.
The basic operation flow of the prediction machine mechanism of the system is designed in detail according to specific service requirements on the basis of the basic flow.
Step two: consensus mechanism for under-design-chain prediction machines
In the network of the under-link computational prophetic machine designed by the invention, the consensus mechanism refers to a mechanism for the under-link computational prophetic machine to achieve consensus on a computation result. In the above work flow, after the calculation is completed by the prediction machine group under the chain participating in the calculation task, multiple results are obtained, and the result agreed by the plurality of nodes is obtained through a consensus mechanism. Specifically, after two-thirds of the computing prediction machines in the computing group complete computing, the agreed result of most of the nodes is selected as the final result. The final result is sent to the chain intelligent contract, and the sent content comprises the result after consensus and a verification file, and the verification file can help the intelligent contract to verify the correctness of the result only in one step. Specifically, the verification method of the smart contract may be zero-knowledge verification or the like.
Step three: design-chain prolog machine excitation mechanism
In the computer-to-speak machine system designed by the invention, an incentive mechanism incentivizes nodes (intelligent devices in the industrial internet) under a chain to provide credible services for getting economic returns, and further starts a computer-to-speak machine network under the chain. In the whole system of the incentive mechanism, the benefits of the intelligent contract are the time delay of the calculation task, the correctness of the calculation result and the minimization of the overhead, and the benefits of participating in the prediction machine of the calculation task are the maximized benefits. The incentive mechanism needs to comprehensively consider both parties, so that benefits between nodes are not conflicted, each node is ensured to obtain positive benefits, and the condition that the service of the node is stopped due to the occurrence of negative benefits is prevented. Meanwhile, a reasonable incentive mechanism can also guarantee fairness among calculation nodes under a chain, namely 'too much labor and too much purpose', rewards are distributed according to contributions, and market control is resisted.
Specifically, the designed incentive mechanism provides better and low-delay computing resources for incentive participation nodes, service of nodes under a chain is scored, and computing prediction machine working groups with abnormal positive and negative score ratios or excessively high negative scores can be eliminated. In terms of reward distribution, the incentive mechanism of a traditional speaker network distributes the same reward to each speaker node participating in the service. In practice, however, the contribution of each serving node is not the same, and the benefit of the partial prediction machine may even be negative under the mechanism of evenly distributing the rewards. Therefore, the traditional prediction machine incentive mechanism cannot guarantee the reasonability of individual allocation, and improvement on fairness is needed. The invention provides a matching algorithm based on node computing power. While achieving the rationality of individual reward distribution and the suitability of incentives. In connection with fig. 4, the proposed matching algorithm is performed in the following steps:
1) the agent prediction machine on the chain records the task related information into the event log of the agent prediction machine, wherein the task related information comprises the calculation task content requested by the intelligent contract, the number of prediction machines required by each calculation task and the reward provided by the intelligent contract. The event log is subscribed by the network of the computer prediction machine under the chain to obtain the information;
2) each unmatched computing speaker sends a request to the smart contract to which the largest reward is provided, wherein the provided reward is inversely proportional to the expected computing latency of the computing speaker;
3) the system matches G calculation prediction machine nodes with the minimum expected calculation time delay with the calculation task, and the calculation prediction machine nodes jointly form a calculation group of the calculation task;
4) when each task is matched with enough computational prediction machines, the matching process is ended;
5) and the calculation prediction machines in the calculation group calculate the distributed tasks, obtain calculation results approved by most members in the group and send the calculation results as final calculation results to the intelligent contract.
In the designed matching mechanism, the intelligent contract executes the calculation task in a mode with less delay, so the performance is superior to the random selection in the traditional consensus mechanism.
The performance of the invention is verified by simulation, and the simulation result is shown in fig. 5. In the simulation, under three different scenes corresponding to 4, 5 and 6 calculation tasks, the change situation of the total delay time along with the increase of the calculation prediction machine under different selection algorithms is compared. After using the matching mechanism, each computing task matches six computing prophetics as its computing group. It can be seen from the figure that, first, under the matching algorithm proposed herein, the total latency is smaller than under the random selection mechanism. Secondly, under the matching algorithm proposed herein, the total latency decreases as the number of prediction machines increases. This is because as the number of the computing prophetic machines increases, more computing prophetic machines with stronger computing power appear in the set, and the computing prophetic machines matched with the computing task have stronger overall computing power, which finally reduces the computing time.
In conclusion, simulation results show that the matching algorithm provided by the invention for the industrial internet computation prediction machine system has lower computation time delay compared with the traditional random matching algorithm.
Claims (8)
1. An industrial internet blockchain trusted computing architecture based on a computational prediction machine is characterized by comprising an industrial internet device layer, a blockchain layer and a prediction machine layer, wherein:
the industrial Internet equipment layer comprises various heterogeneous industrial Internet equipment, and industrial computing tasks are generated in the process of managing the equipment;
the block chain layer comprises an intelligent contract and uses an intelligent contract engine as a carrier to execute industrial computing tasks;
and the prediction machine layer comprises a calculation prediction machine and an on-chain prediction machine and is used for improving the computing power of the intelligent contract.
2. The computing predictive machine-based industrial internet blockchain trusted computing architecture of claim 1, wherein the heterogeneous industrial internet devices are classified into three categories:
(1) the conventional device: only original industrial data can be generated, the capacity of transmitting and receiving data is not available, and the computing capacity and the storage capacity of the traditional equipment are lower than those of intelligent equipment;
(2) the intelligent equipment: the intelligent equipment can become a block chain node or a prediction machine node and has the capability of independently transmitting and receiving data;
(3) the data center comprises: as a centralized, down-link trusted node, a data center is a data hosting facility that is capable of storing multiple types of industrial data.
3. The computing-prophetic-machine-based industrial internet blockchain trusted computing architecture of claim 1, wherein the blockchain layer is comprised of uplink nodes including intelligent contracts, on-chain prophetic machines, and other on-chain nodes; the block chain layer adopts a private chain, and in a block chain network with decentralized industrial internet, each intelligent device is regarded as a legal block chain node with an independent account; meanwhile, the industrial application on the platform takes an intelligent contract as a carrier, and all computing logics are realized based on the intelligent contract; the intelligent contract is deployed in advance based on the multiparty agreement, and the calculation result is linked up to realize tamper resistance and traceability.
4. The computing prophetic-based industrial internet blockchain trusted computing architecture of claim 1, wherein the prophetic layer is comprised of a network of prophetic machines under a chain and a system of prophetic machines on a chain;
the prediction machine network under the chain is composed of calculation prediction machines with sufficient calculation resources, intelligent equipment can become a calculation prediction machine node under the chain as long as the intelligent equipment is successfully registered in the prediction machine system on the chain, and the calculation prediction machines under the chain help the intelligent contracts on the chain to execute calculation tasks;
when the computing resources on the block chain cannot meet the requirements, the intelligent contract sends a request to a prediction machine system on the chain, and after receiving the request, the prediction machine system on the chain assists in operation through a computing prediction machine under the chain to jointly complete a computing task; in this process, the on-chain predictive engine system manages the interaction between the on-chain intelligent contracts and the off-chain computing predictive engines.
5. The computing predictive-based industrial internet blockchain trusted computing architecture of claim 4, wherein the on-chain predictive machine system includes a registration predictive machine, a monitoring predictive machine, a payment predictive machine, and an agent predictive machine, wherein:
(1) registering a predictive speaker: the intelligent equipment under the chain needs to become a prediction machine, besides mortgage and locking a part of token as deposit, the deposit address and the payment address also need to be registered, and the registration process is completed by the registration prediction machine;
(2) monitoring a prediction machine: the monitoring contract of the on-chain predicting machine system is responsible for recording the service quality of the under-chain predicting machine network, including the response delay of the equipment, the response rate and the correct rate of the calculation result, the list of malicious nodes and unprocessed inquiry requests;
(3) payment predicting machine: the intelligent equipment under the chain needs to be registered when the intelligent equipment becomes a predictive speaker node, and a token needs to be mortgage in the registration process; the payment predicting machine serves as a third party in the transaction process of the token, and the intelligent contract on the chain is ensured to pay the payment to the under-chain predicting machine serving the intelligent contract on the chain; paying the service fee paid by the intelligent contract on the temporary conservation chain of the prediction machine until the prediction machine extracts income under the chain of providing the service;
(4) agent prediction machine: the agent prediction machine provides service for the intelligent contracts on the chain, when the intelligent contracts on the chain cannot meet the calculation requirements, the agent prediction machine sends a request to the agent prediction machine, and the agent prediction machine provides a uniform on-chain interface for the intelligent contracts; and after the calculation of the system of the lower-chain calculation predicting machine is finished, the agent predicting machine uses the result in the callback function to asynchronously call back the calculation result to the intelligent contract.
6. A method for operating a block chain trusted computing architecture for an industrial Internet based on a computer prediction machine according to any one of claims 1 to 5, comprising the following steps:
step 1, registering industrial internet intelligent equipment into a legal under-chain calculation predicting machine, issuing a unique account by an under-chain predicting machine system, and paying a set token by the intelligent equipment to serve as a mortgage;
step 2, the intelligent contract obtains data from the chain and triggers a contract calculation task;
step 3, when the calculation capacity on the chain can not meet the requirement, the intelligent contract sends a request to a system of the on-chain prediction machine and calls the 'message' of the agent prediction machine; the request comprises a computing task, and meanwhile, intelligent contract is conducted to freeze service charge in an account; the agent prediction machine records the calculation task in the event log of the agent prediction machine;
step 4, the network of the under-link computing prophetic machine subscribes the event log of the on-link agent prophetic machine, and the network of the under-link computing prophetic machine immediately receives a computing task request after the agent prophetic machine records a computing task;
step 5, the data center feeds back the agent prediction machine according to the task request;
step 6, establishing an excitation mechanism of the prediction machine under the chain, and selecting the calculation prediction machine to execute a calculation task through the excitation mechanism based on the matching algorithm after the network of the calculation prediction machine under the chain receives the request; establishing a consensus mechanism of the prediction machine under the chain, after the calculation task is completed, enabling each node to achieve consensus on the result of the calculation task, and feeding the consensus result back to the agent prediction machine system on the chain;
step 7, the on-chain predicting machine system receives the calculation result of the off-chain calculation predicting machine system by the agent predicting machine, translates the calculation result into a file which can be read by the intelligent contract and then sends the file to the intelligent contract;
step 8, the agent predicting machine sends the service quality of the network of the predicting machine under the chain to the monitoring predicting machine and records the service quality on the chain; meanwhile, the payment predicting machine converts the service fee frozen by the intelligent contract into a state available for extraction, and the service fee is extracted by the lower-chain predicting machine providing the service.
7. The method of claim 6, wherein the mechanism for consensus of the prediction machines in step 6 is a mechanism for consensus of the calculation results of the prediction machines in step 6; after the calculation of the chained prediction machine group participating in the calculation task is completed, obtaining multiple results, and obtaining the result agreed by a plurality of nodes through a consensus mechanism, specifically:
when two thirds of the calculation prediction machines in the calculation group complete calculation, selecting the result agreed by most of the nodes as a final result; the final result is sent to the intelligent contract on the chain, and the sent content comprises the result after the consensus and a verification file, and the verification file ensures that the intelligent contract verifies the correctness of the result in only one step.
8. The method according to claim 6, wherein the matching algorithm in step 6 is as follows:
1) the agent prediction machine on the chain records the task related information into an event log of the agent prediction machine, wherein the task related information comprises the calculation task content requested by the intelligent contract, the number of prediction machines required by each calculation task and the reward provided by the intelligent contract; the event logs are subscribed by the network of the computer prediction machine under the chain to obtain the relevant information of the tasks;
2) each unmatched computing speaker sends a request to the smart contract to which the largest reward is provided, wherein the provided reward is inversely proportional to the expected computing latency of the computing speaker;
3) matching G calculation predicting machine nodes with the minimum expected calculation time delay with a calculation task, wherein the G calculation predicting machine nodes jointly form a calculation group of the calculation task;
4) when each task is matched with enough computational prediction machines, the matching process is ended;
5) and calculating the distributed tasks by the calculation prediction machines in the calculation group, and sending a calculation result approved by the members with the set percentage in the group as a final calculation result to the intelligent contract.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111471798.4A CN114428978A (en) | 2021-12-06 | 2021-12-06 | Industrial Internet block chain trusted computing architecture and method based on computational prediction machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111471798.4A CN114428978A (en) | 2021-12-06 | 2021-12-06 | Industrial Internet block chain trusted computing architecture and method based on computational prediction machine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114428978A true CN114428978A (en) | 2022-05-03 |
Family
ID=81311272
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111471798.4A Pending CN114428978A (en) | 2021-12-06 | 2021-12-06 | Industrial Internet block chain trusted computing architecture and method based on computational prediction machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114428978A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114915556A (en) * | 2022-07-19 | 2022-08-16 | 北京航空航天大学 | Calculation task allocation method for industrial internet profit optimization |
| CN116684430A (en) * | 2023-06-13 | 2023-09-01 | 山东大学 | Distributed bidirectional predictor and blockchain system |
| CN116866045A (en) * | 2023-07-18 | 2023-10-10 | 四川华西集采电子商务有限公司 | Block chain credible predictor decision support system based on access control rule |
-
2021
- 2021-12-06 CN CN202111471798.4A patent/CN114428978A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114915556A (en) * | 2022-07-19 | 2022-08-16 | 北京航空航天大学 | Calculation task allocation method for industrial internet profit optimization |
| CN114915556B (en) * | 2022-07-19 | 2022-10-21 | 北京航空航天大学 | Calculation task allocation method for industrial internet profit optimization |
| CN116684430A (en) * | 2023-06-13 | 2023-09-01 | 山东大学 | Distributed bidirectional predictor and blockchain system |
| CN116684430B (en) * | 2023-06-13 | 2024-03-08 | 山东大学 | Distributed bidirectional predictor and blockchain system |
| CN116866045A (en) * | 2023-07-18 | 2023-10-10 | 四川华西集采电子商务有限公司 | Block chain credible predictor decision support system based on access control rule |
| CN116866045B (en) * | 2023-07-18 | 2024-01-23 | 四川华西集采电子商务有限公司 | Block chain credible predictor decision support system based on access control rule |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114428978A (en) | Industrial Internet block chain trusted computing architecture and method based on computational prediction machine | |
| Yuan et al. | CSEdge: Enabling collaborative edge storage for multi-access edge computing based on blockchain | |
| CN109829718B (en) | Block chain multi-level framework based on storage application scene and operation method thereof | |
| Hammoud et al. | On demand fog federations for horizontal federated learning in IoV | |
| CN111010434B (en) | Optimized task unloading method based on network delay and resource management | |
| CN111405527B (en) | Vehicle-mounted edge computing method, device and system based on volunteer cooperative processing | |
| CN112134959A (en) | Heterogeneous edge resource sharing method based on block chain | |
| CN113364831B (en) | Multi-domain heterogeneous computing network resource credible cooperation method based on block chain | |
| CN112488706B (en) | Cloud service management method and system based on block chain | |
| CN113315978B (en) | Collaborative online video edge caching method based on federal learning | |
| CN110400140A (en) | Layering fragment exchange information processing method and application based on Fabric | |
| CN113282348B (en) | Edge calculation task unloading system and method based on block chain | |
| CN110489218A (en) | Vehicle-mounted mist computing system task discharging method based on semi-Markovian decision process | |
| CN114938292B (en) | Multi-level optimization PBFT consensus method based on node credibility | |
| CN107194810A (en) | Asset allocation system and method | |
| Wu et al. | A profit-aware coalition game for cooperative content caching at the network edge | |
| CN112804361A (en) | Edge alliance game method for content cooperation cache | |
| CN112308297A (en) | Reusable logistics resource sharing system based on block chain and service integration method | |
| CN116366669A (en) | Consensus method based on reputation value weight balance suitable for crowdsourcing system | |
| CN113283778B (en) | Layered convergence federal learning method based on security evaluation | |
| CN111866181B (en) | Block chain-based task unloading optimization method in fog network | |
| CN113361881A (en) | Overlapped organization cooperative control method based on vehicle fog computing architecture | |
| CN109255614A (en) | Digging mine method and device, digging mine machine and block catenary system based on block chain | |
| CN110971707B (en) | Distributed service caching method in mobile edge network | |
| CN115471301B (en) | Unmanned vehicle decentralized order dispatching method and device, storage medium and terminal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |