Method and system for dynamically sharing and configuring computing power of content generated by artificial intelligence

Abstract

The invention provides a computing power dynamic sharing configuration method and system for content generated by artificial intelligence, and the method comprises the steps: S1, enabling a plurality of edge nodes to be connected through a block chain technology in a distributed edge network, and forming a decentralized computing power resource sharing network; s2, acquiring an AIGC task issued by a user, and evaluating the computing power demand of the AIGC task; and S3, according to the real-time computing power resource state of the edge node in the decentralized computing power resource sharing network and the computing power demand of the AIGC task, evaluating the number of nodes required by the current AIGC task, and allocating the computing task based on the number of nodes required by the current AIGC task.

Description

Technical Field

[0001] The present invention relates to the technical field of blockchain, distributed computing, edge computing and artificial intelligence generated content, and specifically, to a method and system for dynamically sharing and configuring computing power for artificial intelligence generated content, and more specifically, to a method and system for dynamically sharing and configuring computing power for artificial intelligence generated content in a distributed edge network based on blockchain technology. Background Art

[0002] With the development of artificial intelligence (AI) technology and distributed computing, artificial intelligence generated content (AIGC) has been widely used in many fields, including content creation, media production, automated generation, intelligent interaction, etc. These AIGC tasks rely on complex AI models and algorithms, such as natural language generation (NLG), image generation, deep learning model training, etc., which require a large amount of computing resources to support model training, reasoning and generation processes. Usually, the execution of these computing tasks depends on centralized data centers and cloud computing platforms. However, with the rapid popularization of edge computing and Internet of Things (IoT) devices, the demand for data generation and computing has gradually migrated to the edge of the network. This centralized computing model has exposed many problems in dealing with distributed computing needs, including: when a large number of AI generation tasks are initiated at the same time, the centralized server is prone to computing resource shortage, resulting in performance degradation and increased task delays. The computing power of edge nodes is relatively limited and cannot provide strong computing support like centralized servers. Especially when performing large-scale AI model training tasks, a single edge node often finds it difficult to complete the task independently. The computing task loads of different edge nodes vary. Some nodes may be overloaded while other nodes are idle, resulting in low utilization of computing resources. Most current distributed computing frameworks use predefined fixed allocation strategies and cannot dynamically adjust computing resources according to actual task requirements. Due to the complexity and diversity of AIGC tasks, the computing power requirements of different tasks are highly dynamic and cannot be efficiently handled by static allocation mechanisms.

[0003] Therefore, there is an urgent need for a method and device for dynamic sharing and scheduling of computing power for AI-generated content in a distributed edge computing network based on blockchain. Summary of the invention

[0004] In view of the defects in the prior art, the purpose of the present invention is to provide a method for dynamically sharing and configuring computing power for artificial intelligence-generated content.

[0005] A method for dynamically sharing and configuring computing power for artificial intelligence-generated content provided by the present invention includes:

[0006] Step S1: In a distributed edge network, multiple edge nodes are connected through blockchain technology to form a decentralized computing resource sharing network;

[0007] Step S2: Obtain AIGC tasks posted by users and evaluate the computing power requirements of AIGC tasks;

[0008] Step S3: According to the real-time computing resource status of the edge nodes in the decentralized computing resource sharing network and the computing power requirements of the AIGC task, the number of nodes required for the current AIGC task is evaluated, and computing tasks are allocated based on the number of nodes required for the current AIGC task.

[0009] Preferably, step S1 includes: the distributed edge network includes multiple edge nodes, and the multiple edge nodes include IoT devices, edge servers and mobile devices; multiple edge nodes that have been registered on the blockchain are connected through blockchain technology, and resources of the multiple edge nodes are dynamically monitored through smart contracts.

[0010] Preferably, step S2 comprises:

[0011] Step S2.1: Obtain the AIGC task posted by the user; wherein the AIGC task includes: task type, task data volume, AI model parameters for generating the AIGC task, and execution time requirements;

[0012] Step S2.2: Calculate the complexity coefficient of the AIGC task posted by the current user, and evaluate the computing power requirement based on the complexity coefficient of the task.

[0013] Preferably, the step S2.2 comprises:

[0014] Step S2.2.1: Calculate the complexity coefficient of the AIGC task issued by the current user;

[0015] C=C type *C Data *C Model *C Time

[0016] Where C represents the complexity coefficient; C type Indicates the task type; C Data Data volume; C Model Represents the AI ​​model parameters for generating AIGC tasks; C Time Indicates execution time;

[0017] Step S2.2.2: Build a computing power demand prediction model based on the machine learning algorithm, and use the historical task execution data to train the computing power demand prediction model to obtain the trained computing power demand prediction model; use the trained computing power demand prediction model to predict the computing power demand based on the complexity coefficient of the AIGC task released by the current user.

[0018] Preferably, the step S3 comprises:

[0019] Step S3.1: The smart contract obtains the status information of each edge node in real time through the blockchain, and the status information of the edge node includes the current load of the edge node, the available computing power and the network status information;

[0020] Step S3.2: Filter edge nodes that meet the preset requirements based on the computing power requirements of the current AIGC task and the acquired status information of each edge node;

[0021] Step S3.3: Select edge nodes based on the computing power scheduling mechanism according to the computing power requirements of the current AIGC task and the edge nodes that meet the preset requirements;

[0022] The computing power scheduling mechanism includes: determining the number of selected edge nodes based on the computing power requirements of the current AIGC task and the edge nodes that meet the preset requirements and maximizing the number of edge nodes;

[0023] When the computing power of the edge node is greater than the preset value and is in an idle state, the current edge node is preferentially scheduled to participate in task calculation;

[0024] When the load of the edge node is greater than the preset value, the tasks assigned to the current node are automatically reduced to achieve balanced computing load of the entire network;

[0025] When a task meets the preset requirements, the nodes are scheduled to participate in the execution first to ensure that the task is completed in time.

[0026] Preferably, smart contracts are used to monitor each AIGC task in real time, and the task execution progress and computing power usage of each node are recorded through the distributed ledger in the blockchain network; logs of all task execution operations are recorded on the blockchain.

[0027] Preferably, the method further includes: when the smart contract detects that there is an abnormality in the status information of a certain edge node, dynamically scheduling other available nodes to take over the task of the current edge node and continue to execute.

[0028] Preferably, the method further includes: analyzing the performance of each edge node based on the historical task execution data recorded in the blockchain, and dynamically allocating computing tasks according to the performance of each edge node obtained from the analysis.

[0029] Preferably, the method further includes: realizing computing resource sharing and task scheduling among multiple blockchain networks based on smart contracts through cross-chain computing.

[0030] Preferably, when the task is completed, the smart contract automatically settles the remuneration based on the computing power contributed by each edge node and the efficiency of task completion.

[0031] A system for dynamically sharing and configuring computing power for artificial intelligence-generated content provided by the present invention includes:

[0032] Module M1: In a distributed edge network, multiple edge nodes are connected through blockchain technology to form a decentralized computing resource sharing network;

[0033] Module M2: Obtain AIGC tasks posted by users and evaluate the computing power requirements of AIGC tasks;

[0034] Module M3: According to the real-time computing resource status of the edge nodes in the decentralized computing resource sharing network and the computing power requirements of the AIGC task, the number of nodes required for the current AIGC task is evaluated, and computing tasks are allocated based on the number of nodes required for the current AIGC task.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] 1. The present invention uses blockchain smart contracts to achieve on-demand scheduling and sharing of computing power between nodes in a distributed edge network, which is efficiently utilized and ensures efficient execution of AIGC tasks;

[0037] 2. The present invention provides a dynamic computing power allocation strategy to flexibly adjust computing power resources in a distributed network according to the complexity of tasks and real-time requirements, thereby avoiding resource waste and node overload;

[0038] 3. Record and monitor the task execution process through blockchain to ensure the transparency and fairness of task allocation and prevent unfair operations in the use of computing power;

[0039] 4. Provide a decentralized computing power scheduling mechanism to enhance the scalability and elasticity of the system and meet the growing AIGC task requirements;

[0040] 5. Through real-time computing power status monitoring of smart contracts and edge nodes, the allocation of computing power resources can be dynamically adjusted according to task requirements and node load conditions to avoid waste of computing power resources and ensure the efficiency of task execution.

[0041] 6. The tamper-proof nature of blockchain technology ensures transparent records of all operations during task execution. Users can monitor the progress of task execution in real time, avoid unfair operations in resource use, and improve the security of the system.

[0042] 7. The system can dynamically adjust the allocation of computing resources according to the priority and urgency of the task to ensure that high-priority tasks are processed in a timely manner. At the same time, by real-time monitoring of the load status of each node, the system can balance the use of computing resources of each node to avoid single-point overload.

[0043] 8. Automatically manage task settlement through smart contracts to ensure fair distribution of computing resources, provide timely computing power rewards to participating nodes, and encourage nodes to continuously contribute computing power resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0045] Figure 1 A system architecture diagram for dynamically sharing computing power configured for AI-generated content. DETAILED DESCRIPTION

[0046] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0047] Example 1

[0048] According to the method and system for dynamic sharing configuration of computing power for artificial intelligence-generated content provided by the present invention, it is intended to solve the problems of uneven scheduling of computing power resources, low efficiency, and insufficient transparency of task execution in existing edge computing networks; by introducing blockchain technology and smart contract mechanism, the present invention can dynamically share computing power resources between nodes, schedule on demand, and realize efficient execution of artificial intelligence-generated content tasks. Through a decentralized computing power sharing system, the present invention avoids the bottleneck and inefficiency of the centralized computing power scheduling architecture, significantly improves the utilization efficiency of computing power resources and task processing capabilities in edge networks, and ensures the transparency and fairness of the task execution process.

[0049] The method for dynamically sharing and configuring computing power for artificial intelligence-generated content includes:

[0050] Step S1: In the distributed edge network, each node is interconnected through blockchain technology to form a decentralized computing power sharing network; each node is a device that can provide computing power;

[0051] In this embodiment, the scheduling of tasks and the allocation of computing power are managed through blockchain smart contracts. Each participating node registers its own computing power resources through smart contracts and participates in computing power sharing according to the rules of the blockchain network. Smart contracts predefine the mechanisms for computing power resource allocation, task execution monitoring, and remuneration settlement to ensure the automation and transparency of task execution.

[0052] Step S2: Obtain AIGC tasks posted by users and evaluate the computing power requirements of AIGC tasks;

[0053] In this embodiment, when a user publishes an AIGC task, the task information is transmitted to all edge nodes through the blockchain network. The task publisher will describe the computing power requirements of the task in detail, including the complexity of the task, the expected execution time, and the resource requirements. The system evaluates the computing power requirements of the task based on factors such as the scale of the task, the execution time limit, and the amount of calculation, and automatically selects the appropriate node to participate in the calculation through the smart contract.

[0054] Step S3: According to the real-time computing resource status of the edge nodes in the decentralized computing resource sharing network and the computing power requirements of the AIGC task, the number of nodes required for the current AIGC task is evaluated, and computing tasks are allocated on demand to achieve dynamic allocation.

[0055] In this embodiment, the smart contract dynamically allocates computing tasks based on the real-time computing power status of the edge node and the computing power requirements of the task. When a task is initiated, the system will query the current computing power load of each node in real time through the distributed ledger in the blockchain, and automatically select the appropriate node to perform the task according to the task requirements. Through the automated computing power scheduling mechanism of the smart contract, the system can dynamically adjust the computing power allocation according to the real-time status of the node during the task execution process.

[0056] On-demand allocation: When a node has sufficient computing power and is idle, it will be scheduled first to participate in task calculation;

[0057] Load balancing: If a node is overloaded, the system will automatically reduce the tasks assigned to that node and reselect other nodes with lighter loads to achieve load balancing of the computing power of the entire network.

[0058] Resource tilt for high-priority tasks: For tasks of high importance or strict time requirements, the system will prioritize scheduling more computing nodes to participate in the execution to ensure that the tasks are completed in time.

[0059] The method for dynamically sharing and configuring computing power for artificial intelligence-generated content also includes:

[0060] Task execution and real-time monitoring: After the task is assigned to each edge node, each node will participate in the execution of the task according to its own computing power. The system records the task execution progress and computing power usage of each node through the blockchain. The smart contract will monitor the task completion of each node to ensure that the task is completed within the specified time, and ensure the transparency of the calculation process and the immutability of the data.

[0061] Computing power settlement and reward mechanism: After the task is completed, the system will automatically settle the computing power reward based on the computing power provided by each node and the task completion time. The smart contract distributes the reward to each participating node in proportion to the computing power contribution and execution effect of the task. This settlement process is carried out through the blockchain network to ensure the fairness, transparency and immutability of the settlement.

[0062] Fault tolerance and dynamic adjustment mechanism for task execution

[0063] During the task execution process, if an edge node cannot execute the task normally due to problems such as excessive load, failure or network delay, the system will detect the anomaly through the smart contract and make dynamic adjustments to reschedule other available nodes to participate in the task execution. This mechanism can effectively improve the reliability of the system and the stability of task execution, and avoid task interruption due to the failure of a single node.

[0064] Example 2

[0065] Embodiment 2 is a preferred embodiment of Embodiment 1

[0066] A method for dynamically sharing and configuring computing power for artificial intelligence-generated content provided by the present invention includes:

[0067] Step 1: Blockchain network and node deployment;

[0068] In this embodiment, in a distributed edge network, multiple edge nodes are connected through blockchain technology to form a decentralized computing resource sharing network. These nodes can be IoT devices, edge servers, mobile devices, etc. They have certain computing capabilities and can be interconnected through the blockchain network. Each node participates in the distributed computing power sharing network by registering on the blockchain, and dynamically manages resources through smart contracts.

[0069] The smart contract of the blockchain network includes task allocation, resource monitoring and settlement modules. Smart contracts are deployed by central nodes or users and manage computing resources according to the rules in the contract. The contract clearly defines the computing power contribution method of participating nodes, task allocation rules and computing power settlement mechanism.

[0070] Step 2: AIGC task release and demand assessment;

[0071] The publisher of AIGC tasks submits tasks through the blockchain network; in this embodiment, the publisher of AIGC tasks is the AI ​​content generation system;

[0072] The task description includes: task type, task data volume, AI model parameters for generating AIGC tasks, and execution time requirements;

[0073] By analyzing the complexity of the task, the required computing resources are calculated, and the number of nodes required is evaluated based on the scale of the task; after the evaluation, the smart contract will allocate computing tasks on demand based on the currently available nodes in the edge network.

[0074] The specific implementation process includes:

[0075] Step 2.1: Task complexity analysis; Task complexity is a key factor affecting computing power requirements, which usually depends on the type of task, data volume, model scale and parameters, etc. In order to evaluate the complexity of the task, the system will analyze based on the following aspects: 1) Task type. Different types of AIGC tasks, such as image generation, text generation, video processing, and deep learning model training, have different requirements for computing power; 2) Data volume; The system will analyze the size of the task input data, such as image resolution, text length, or video frame number; 3) Model parameters; If the task involves the training or reasoning of an AI model, the system will evaluate the computing power requirements based on the number of model parameters, such as the number of neural network layers, the number of parameters, and the model scale; 4) Execution time requirements; If the task has a clear execution time requirement, such as the need to be completed within a certain time, the system will include this requirement in the evaluation scope and improve the demand assessment for computing power resources accordingly.

[0076] Step 2.2: Calculate the required computing resources; after completing the task complexity analysis, estimate the complexity coefficient according to the preset complexity coefficient calculation model;

[0077] The core of the computational model is as follows:

[0078] C=C type *C Data *C Model *C Time

[0079] Where C represents the complexity coefficient; C type Indicates the task type; C Data Data volume; C Model Represents the AI ​​model parameters for generating AIGC tasks; C Time Indicates execution time;

[0080] Calculate computing power requirements based on the complexity coefficient of the task;

[0081] A computing power demand prediction model based on historical task execution data was established. Through machine learning algorithms (such as linear regression, neural network, decision tree, etc.), the model can predict the required computing power according to the task complexity coefficient C.

[0082] Step 2.3: Node quantity evaluation; After obtaining the computing power resources required for the task, the system will evaluate the number of nodes required for the task based on the computing power resource status of each node in the current edge network.

[0083] Specifically, node status monitoring: obtain the computing power status of each edge node in real time through the blockchain network, including information such as the number of available CPU / GPU cores, memory capacity, and network bandwidth.

[0084] Node screening: Filter out nodes that meet the requirements based on the computing resources required for the task. The screening criteria include the node's computing power, current load, available memory, and network bandwidth. For example, if a task requires 8 GPU cores, but a node only has 2 GPU cores, then the node needs to participate in the task calculation together with other nodes.

[0085] Calculation of the number of nodes: Calculate the number of nodes required based on the total computing power requirements of the task and the available resources of each node. The principle for determining the number of nodes is to select the largest number of nodes. For example, if the task requires 32 GPU cores and each node can provide an average of 4 GPU cores, 8 nodes are required to participate in the calculation. At the same time, the system will also consider factors such as network transmission delay and node reliability to ensure that the number and combination of nodes selected can meet the task requirements.

[0086] During the task execution process, the computing power demand may change due to the actual task execution progress and node status changes. The present invention dynamically adjusts the computing power allocation through blockchain smart contracts based on real-time monitoring information. For example, when a node is overloaded or fails, the system will automatically increase the participation of other nodes or reallocate tasks to ensure that the task can be completed on time.

[0087] Step 3: Task scheduling and computing power allocation;

[0088] Dynamic scheduling of tasks is automatically achieved through smart contracts; the contract will monitor all edge nodes registered in the blockchain network in real time, check each node’s current load, available computing power, network status and other information, and schedule appropriate nodes to participate in the calculation according to task requirements.

[0089] On-demand allocation: Smart contracts select appropriate nodes based on the complexity of the task to avoid over-use of computing resources. For example, simple tasks will be assigned to nodes with low load, while complex tasks will be assigned to nodes with stronger computing power.

[0090] Dynamic adjustment: During the task execution, the contract will continuously monitor the running status of the nodes. If a node's performance degrades due to insufficient resources or network delays, the contract will reschedule other nodes to take over the calculation to ensure that the task can be completed on time.

[0091] For example, when an AIGC task is released, the smart contract detects that five edge nodes are idle. After calculating the computing power of each node, the contract splits the task into multiple subtasks and assigns them to the five nodes for execution. Each node receives the corresponding subtask and performs the calculation locally.

[0092] Step 4: Task execution monitoring;

[0093] Once the task is assigned to each node, the smart contract will monitor the task execution in real time; the distributed ledger in the blockchain network will record the task execution progress and computing power usage of each node; all task execution operation logs will be recorded on the blockchain to ensure the transparency and immutability of task execution;

[0094] Progress monitoring: When each node performs a task, it will regularly report the progress of the task to the smart contract. The contract will record this progress information, and users can query the execution status of the task through the blockchain.

[0095] Fault-tolerance mechanism: If a node exits due to a fault or is overloaded during task execution, causing the task to be delayed, the smart contract will automatically detect the anomaly through the blockchain network and dynamically schedule other available nodes to take over the task to ensure the continuity of task execution.

[0096] Step 5: Task completion and settlement mechanism;

[0097] When the task is completed, the smart contract will automatically settle the reward according to the computing power contributed by each node and the efficiency of task completion. The contract will distribute the reward according to the execution of the task and the proportion of the computing power contributed by the node to ensure the fairness and transparency of the settlement.

[0098] Task completion report: After each node completes the assigned subtask, it uploads the execution results to the blockchain and submits a task completion report. The smart contract verifies the task execution status of all nodes and confirms whether the task is completed as expected.

[0099] Reward distribution: The smart contract distributes rewards according to the pre-set reward rules and the computing power contribution of each node. The payment of rewards is encrypted through the blockchain to ensure the transparency of the transaction.

[0100] Example 3

[0101] Embodiment 3 is a preferred embodiment of Embodiment 1 or Embodiment 2

[0102] A method for dynamic sharing configuration of computing power for artificial intelligence-generated content provided by the present invention includes: dynamic adjustment of computing power and load balancing mechanism;

[0103] Real-time computing power monitoring and load balancing:

[0104] In a distributed edge network, the computing power resources of edge nodes are dynamically changing and may change at any time due to the load of other tasks. To ensure the continuous and efficient execution of tasks, the present invention monitors the load of nodes in real time through smart contracts and performs load balancing according to the changes in the computing power of nodes.

[0105] Dynamic adjustment of computing power: During the execution of a task, if a node's computing power resources are short (for example, network interruption or other computing tasks occupying resources), the smart contract will automatically reduce the task load of the node and transfer part of the computing tasks to other low-load nodes.

[0106] Fault recovery: When a node completely loses its response, the smart contract will immediately stop the task of that node and reschedule other nodes to take over the task through the blockchain network to ensure that the task is completed without affecting the overall execution progress.

[0107] Task priority scheduling: In order to ensure that high-priority tasks can obtain sufficient computing resources in a timely manner, the smart contract will prioritize the tasks according to their importance when scheduling them. High-priority tasks will be assigned to nodes with more computing resources, while low-priority tasks will be assigned to nodes with fewer resources.

[0108] High-priority tasks: For example, when an AIGC task is required to be completed in a shorter period of time, the smart contract will prioritize allocating more computing resources to the task, and suspend or postpone the execution of low-priority tasks when necessary to ensure that high-priority tasks are completed on time.

[0109] Low-priority tasks: For low-priority tasks, the system will choose to execute them when the node is idle to avoid resource competition with high-priority tasks.

[0110] Example 4

[0111] Embodiment 4 is a preferred embodiment of Embodiment 1 or Embodiment 2

[0112] A method for dynamic sharing configuration of computing power for artificial intelligence-generated content provided by the present invention includes: automated optimization of smart contracts;

[0113] As the number of tasks and computing nodes increases, smart contracts need to have scalability and automatic optimization capabilities. The smart contracts in this invention continuously optimize the computing power allocation strategy by monitoring node performance and task execution in real time.

[0114] Optimization based on historical data: The system can analyze the performance of each node based on the historical task execution data recorded in the blockchain, and dynamically adjust the computing power allocation rules. For example, some nodes may perform better in a specific time period, and smart contracts can give priority to these nodes to execute time-sensitive tasks; among them, the performance of each node includes: 1) Computing power: including CPU / GPU / memory capacity and utilization. 2) Network performance: network bandwidth / network latency / network stability. 3) Storage performance: storage capacity / storage read and write speed. 4) Task execution history: completion rate / number of task failures. 5) Energy efficiency: power consumption and computing power ratio. 6) Task parallel processing capability: the ability of the node to handle multiple tasks at the same time, and computing performance under high load conditions.

[0115] Intelligent scheduling rule optimization: The contract can automatically optimize the task scheduling algorithm according to the execution efficiency of the task, the response speed of the node and the load status to ensure the fairness and efficiency of the computing power allocation.

[0116] Scalability and cross-platform application: The smart contract mechanism of the present invention is highly scalable and can be applied to different blockchain platforms (such as Ethereum, Hyperledger, etc.) and various distributed computing scenarios. By introducing cross-chain technology, contracts can realize the sharing of computing resources and task scheduling among multiple blockchain networks, further expanding the application scenarios of the system.

[0117] The present invention also provides a system for dynamically sharing and configuring computing power for artificial intelligence-generated content. The system for dynamically sharing and configuring computing power for artificial intelligence-generated content can be implemented by executing the process steps of the method for dynamically sharing and configuring computing power for artificial intelligence-generated content, that is, those skilled in the art can understand the method for dynamically sharing and configuring computing power for artificial intelligence-generated content as a preferred implementation of the system for dynamically sharing and configuring computing power for artificial intelligence-generated content.

[0118] The present invention realizes the dynamic sharing and scheduling of computing power of artificial intelligence generated content (AIGC) tasks in distributed edge networks through blockchain smart contracts. Through smart contracts, the system can dynamically allocate computing power resources according to the requirements of tasks and the resource status of nodes to ensure the efficiency, transparency and security of task execution. This decentralized scheduling mechanism greatly improves the utilization rate of computing power resources in edge networks and ensures fairness and reliability in the task execution process.

[0119] Those skilled in the art know that, in addition to realizing the system and its various devices, modules, and units provided by the present invention in a purely computer-readable program code, it is entirely possible to realize the same functions in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered as a hardware component, and the devices, modules, and units included therein for realizing various functions can also be regarded as structures within the hardware component; the devices, modules, and units for realizing various functions can also be regarded as both software modules for realizing the method and structures within the hardware component.

[0120] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A method for dynamically sharing and configuring computing power for artificial intelligence-generated content, characterized in that: include: Step S1: In a distributed edge network, multiple edge nodes are connected through blockchain technology to form a decentralized computing resource sharing network; Step S2: Obtain AIGC tasks posted by users and evaluate the computing power requirements of AIGC tasks; Step S3: According to the real-time computing resource status of the edge nodes in the decentralized computing resource sharing network and the computing power requirements of the AIGC task, the number of nodes required for the current AIGC task is evaluated, and computing tasks are allocated based on the number of nodes required for the current AIGC task.

2. The method for dynamic sharing configuration of computing power for artificial intelligence-generated content according to claim 1, characterized in that: The step S1 includes: the distributed edge network includes multiple edge nodes, and the multiple edge nodes include IoT devices, edge servers and mobile devices; multiple edge nodes that have been registered on the blockchain are connected through blockchain technology, and resources of the multiple edge nodes are dynamically monitored through smart contracts.

3. The method for dynamic sharing configuration of computing power for artificial intelligence-generated content according to claim 1, characterized in that: The step S2 comprises: Step S2.1: Obtain the AIGC task posted by the user; wherein the AIGC task includes: task type, task data volume, AI model parameters for generating the AIGC task, and execution time requirements; Step S2.2: Calculate the complexity coefficient of the AIGC task posted by the current user, and evaluate the computing power requirement based on the complexity coefficient of the task; The step S2.2 comprises: Step S2.2.1: Calculate the complexity coefficient of the AIGC task issued by the current user; C=C type *C Data *C Model *C Time Where C represents the complexity coefficient; C type Indicates the task type; C Data Data volume; C Model Represents the AI ​​model parameters for generating AIGC tasks; C Time Indicates execution time; Step S2.2.2: Build a computing power demand prediction model based on the machine learning algorithm, and use the historical task execution data to train the computing power demand prediction model to obtain the trained computing power demand prediction model; use the trained computing power demand prediction model to predict the computing power demand based on the complexity coefficient of the AIGC task released by the current user.

4. The method for dynamically sharing and configuring computing power for artificial intelligence-generated content according to claim 1, characterized in that: The step S3 comprises: Step S3.1: The smart contract obtains the status information of each edge node in real time through the blockchain, and the status information of the edge node includes the current load of the edge node, the available computing power and the network status information; Step S3.2: Filter edge nodes that meet the preset requirements based on the computing power requirements of the current AIGC task and the acquired status information of each edge node; Step S3.3: Select edge nodes based on the computing power scheduling mechanism according to the computing power requirements of the current AIGC task and the edge nodes that meet the preset requirements; The computing power scheduling mechanism includes: determining the number of selected edge nodes based on the computing power requirements of the current AIGC task and the edge nodes that meet the preset requirements and maximizing the number of edge nodes; When the computing power of the edge node is greater than the preset value and is in an idle state, the current edge node is preferentially scheduled to participate in task calculation; When the load of the edge node is greater than the preset value, the tasks assigned to the current node are automatically reduced to achieve balanced computing load of the entire network; When a task meets the preset requirements, the nodes are scheduled to participate in the execution first to ensure that the task is completed in time.

5. The method for dynamic sharing configuration of computing power for artificial intelligence generated content according to claim 1, characterized in that: Smart contracts are used to monitor each AIGC task in real time, and the distributed ledger in the blockchain network records the task execution progress and computing power usage of each node; logs of all task execution operations are recorded on the blockchain.

6. The method for dynamic sharing configuration of computing power for artificial intelligence generated content according to claim 1, characterized in that: The method also includes: when the smart contract detects that there is an abnormality in the status information of a certain edge node, dynamically scheduling other available nodes to take over the task of the current edge node and continue to execute.

7. The method for dynamic sharing configuration of computing power for artificial intelligence-generated content according to claim 1, characterized in that: The method further includes: analyzing the performance of each edge node based on the historical task execution data recorded in the blockchain, and dynamically allocating computing tasks according to the performance of each edge node obtained by the analysis; The performance of each edge node includes: computing power, network performance, storage performance, completion rate / number of task failures, energy efficiency, and task parallel processing capability.

8. The method for dynamic sharing configuration of computing power for artificial intelligence-generated content according to claim 1, characterized in that: The method also includes: realizing computing resource sharing and task scheduling among multiple blockchain networks based on smart contracts through cross-chain computing.

9. The method for dynamic sharing configuration of computing power for artificial intelligence generated content according to claim 1, characterized in that: When the task is completed, the smart contract automatically settles the remuneration based on the computing power contributed by each edge node and the efficiency of task completion.

10. A system for dynamically sharing and configuring computing power for artificial intelligence-generated content, characterized in that: include: Module M1: In a distributed edge network, multiple edge nodes are connected through blockchain technology to form a decentralized computing resource sharing network; Module M2: Obtain AIGC tasks posted by users and evaluate the computing power requirements of AIGC tasks; Module M3: According to the real-time computing resource status of the edge nodes in the decentralized computing resource sharing network and the computing power requirements of the AIGC task, the number of nodes required for the current AIGC task is evaluated, and computing tasks are allocated based on the number of nodes required for the current AIGC task.

Images