Block chain-based computing power resource dynamic scheduling system and method

Abstract

The invention discloses a computing power resource dynamic scheduling system and method based on a block chain. The system comprises a resource monitoring agent module which collects state resource signals in real time; the core trust layer module authenticates node identity through block chain smart contract registration, issues a task demand contract, defines a scheduling strategy rule, and forms a strategy constraint signal; the scheduling coordinator module receives a dynamic resource signal and a strategy constraint signal, and outputs a resource allocation instruction signal through an optimization decision algorithm; the task execution gateway module receives the resource allocation instruction signal and generates a task state feedback signal; and the dynamic reputation library module continuously receives the strategy constraint signal and the task state feedback signal, updates a node dynamic reputation score, forms a reputation weight signal, and feeds back the reputation weight signal to the scheduling coordinator module. The computing power resource dynamic scheduling system and method based on the block chain can solve the problems of trust deficiency and insufficient dynamics in computing power resource scheduling.

Description

Technical Field

[0001] The present invention relates to the interdisciplinary field of blockchain and cloud computing, and in particular to a system and method for dynamically scheduling computing resources based on blockchain. Background Art

[0002] Traditional computing power scheduling relies on centralized platforms to manage resource allocation, posing risks of single points of failure and data tampering. Users need to trust the fairness and security of the platform, but in practice, platforms may manipulate pricing, conceal resource status, or leak sensitive data. Furthermore, centralized architectures experience delayed resource status updates, making it difficult to respond promptly to sudden load fluctuations, resulting in delays for high-priority tasks and idle resources. Early decentralized solutions attempted to replace centralized platforms with blockchains, such as building computing power marketplaces based on smart contracts to enable direct transactions between resource providers and demanders. While these solutions improved transparency, they placed all scheduling decisions on-chain, making them limited by blockchain performance and unable to support high-frequency, real-time scheduling requirements. The high gas costs and network latency resulting from complex decision-making prevented the system from handling urgent computing tasks. Existing hybrid architectures attempt to move some functions off-chain, but lack effective trust mechanisms: off-chain schedulers can act maliciously and are difficult to hold accountable, and resource nodes lack the means to constrain the misreporting of performance metrics.

[0003] Furthermore, the coordinated scheduling of nodes across heterogeneous resources requires a comprehensive assessment of multi-dimensional dynamic factors such as real-time computing power, network topology, and historical reputation. Current solutions either rely on simple static weight models or employ unverified black-box algorithms, resulting in scheduling decisions that deviate from the optimal solution. In the trust verification phase, existing technologies often employ full-scale repeated computation or trusted hardware authentication. The former results in secondary consumption of computing power, while the latter increases node deployment costs and compatibility thresholds. More critically, the disconnect between resource dynamics and reputation evaluation means that high-performance but low-reputation nodes may be overscheduled, while high-reputation nodes cannot be downgraded in a timely manner when their performance suddenly degrades, reducing the overall reliability of the system. Therefore, there is an urgent need for a dynamic optimization mechanism that integrates the blockchain's trusted foundation with efficient off-chain scheduling capabilities and can couple resource status and reputation evolution in real time. Summary of the Invention

[0004] In view of the shortcomings of the above-mentioned prior art, the purpose of the present invention is to provide a blockchain-based dynamic scheduling system and method for computing power resources, which is used to solve the problems of lack of trust and insufficient dynamism in computing power resource scheduling. The present invention constructs a core trust layer constraint strategy on the chain, collects multi-dimensional resource data in real time off the chain, and generates optimization instructions by integrating a dynamic scoring model with a reputation feedback mechanism by the scheduling coordinator. The task gateway drives distributed execution and feedbacks on-chain evidence, thus achieving a closed loop of accurate resource value assessment, dynamic matching of task requirements, and trusted traceability of the scheduling process in a decentralized environment.

[0005] The present invention provides a blockchain-based computing resource dynamic scheduling system, comprising: Resource monitoring agent module, which collects hardware performance indicators, real-time load status and network topology data of distributed nodes in real time to form dynamic resource signals; The core trust layer module registers and authenticates node identities through blockchain smart contracts, publishes task requirement contracts, and defines scheduling policy rules to form policy constraint signals; The scheduling coordinator module receives dynamic resource signals and policy constraint signals, generates resource value assessment results based on a multi-dimensional dynamic scoring model, integrates task service quality requirements with real-time market supply and demand status, and outputs resource allocation instruction signals through an optimized decision-making algorithm; The task execution gateway module receives resource allocation instruction signals, triggers the distribution and collaborative execution of cross-node computing tasks, and generates task status feedback signals; Dynamic reputation database module,The dynamic reputation database module continuously receives policy constraint signals and task status feedback signals, updates the node dynamic reputation score, forms a reputation weight signal, and feeds it back to the resource value evaluation process of the scheduling coordinator module.

[0006] In one embodiment of the present invention, the resource monitoring agent module further includes a distributed probe unit and a data fusion unit; it is characterized in that: the distributed probe unit is deployed locally at each computing power node, and periodically collects underlying indicators including central processing unit core occupancy, graphics processing unit memory utilization, network round-trip delay and bandwidth availability; the data fusion unit receives original indicator data from all probe units, eliminates instantaneous noise interference through time series analysis, and integrates the network topology relationship between nodes to generate a structured dynamic resource signal; the dynamic resource signal is encapsulated as a standard data packet containing a timestamp and a unique node identifier, and is pushed to the data input interface of the scheduling coordinator module in real time.

[0007] In one embodiment of the present invention, the scheduling policy rules defined in the core trust layer module include a service quality grading strategy and a market constraint strategy; the characteristics are: the service quality grading strategy automatically generates differentiated scheduling priority labels based on the deadline threshold, minimum computing power guarantee level and fault tolerance redundancy requirements declared in the task requirement contract; the market constraint strategy presets the resource pricing floating range, transaction fee ratio and breach penalty coefficient through the smart contract; the two together constitute the executable terms in the policy constraint signal, and are synchronized in real time to the policy parsing engine of the scheduling coordinator module through the blockchain event monitoring mechanism.

[0008] In one embodiment of the present invention, the multi-dimensional dynamic scoring model in the scheduling coordinator module includes a hardware benchmark score, a real-time performance score and a network quality score; it is characterized in that: the hardware benchmark score calculates the initial static score based on the central processing unit model, graphics processor architecture and persistent storage capacity provided when the node is registered; the real-time performance score dynamically adjusts the weight by analyzing the load fluctuation trend and resource idle window in the dynamic resource signal; the network quality score is generated based on the topological hop count and historical transmission stability from the node to the task data source; the three are fused into a resource value evaluation result according to a preset weighted formula, among which the weight of the real-time performance score adaptively increases as the task urgency increases.

[0009] In one embodiment of the present invention, the optimization decision algorithm includes a demand matching stage and a resource allocation stage; it is characterized in that: in the demand matching stage, the available node pool is screened according to the geographical affinity constraints in the task service quality requirements, and a preliminary sorting is performed based on the resource value evaluation results; in the resource allocation stage, a heuristic search algorithm is used to generate a node combination plan with the goal of maximizing global resource utilization under the dual constraints of deadline and budget; the final output resource allocation instruction signal includes a target node list, a task slice mapping relationship and an execution timing control parameter.

[0010] In one embodiment of the present invention, the task execution gateway module further includes a task slicing unit and a fault-tolerant control unit; the characteristics are: the task slicing unit splits the complex computing task into several independent sub-task packages according to the mapping relationship in the resource allocation instruction signal, and adds a data integrity check code; the fault-tolerant control unit monitors the sub-task execution timeout or verification failure event, and automatically triggers the task redistribution process based on the backup node list; the task status feedback signal contains the actual start time, execution time and verification result log of each sub-task package, and is sent to the core trust layer module after being digitally signed.

[0011] In one embodiment of the present invention, the process of updating the dynamic reputation score of a node by the dynamic reputation library module includes behavior analysis and score reconstruction; it is characterized in that: the behavior analysis unit extracts the task completion time deviation rate and result verification pass rate indicators from the task status feedback signal, and parses the node's historical violation records from the policy constraint signal; the score reconstruction unit inputs the above indicators into the reputation decay model, wherein the recent behavior data is calculated using exponential weighted accumulation, and the reputation base is calibrated in combination with the total online time of the node. The final generated reputation weight signal includes a reputation level label and a scheduling recommendation weight value.

[0012] In one embodiment of the present invention, the blockchain-based dynamic scheduling system for computing power resources further includes a cross-chain verification relay module; its characteristics are: the module is deployed between the core trust layer module and the external blockchain network, and automatically generates a cross-chain resource query request when a task requirement contract involving cross-chain computing power resource scheduling is detected; the credibility certificate and resource quotation data of the node on the external chain are obtained through the relay bridge, and after format conversion, they are input into the scheduling coordinator module as an extension policy constraint signal; the settlement data generated by the task execution gateway module is forwarded by the relay module to the external chain for atomic cross-chain settlement.

[0013] In one embodiment of the present invention, the reputation weight signal feedback mechanism of the dynamic reputation library module includes a dual mode of active push and passive response; its characteristics are: in the active push mode, whenever the change in the node reputation score exceeds the set threshold, an incremental update signal is immediately sent to the scheduling coordinator module; in the passive response mode, the scheduling coordinator module sends a reputation query request before starting resource evaluation, and the dynamic reputation library module returns the latest full reputation snapshot; the two modes are distinguished by a signal type identifier, and the scheduling coordinator module dynamically switches and calls according to the real-time decision-making requirements.

[0014] The present invention also includes a method for dynamically scheduling computing resources based on blockchain, comprising: S1: Real-time collection of hardware performance indicators, real-time load status, and network topology data of distributed nodes to form dynamic resource signals; S2: Register and authenticate node identities through blockchain smart contracts, publish task requirement contracts, and define scheduling policy rules to form policy constraint signals; S3: Receives dynamic resource signals and policy constraint signals, generates resource value assessment results based on a multi-dimensional dynamic scoring model, integrates task service quality requirements with real-time market supply and demand status, and outputs resource allocation instruction signals through an optimized decision-making algorithm; S4: Receives resource allocation instruction signals, triggers the distribution and coordinated execution of cross-node computing tasks, and generates task status feedback signals; S5: Continuously receive policy constraint signals and task status feedback signals, update the node dynamic reputation score, form a reputation weight signal, and feed it back to the resource value evaluation process of the scheduling coordinator module.

[0015] The blockchain-based dynamic scheduling system and method for computing resources provided by the present invention builds a core trust layer constraint strategy on the chain, collects multi-dimensional resource data in real time off the chain, and generates optimization instructions by integrating the dynamic scoring model and the reputation feedback mechanism by the scheduling coordinator. The task gateway drives distributed execution and feeds back the on-chain evidence, thus realizing a closed loop of accurate resource value assessment, dynamic matching of task requirements, and trusted traceability of the scheduling process in a decentralized environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 This is the system architecture diagram of the dynamic scheduling system for computing resources based on blockchain; Figure 2 A schematic diagram showing the operation process of the dynamic scheduling system for computing resources on the blockchain; Figure 3 The flowchart of the method for dynamic scheduling of computing resources based on blockchain. DETAILED DESCRIPTION

[0018] The following describes the embodiments of the present invention through specific examples. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments. The details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other unless they conflict.

[0019] It should be noted that the illustrations provided in the following embodiments are merely schematic illustrations of the basic concept of the present invention. Therefore, the illustrations only show components related to the present invention and are not drawn according to the number, shape, and size of components in actual implementation. In actual implementation, the type, quantity, and proportion of each component may be changed arbitrarily, and the component layout may also be more complex.

[0020] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the embodiments of the present invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring the embodiments of the present invention.

[0021] See Figure 1-3, which shows the blockchain-based dynamic scheduling system and method for computing power resources of the present invention. The blockchain-based dynamic scheduling system for computing power resources of the present invention includes a resource monitoring agent module, which collects hardware performance indicators, real-time load status, and network topology data of distributed nodes in real time to form dynamic resource signals; a core trust layer module, which registers and authenticates node identities through blockchain smart contracts, publishes task requirement contracts, and defines scheduling policy rules to form policy constraint signals; a scheduling coordinator module, which receives dynamic resource signals and policy constraint signals, generates resource value assessment results based on a multi-dimensional dynamic scoring model, integrates task service quality requirements with real-time market supply and demand status, and outputs resource allocation instruction signals through an optimized decision algorithm; a task execution gateway module, which receives resource allocation instruction signals, triggers the distribution and coordinated execution of cross-node computing power tasks, and generates task status feedback signals; and a dynamic reputation library module, which continuously receives policy constraint signals and task status feedback signals, updates node dynamic reputation scores, forms reputation weight signals, and feeds back to the resource value assessment process of the scheduling coordinator module.

[0022] like Figure 1As shown in the figure, the blockchain-based dynamic computing resource scheduling system comprises a collaborative operating mechanism of five core modules. The resource monitoring agent module, serving as the system perception layer, is embedded in each distributed computing node via a lightweight agent program and continuously captures three key dynamic metrics: hardware performance metrics encompass static attributes such as the CPU base frequency, the number of GPU stream processors, and the persistent storage read / write rate; real-time load status includes fluctuating data such as the instantaneous CPU core utilization, the GPU memory usage ratio, and the memory swap frequency; and network topology data, obtained through inter-node probe communication, encompasses network quality parameters such as the number of transmission path hops, round-trip latency, and packet loss rate. These raw metrics are sampled at millisecond frequencies, cleaned to remove sensor noise, and aggregated into structured dynamic resource signals according to preset time windows. This signal, encapsulated in binary code with a unique node identifier, a timestamp sequence, and multi-dimensional indicator key-value pairs, is then pushed to the message middleware via a publish-subscribe model. The core trust layer module is built on a permissioned blockchain network. Its smart contracts include three core logics: node registration contract, task release contract, and policy management contract. The node registration contract requires resource providers to pledge digital assets and submit hardware certification reports to generate trusted identity credentials. The task release contract parses the task descriptor submitted by the demander and generates a task requirement contract containing the service quality level, budget cap, and data source location. The policy management contract allows authorized nodes to vote on the dynamic update of scheduling policy rules, and finally compiles the identity credentials, task contracts, and policy rules into a policy constraint signal, which is broadcast to the entire network in the form of a blockchain transaction event. The scheduling coordinator module serves as the decision-making center and sets up a dedicated message queue to receive dynamic resource signals and policy constraint signals. The signal parsing engine first separates the node identity whitelist, task priority label and market constraint parameters in the policy constraint signal, and decodes the dynamic resource signal into a real-time node status matrix. The multi-dimensional dynamic scoring model runs the scoring process based on this matrix: the hardware benchmark score calls the static parameters stored in the registration contract to calculate the theoretical computing power value, the real-time performance score is dynamically downgraded according to the degree to which the load indicator deviates from the baseline, and the network quality score combines the task data source location and node network indicators to calculate the transmission cost. The three scoring inputs are weighted and fused into a unit to generate a resource value assessment result, where the weighting coefficient is dynamically adjusted by the service quality requirements in the policy constraint signal. The optimization decision algorithm is then started: the demand matching stage filters the node pool according to the task geographical affinity constraint, and the resource allocation stage uses the taboo search algorithm to solve the global resource utilization maximization plan under the deadline and budget constraints, and finally generates a resource allocation instruction signal containing a target node address list, a task shard mapping table and an execution timeout threshold.After receiving the command signal, the task dispatcher splits the computational task into atomic subtask packages based on a mapping table and distributes them to the target node via a peer-to-peer encrypted channel. The node's local execution engine runs the subtasks and generates a local status report containing execution timestamps and resource consumption logs. The collaborative controller aggregates all local reports, verifies data consistency, and assembles them into a global task status feedback signal, which is then digitally signed and sent to the blockchain network. The dynamic reputation library module, deployed in an off-chain database, monitors the blockchain to obtain the node's initial reputation value from the policy constraint signal and the execution results from the task status feedback signal. The reputation update engine periodically scans new data, and the behavior analyzer extracts the task completion time deviation rate, result verification pass rate, and resource data authenticity mark. These are input into a reputation model based on an exponential decay function: recent behavior data is assigned a higher weight, while malicious behavior records trigger a punitive score reduction. The final output is a dynamic reputation score. This score is integrated with the node's online time and historical stability indicators to form a reputation weight signal, which is then called back to the score fusion unit of the scheduling coordinator module via an application programming interface, achieving a closed-loop coupling between resource value assessment and reputation feedback.

[0023] Furthermore, the resource monitoring agent module adopts a two-layer architecture of distributed probe units and data fusion units. The distributed probe unit is embedded in each computing node in the form of an operating system kernel module, and includes three types of collectors: hardware performance probes, real-time load probes, and network topology probes. The hardware performance probe is activated when the node is registered, obtains the instruction set extension support flag by accessing the CPU model-specific register, calls the graphics processor driver application interface to read the number of shader cores and video memory bandwidth data, and scans the interface protocol and queue depth parameters of the persistent storage device; the real-time load probe polls system resource statistics at a frequency of seconds, including the interrupt request rate of each core of the CPU, the active cycle ratio of the graphics processor computing unit, the memory page fault exception count, and the network adapter buffer overflow event; after the network topology probe is started, it actively sends link layer discovery protocol packets to adjacent nodes, builds an initial topology map based on the response delay and path trajectory, and then dynamically maintains the topology relationship by monitoring the Open Shortest Path First protocol routing update message. The data fusion unit, deployed at the regional aggregation node, receives the raw metrics stream reported by the probe units. Its data processing pipeline consists of four operations: a noise filtering layer uses a sliding window average algorithm to eliminate instantaneous jitter, for example, applying a Hanning window smoothing to CPU utilization. A data alignment layer corrects node timestamps based on the network time protocol server to ensure cross-node metrics are synchronized. A correlation analysis layer detects causal relationships between hardware performance and real-time load, for example, identifying whether a sudden increase in video memory utilization is accompanied by idle GPU compute units. A topology integration layer injects inter-node network latency metrics into the initial topology graph to generate a weighted dynamic topology matrix. The resulting structured dynamic resource signal uses a protocol buffer encoding scheme: a nested data structure is defined. The top layer contains a global sequence number and data collection round number, the middle layer is indexed by node unique identifiers, and the bottom layer is divided into static attribute segments, dynamic load segments, and network topology segments. The static attribute segment solidifies the hardware benchmark parameters after registration and authentication, the dynamic load segment stores metric statistics within the most recent time window, and the network topology segment records a list of path weights from the node to the core switching nodes. The signal is multicast to the designated port of the scheduling coordinator cluster via the User Datagram Protocol, and the receiving end enables cyclic redundancy check to ensure data transmission integrity.

[0024] Specifically, the scheduling policy rule system in the core trust layer module is divided into two subsystems: service quality grading strategy and market constraint strategy. The service quality grading strategy acts on the task release contract parsing stage: when the demander submits a computing task descriptor, the policy engine scans the service quality declaration field in the descriptor, including the task's latest completion timestamp, minimum computing power guarantee level, and fault tolerance redundancy requirements; time-sensitive tasks are divided into urgency levels based on the difference between the deadline and the current blockchain time, and the corresponding scheduling priority label is generated; the computing power guarantee level is mapped to the minimum threshold combination of the number of CPU cores, GPU floating-point capability, and memory capacity; the fault tolerance redundancy requirement is converted into the number of task replicas and the fault switching time window parameter, the three of which together constitute a multi-dimensional vector of the task priority label. Market constraint strategies are configured through on-chain governance contracts: A floating resource pricing rule defines a benchmark computing power price curve, which adjusts daily based on historical market supply and demand ratios and sets an upper limit on price fluctuations per transaction. Transaction fee ratios are designed according to a tiered model, with large transactions receiving rate discounts but requiring additional on-chain settlement gas fees. The default penalty coefficient matrix encompasses two scenarios: provider default and demand default. The former covers situations like node disconnection, insufficient computing power, and erroneous results, while the latter focuses on payment timeouts and misrepresented task descriptions. Penalty amounts are calculated based on the collateral ratio and the severity of the default. The executable clause generation process for the policy rules is as follows: a task priority label vector is input into the policy compiler, generating a binary code snippet containing the task scheduling weight coefficients. Market constraint parameters are compiled into smart contract code blocks for the resource pricing function, fee calculation function, and default adjudication function. These functions are then verified by consensus among blockchain nodes and written into a new block as the payload of the policy constraint signal. The blockchain event listening mechanism achieves policy synchronization: the blockchain listener deployed by the scheduling coordinator module subscribes to the storage change events of the policy management contract. When it detects that a new block contains a policy update transaction, it triggers the event parser to extract the policy operation code; after the operation code is simulated and executed in a secure sandbox environment, it outputs an incremental update package of the policy constraint signal and updates the local policy cache library through the remote procedure call interface; the policy cache library adopts a version control mechanism to ensure that scheduling decisions are always based on the latest effective policy version, while retaining historical versions for audit traceability.

[0025] In one embodiment of the present invention, the multi-dimensional dynamic scoring model in the scheduling coordinator module is composed of three core dimensions: hardware benchmark score, real-time performance score, and network quality score. The hardware benchmark score calculation process calls the node registration information stored in the core trust layer module, extracts the CPU instruction set extension capability flag, combines the number of cores in the graphics processor unified computing device architecture and the random read and write rate indicators of persistent storage, and uses a static weighted formula to generate a theoretical computing capability index. The instruction set extension capability is graded and assigned according to key technologies such as the support vector machine instruction set and matrix extension instruction set. The graphics processor parameters are converted into theoretical floating-point performance values ​​based on the number of stream processors and the ratio of texture units. Storage performance is quantified through queue depth test data. Real-time performance scores are dynamically calculated based on a second-level sampling sequence reported by the resource monitoring agent module. A time series model of CPU core utilization is established, and future short-term load trends are predicted using an autoregressive integral sliding average algorithm. Graphics processor performance evaluation incorporates a multiplication factor of the actual memory bandwidth utilization and the active period of the computing unit. Memory subsystem performance is dynamically downgraded based on page fault frequency and swap partition utilization. The final performance score is a function of the negative deviation of the measured value from the baseline theoretical value. Network quality scores are generated based on specific task requirements. First, the geographic coordinates of the task data source in the policy constraint signal are parsed, and the number of network topology hops from the target node to the data source are calculated. The average round-trip delay and standard deviation of the packet loss rate for this path are calculated using historical transmission logs. A network congestion perception coefficient is introduced, which is dynamically adjusted by monitoring the recent rate of change in the transmission control protocol congestion window size. The three scoring inputs are fused into the engine to perform weighted calculations: the hardware benchmark score is fixed at the baseline value, the real-time performance score is positively correlated with the urgency stated in the task requirement contract, and the network quality score is automatically increased for data transmission-intensive tasks. The fusion engine has a built-in normalization processing unit to eliminate dimensional differences, applies a score attenuation factor to ultra-low credibility nodes, and outputs a resource value assessment result matrix containing the three-dimensional sub-scores and comprehensive score values ​​of each node for the current task. This matrix serves as the core input parameter of the optimization decision algorithm.

[0026] like Figure 2As shown, the optimization decision algorithm is divided into two sequentially executed logical units: the demand matching phase and the resource allocation phase. The demand matching phase first constructs a node screening funnel: a geographic affinity constraint parses the location-sensitive flag in the task demand contract. If this constraint is enabled, only nodes within the same autonomous system as the task's data source are selected. A computing power threshold constraint filters low-performance nodes based on the task's minimum computing power guarantee level; and a reputation access constraint excludes blacklisted nodes marked by the dynamic reputation library module. Nodes passing through the funnel form an initial pool of available nodes, which are then sorted in descending order by the comprehensive score from the resource value assessment results. The resource allocation phase employs an improved tabu search algorithm to solve a multi-constrained optimization problem: the solution space is defined as a combination of subsets of the available node pool, and the objective function is to maximize global resource utilization, defined as the ratio of the total task computational effort to the sum of the actual node computing power. Constraints include the task deadline threshold and the demander's budget cap. The time constraint uses the critical path method to estimate the maximum subtask execution time, while the budget constraint integrates the resource pricing function with the estimated execution time. During the algorithm's initialization phase, a random feasible solution is generated as the current solution. Neighborhood operations employ a node replacement strategy—randomly selecting the lowest-scoring node in the current solution and searching for a higher-scoring, constraint-satisfying replacement node among the unselected nodes. A taboo table records the identifiers of recently replaced nodes, prohibiting them from re-entering the solution set within a set number of iterations to prevent cyclic oscillation. The amnesty rule is activated when the objective function value of a candidate solution outperforms the historical optimal solution, allowing for violations of the taboo restriction. The algorithm terminates when the maximum number of iterations is reached or the number of consecutive, unimproved iterations exceeds the limit. The final output resource allocation instruction signal contains a four-tuple structure: a target node address list identifies the physical nodes participating in the computation; a task shard mapping table defines the binding relationships between subtasks and nodes after the complex task is decomposed; execution timing control parameters specify the time window for each subtask to initiate and the location of synchronization barriers; and a dynamic pricing list lists the actual settlement price per unit of computing power for each node, generated based on real-time market supply and demand within the floating range allowed by the policy constraint signal.

[0027] Furthermore, the task execution gateway module ensures distributed execution reliability through a collaborative mechanism between the task sharding unit and the fault-tolerant control unit. Upon receiving a resource allocation command signal, the task sharding unit activates the sharding engine. Compute-intensive tasks are split into several homogeneous subtask packages based on data parallelism. Each package uses the same computational logic but processes different data blocks. For pipeline tasks, a functional splitting strategy is employed, dividing the processing flow into heterogeneous subtask stages that are executed sequentially. Each subtask package is accompanied by an integrity check mechanism: data block tasks generate a Merkle tree root hash value and write it into the packet header; algorithmic logic tasks inject a lightweight cyclic redundancy check (CRC) code. The sharding engine also generates a task dependency graph, annotating the predecessor and successor constraints between subtasks. Subtask packages are distributed to target nodes via an encrypted transmission channel, using quantum-resistant encryption algorithms to encapsulate the transmission payload. The fault-tolerant control unit deploys a heartbeat monitor and timeout timer for dual security. The heartbeat monitor receives execution progress messages from each node at a sub-second frequency, containing the percentage of processed data and a snapshot of resource consumption. The timeout timer initializes the maximum allowed execution time for each subtask based on execution timing control parameters. When a subtask is detected to have stalled or timed out, the exception handler initiates a three-level response process: the primary response sends an execution status query request to the target node. If it is confirmed to be a temporary failure, the deadline is extended; the intermediate response selects a node of the same specification from the backup node list, redistributes the subtask package, and migrates the intermediate state data; the advanced response executes the same subtask in parallel on multiple backup nodes, and uses a majority consensus mechanism to determine the correct result. After all subtasks are completed, the result aggregator verifies the integrity check code of each subtask package and merges them to generate the final calculation result. The task status feedback signal contains a structured log sequence: each subtask package records the actual start timestamp, completion timestamp, number of CPU clock cycles consumed, and number of GPU stream processor instructions; the abnormal event is individually marked with a type code and processing process; the entire task is attached with a digital signature and result hash value, and submitted to the core trust layer module through the blockchain application program interface.

[0028] like Figure 3The figure shows a blockchain-based dynamic computing resource scheduling method of the present invention. S1: Real-time collection of hardware performance indicators, real-time load status, and network topology data of distributed nodes to form dynamic resource signals. S2: Node identity registration and authentication, task requirement contracts, and scheduling policy rules are published through blockchain smart contracts to form policy constraint signals. S3: Receive the dynamic resource signal and policy constraint signal, generate resource value assessment results based on a multi-dimensional dynamic scoring model, integrate task service quality requirements with real-time market supply and demand status, and output resource allocation instruction signals through an optimized decision algorithm. S4: Receive the resource allocation instruction signal, trigger the distribution and collaborative execution of cross-node computing tasks, and generate a task status feedback signal. S5: Continuously receive the policy constraint signal and the task status feedback signal, update the node dynamic reputation score, form a reputation weight signal, and feed it back to the resource value assessment process of the scheduling coordinator module.

[0029] Specifically, the dynamic reputation database module's node dynamic reputation score update mechanism is implemented in tandem by a behavior parsing unit and a score reconstruction unit. The behavior parsing unit monitors two key data streams: extracting the initial reputation seed value at node registration and a summary of historical violation records from the policy constraint signal; parsing three types of behavioral evidence from the task status feedback signal: the task completion time deviation rate is calculated by comparing the actual completion timestamp with the planned completion time window in the resource allocation instruction signal; the result verification pass rate is calculated by counting the number of subtask package verification failures as a percentage of the total task count; and the resource data authenticity tag is added by the resource monitoring agent module upon detecting abnormal data fluctuations. The score reconstruction unit implements a time-decay-based reputation model: a reputation base value update formula is established, in which the weight of the initial reputation seed value decays exponentially with online time, and recent behavioral data is weighted and accumulated using a sliding window. The specific calculation process defines the behavioral evidence weight coefficient: the portion of the task completion time deviation rate exceeding the threshold is amplified by a quadratic function penalty factor; a result verification failure is considered a serious violation, triggering a step-by-step score reduction; and a fixed penalty value is immediately imposed upon confirmed false reporting of resource data. The model introduces a positive incentive mechanism: nodes that successfully complete tasks continuously and maintain stable resource data receive accelerated points, with the rate of point growth positively correlated with task complexity. Normalization is performed before scoring output: the original credit score is mapped to a standardized range of zero to one hundred, and a credit cap is set for long-term, highly stable nodes. The resulting credit weight signal contains three data components: a dynamic credit score value with two decimal places of precision; credit level labels divided into four levels: trusted, observed, restricted, and blacklisted, based on the score range; and a scheduling recommendation weight value that is a linear function of the credit score and is used to adjust node sorting priorities in the resource value assessment of the scheduling coordinator module. This signal is asynchronously pushed to the scheduling coordinator module's credit input buffer via a message queue and simultaneously written to a distributed database for audit queries.

[0030] The blockchain-based computing resource dynamic scheduling system and method of the present invention builds a core trust layer constraint strategy on the chain, collects multi-dimensional resource data in real time off the chain, and generates optimization instructions by integrating the dynamic scoring model and the reputation feedback mechanism by the scheduling coordinator. The task gateway drives distributed execution and feeds back on-chain evidence, thus realizing a closed loop of accurate resource value assessment, dynamic matching of task requirements, and trusted traceability of the scheduling process in a decentralized environment.

[0031] Therefore, the problems of lack of trust and insufficient dynamism in computing power resource scheduling are solved through the blockchain-based computing power resource dynamic scheduling system and method of the present invention.

[0032] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the present invention. Anyone skilled in the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by one of ordinary skill in the art without departing from the spirit and technical principles disclosed herein are intended to be covered by the claims of the present invention.

Claims

1. The dynamic scheduling system of computing resources based on blockchain is characterized by: include: A resource monitoring agent module, which collects hardware performance indicators, real-time load status and network topology data of distributed nodes in real time to form dynamic resource signals; A core trust layer module, which registers and authenticates node identities through blockchain smart contracts, publishes task requirement contracts, and defines scheduling policy rules to form policy constraint signals; a scheduling coordinator module, which receives the dynamic resource signal and the policy constraint signal, generates a resource value assessment result based on a multi-dimensional dynamic scoring model, integrates the task service quality requirements with the real-time market supply and demand status, and outputs a resource allocation instruction signal through an optimized decision algorithm; A task execution gateway module receives a resource allocation instruction signal, triggers the distribution and collaborative execution of cross-node computing tasks, and generates a task status feedback signal; A dynamic reputation database module continuously receives the policy constraint signal and the task status feedback signal, updates the node dynamic reputation score, forms a reputation weight signal, and feeds back to the resource value evaluation process of the scheduling coordinator module.

2. The blockchain-based computing resource dynamic scheduling system according to claim 1 is characterized in that: The resource monitoring agent module further includes a distributed probe unit and a data fusion unit; it is characterized in that: the distributed probe unit is deployed locally at each computing power node, and periodically collects underlying indicators including central processing unit core occupancy, graphics processing unit memory utilization, network round-trip delay and bandwidth availability; the data fusion unit receives original indicator data from all probe units, eliminates instantaneous noise interference through time series analysis, and integrates the network topology relationship between nodes to generate a structured dynamic resource signal; the dynamic resource signal is encapsulated as a standard data packet containing a timestamp and a unique node identifier, and is pushed to the data input interface of the scheduling coordinator module in real time.

3. The blockchain-based computing resource dynamic scheduling system according to claim 1 is characterized in that: The scheduling policy rules defined in the core trust layer module include a service quality grading strategy and a market constraint strategy. The characteristics of the strategy are that the service quality grading strategy automatically generates differentiated scheduling priority labels based on the deadline threshold, minimum computing power guarantee level, and fault-tolerance redundancy requirements declared in the task requirement contract. The market constraint strategy presets the resource pricing floating range, transaction fee ratio, and default penalty coefficient through smart contracts. The two together constitute the executable terms in the policy constraint signal and are synchronized in real time to the policy parsing engine of the scheduling coordinator module through the blockchain event monitoring mechanism.

4. The blockchain-based computing resource dynamic scheduling system according to claim 1 is characterized in that: The multi-dimensional dynamic scoring model in the scheduling coordinator module includes hardware benchmark scores, real-time performance scores and network quality scores; its characteristics are: the hardware benchmark score calculates the initial static score based on the central processing unit model, graphics processor architecture and persistent storage capacity provided when the node is registered; the real-time performance score dynamically adjusts the weight by analyzing the load fluctuation trend and resource idle window in the dynamic resource signal; the network quality score is generated based on the topological hop count and historical transmission stability from the node to the task data source; the three are fused into the resource value evaluation result according to a preset weighted formula, among which the weight of the real-time performance score adaptively increases as the task urgency increases.

5. The blockchain-based computing resource dynamic scheduling system according to claim 1 is characterized in that: The optimization decision algorithm includes a demand matching stage and a resource allocation stage; its characteristics are: in the demand matching stage, the available node pool is screened according to the geographical affinity constraints in the task service quality requirements, and a preliminary ranking is performed based on the resource value assessment results; in the resource allocation stage, a heuristic search algorithm is used to generate a node combination plan with the goal of maximizing global resource utilization while satisfying the dual constraints of deadline and budget; the final output resource allocation instruction signal includes a target node list, task shard mapping relationship and execution timing control parameters.

6. The blockchain-based computing resource dynamic scheduling system according to claim 1 is characterized in that: The task execution gateway module further includes a task slicing unit and a fault-tolerant control unit; its characteristics are: the task slicing unit splits the complex computing task into several independent subtask packages according to the mapping relationship in the resource allocation instruction signal, and adds a data integrity check code; the fault-tolerant control unit monitors the subtask execution timeout or verification failure event, and automatically triggers the task redistribution process based on the backup node list; the task status feedback signal contains the actual start time, execution time and verification result log of each subtask package, and is sent to the core trust layer module after being digitally signed.

7. The blockchain-based computing resource dynamic scheduling system according to claim 1 is characterized in that: The process of updating the node dynamic reputation score by the dynamic reputation library module includes behavior analysis and score reconstruction; it is characterized in that: the behavior analysis unit extracts the task completion time deviation rate and result verification pass rate indicators from the task status feedback signal, and analyzes the node's historical violation records from the policy constraint signal; the score reconstruction unit inputs the above indicators into the reputation decay model, in which the recent behavior data is calculated using exponential weighted accumulation, and the reputation base is calibrated in combination with the total online time of the node. The final generated reputation weight signal includes the reputation level label and the scheduling recommendation weight value.

8. The blockchain-based computing resource dynamic scheduling system according to claim 1 is characterized in that: The blockchain-based computing power resource dynamic scheduling system further includes a cross-chain verification relay module; its characteristics are: this module is deployed between the core trust layer module and the external blockchain network, and automatically generates a cross-chain resource query request when a task requirement contract involving cross-chain computing power resource scheduling is detected; obtains the credibility proof and resource quotation data of the node on the external chain through the relay bridge, and inputs the data into the scheduling coordinator module as an extended policy constraint signal after format conversion; the settlement data generated by the task execution gateway module is forwarded by the relay module to the external chain for atomic cross-chain settlement.

9. The blockchain-based computing resource dynamic scheduling system according to claim 1 is characterized in that: The reputation weight signal feedback mechanism of the dynamic reputation library module includes both active push and passive response modes. Its characteristics are as follows: in the active push mode, whenever the change in the node reputation score exceeds a set threshold, an incremental update signal is immediately sent to the scheduling coordinator module; in the passive response mode, the scheduling coordinator module sends a reputation query request before starting resource evaluation, and the dynamic reputation library module returns the latest full reputation snapshot. The two modes are distinguished by a signal type identifier, and the scheduling coordinator module dynamically switches between them based on the real-time decision-making requirements.

10. The method for dynamic scheduling of computing resources based on a blockchain computing resource dynamic scheduling system according to any one of claims 1 to 9, characterized in that: include: S1: Real-time collection of hardware performance indicators, real-time load status, and network topology data of distributed nodes to form dynamic resource signals; S2: Register and authenticate node identities through blockchain smart contracts, publish task requirement contracts, and define scheduling policy rules to form policy constraint signals; S3: Receive the dynamic resource signal and the policy constraint signal, generate a resource value assessment result based on a multi-dimensional dynamic scoring model, integrate the task service quality requirements with the real-time market supply and demand status, and output a resource allocation instruction signal through an optimized decision algorithm; S4: Receives resource allocation instruction signals, triggers the distribution and coordinated execution of cross-node computing tasks, and generates task status feedback signals; S5: Continuously receive the policy constraint signal and the task status feedback signal, update the node dynamic reputation score, form a reputation weight signal, and feed it back to the resource value evaluation process of the scheduling coordinator module.

Images