Distributed computing task scheduling security guarantee method and system based on dynamic trust evaluation

Abstract

The invention discloses a distributed computing task scheduling security guarantee method and system based on dynamic trust evaluation, and the method comprises the steps: collecting and preprocessing the operation state data of each computing node in real time, and dynamically calculating and updating the trust degree of each node according to the operation state data; according to the credibility and in combination with node resource loads, the tasks are only allocated to nodes with the credibility not lower than a preset threshold value and sufficient resources to be executed; the node state is continuously monitored in the task execution process, when a node fault is detected or the node trust degree is lower than the threshold value, task rescheduling is triggered, and the affected task is migrated to the remaining nodes with high trust degree and sufficient resources to continue to be executed; the key data of task allocation, node selection and task completion state are written into the block chain by using the characteristics of decentralization, non-tampering and encryption protection of the block chain so as to ensure the transparency and data security of the task execution process.

Description

Technical Field

[0001] This invention belongs to the field of distributed task scheduling and security technology, and particularly relates to a distributed computing task scheduling security method and system based on dynamic trust assessment. Background Technology

[0002] With the widespread application of cloud computing and distributed computing, ensuring data security and privacy in large-scale, highly complex computing environments has become a crucial research issue. Existing distributed computing task scheduling methods typically rely on static resource allocation strategies or task queue-based scheduling algorithms. While these methods are effective in certain scenarios, they often reveal significant shortcomings when facing dynamically changing environments. In practical applications, especially in the concurrent execution of multiple tasks in cloud computing environments, the lack of dynamic assessment of the real-time trust level of computing nodes in task scheduling systems leads to tasks not being assigned to the most reliable nodes, thus impacting system security and performance.

[0003] Furthermore, while existing blockchain technology can ensure tamper-proof capabilities during task execution and provides a certain level of security in task scheduling, it lacks sufficient security measures to address dynamic changes in nodes and network attacks. While blockchain's consensus mechanism can ensure the reliability of task execution to some extent, it still lacks effective dynamic protection and adaptive adjustment mechanisms for real-time responses to security threats and privacy violations from computing nodes in distributed computing environments. This lack of resilience and real-time protection limits the widespread application of blockchain in distributed computing task scheduling, especially in complex and ever-changing cloud computing environments, where the efficiency and security of task scheduling remain pressing challenges.

[0004] CN118656183A discloses a decentralized task scheduling system that ensures tamper-proof task execution through a blockchain consensus mechanism and stores task scheduling information in the form of a DAG graph. This method improves the efficiency of task scheduling, especially in terms of concurrent task execution and system throughput. However, existing systems still suffer from insufficient trust assessment and node security guarantees. In particular, when facing dynamic node changes and network attacks, existing technologies fail to provide sufficient security protection and adaptive capabilities, which may lead to task scheduling being affected by unreliable nodes, impacting the overall security and execution efficiency of the system. Summary of the Invention

[0005] Purpose of the invention: This invention provides a method and system for ensuring the security of distributed computing task scheduling based on dynamic trust assessment. It aims to solve the problems of low task execution reliability, high risk of data privacy leakage, and insufficient system throughput and overall security caused by the lack of real-time trust assessment and adaptive security protection mechanisms in existing distributed computing task scheduling technologies when facing dynamic changes in nodes, network attacks or failures.

[0006] Technical Solution: This invention provides a method for ensuring the security of distributed computing task scheduling based on dynamic trust assessment, comprising:

[0007] S1. Collect and preprocess the running status data of each computing node in real time, and dynamically calculate and update the trust level of each node based on the running status data.

[0008] S2. Based on the trust level and combined with the node resource load, an intelligent selection strategy is adopted to select a node with a trust level not lower than a preset threshold and sufficient resources to perform the task.

[0009] S3. During task execution, continuously monitor the node status. When a node failure or node trust level is detected to be lower than the threshold, trigger the task rescheduling rule to migrate the affected task to the remaining high-trust node with sufficient resources for continued execution.

[0010] S4. Utilize the decentralized, immutable, and cryptographically protected characteristics of blockchain to write key data such as task allocation, node selection, and task completion status into the blockchain.

[0011] Furthermore, the calculation and updating of the trust level of each node as described in S1 includes:

[0012] Collect task execution history, load data, inter-node interaction information, and behavior monitoring data; calculate task execution success rate, node load and performance, inter-node interaction quality, and abnormal node behavior.

[0013] The formula for the task execution success rate is:

[0014]

[0015] The formula for calculating node load and performance is as follows:

[0016]

[0017] Among them, β1, β2, and β3 are the weights for CPU utilization, memory utilization, and network bandwidth utilization.

[0018] The formula for calculating the interaction quality between nodes is:

[0019]

[0020] The effective data transmission rate is the amount of data successfully transmitted, excluding failed data due to delays or packet loss. The total data transmission rate refers to the amount of all data transmitted, including both successfully transmitted and failed data.

[0021] The formula for calculating abnormal node behavior is as follows:

[0022]

[0023] Where m is the type of abnormal behavior, β 4i It represents the weight of each abnormal behavior, indicating its degree of impact on the node's trust level;

[0024] Using a weighted model T n (t)=w1S n (t)+w2L n (t)+w3C n (t)+w4A n (t) Calculate the node trust level, where w1, w2, w3, w4 are the weights of each evaluation factor, and S n (t) represents the task execution success rate, L n (t) represents node load and performance, C n (t) represents the interaction quality between nodes, A n (t) indicates abnormal node behavior; T n (t) represents the current trust level of the node;

[0025] After each task is completed, based on the task execution results and node status feedback, proceed as follows:

[0026] T n (t+1)=T n (t)+αΔT n (t) Update node trust, where T n (t+1) represents the trust level of node n at time t+1, α is the learning rate, and ΔT n (t) is the change in trust level of node n at time t.

[0027] Furthermore, the intelligent selection strategy described in S2 includes:

[0028] Set a trust threshold T threshold ;

[0029] Before each task scheduling, the current trust level T of each node is obtained in real time. n Given the load of CPU, memory, and network resources (t), calculate resource availability and the node resource availability R. n (t);

[0030] Calculate the overall score.n (t)=α1T n (t)+α2R n (t), determines the task priority, where Score n (t) represents the comprehensive score of node n at time t, and α1 and α2 are the weights of the influence of trust and resource load on task scheduling decisions in task scheduling.

[0031] If the trust level T n (t) <T threshold If node resources are nearing saturation, the task rescheduling rule will be triggered.

[0032] In meeting the trust level T n (t)≥T threshold In the task scheduling pool, nodes are selected to execute tasks based on task priority and task type.

[0033] Furthermore, the rescheduling rules in S3 include:

[0034] Real-time monitoring of the trust level and task execution status of each node;

[0035] When the node trust level T n (t) is lower than the preset threshold T threshold If a node fails, that node will be removed from the task scheduling pool.

[0036] The dynamic trust assessment algorithm is executed to reassess the trust level and resource load of the remaining nodes, and the overall score is recalculated. n (t), and select nodes with a trust level greater than the threshold and sufficient resources according to the intelligent selection strategy;

[0037] The affected tasks will be migrated to a new node to continue execution, and the task status will be continuously monitored. If an anomaly occurs again, the rescheduling rule will be triggered again until the task is successfully completed.

[0038] Furthermore, S4 specifically includes:

[0039] By leveraging the decentralized, immutable, and cryptographic protection features of blockchain, key data such as each task allocation, node selection, task completion status, and node behavior monitoring are written into the blockchain as transactions to ensure the integrity, transparency, and traceability of all scheduling-related data, and to provide a reliable data source for dynamic trust assessment.

[0040] Through smart contracts deployed on the blockchain, the smart selection strategy, rescheduling rules and dynamic trust evaluation algorithm of S2 and S3 are automatically executed; the smart contract automatically verifies the task execution results submitted by the node, and dynamically calculates and updates the node's trust value based on the predefined trust degree algorithm and the on-chain historical behavior records, and writes the tamper-proof update result into the blockchain as the core basis for subsequent scheduling.

[0041] When the smart contract detects through real-time monitoring that the node's trust level is below the threshold T, threshold When malicious behavior is detected, a preset emergency mechanism is automatically triggered, including automatic alarm, node isolation, or task reassignment.

[0042] This invention also provides a distributed computing task scheduling security system based on dynamic trust assessment, comprising:

[0043] The trust calculation module is used to collect and preprocess the running status data of each computing node in real time, and dynamically calculate and update the trust level of each node based on the running status data.

[0044] The intelligent selection module is used to select nodes with a trust level not lower than a preset threshold and sufficient resources to perform tasks based on the trust level and node resource load, using an intelligent selection strategy.

[0045] The rescheduling module is used to continuously monitor the node status during task execution. When a node failure or node trust level is lower than the threshold is detected, the task rescheduling rule is triggered to migrate the affected task to the remaining high-trust nodes with sufficient resources to continue execution.

[0046] The blockchain module is used to leverage the decentralized, immutable, and cryptographically protected characteristics of blockchain to write key data such as task allocation, node selection, and task completion status into the blockchain.

[0047] Furthermore, the trust calculation module calculates and updates the trust level of each node, including:

[0048] Collect task execution history, load data, inter-node interaction information, and behavior monitoring data; calculate task execution success rate, node load and performance, inter-node interaction quality, and abnormal node behavior.

[0049] The formula for the success rate of task execution is:

[0050]

[0051] The formula for calculating node load and performance is as follows:

[0052]

[0053] Among them, β1, β2, and β3 are the weights for CPU utilization, memory utilization, and network bandwidth utilization.

[0054] The formula for calculating the interaction quality between nodes is:

[0055]

[0056] The effective data transmission rate is the amount of data successfully transmitted, excluding failed data due to delays or packet loss. The total data transmission rate refers to the amount of all data transmitted, including both successfully transmitted and failed data.

[0057] The formula for calculating abnormal node behavior is as follows:

[0058]

[0059] Where m is the type of abnormal behavior, β 4i It represents the weight of each abnormal behavior, indicating its degree of impact on the node's trust level;

[0060] Using a weighted model T n (t)=w1S n (t)+w2L n (t)+w3C n (t)+w4A n (t) Calculate the node trust level, where w1, w2, w3, w4 are the weights of each evaluation factor, and S n (t) represents the task execution success rate, L n (t) represents node load and performance, C n (t) represents the interaction quality between nodes, A n (t) indicates abnormal node behavior; T n (t) represents the current trust level of the node;

[0061] After each task is completed, based on the task execution results and node status feedback, proceed as follows:

[0062] T n (t+1)=T n (t)+αΔT n (t) Update node trust, where T n (t+1) represents the trust level of node n at time t+1, α is the learning rate, and ΔT n (t) is the change in trust level of node n at time t.

[0063] Furthermore, the intelligent selection strategy in the intelligent selection module includes:

[0064] Set a trust threshold T threshold ;

[0065] Before each task scheduling, the current trust level T of each node is obtained in real time. n Given the load of CPU, memory, and network resources (t), calculate resource availability and the node resource availability R. n (t);

[0066] Calculate the overall score. n (t)=α1T n (t)+α2R n (t), determines the task priority, where Score n (t) represents the comprehensive score of node n at time t, and α1 and α2 are the weights of the influence of trust and resource load on task scheduling decisions in task scheduling.

[0067] If the trust level T n (t) <T threshold If node resources are nearing saturation, the task rescheduling rule will be triggered.

[0068] In meeting the trust level T n (t)≥T threshold In the task scheduling pool, nodes are selected to execute tasks based on task priority and task type.

[0069] Furthermore, the rescheduling rules in the rescheduling module include:

[0070] Real-time monitoring of the trust level and task execution status of each node;

[0071] When the node trust level T n (t) is lower than the preset threshold T threshold If a node fails, that node will be removed from the task scheduling pool.

[0072] The dynamic trust assessment algorithm is executed to reassess the trust level and resource load of the remaining nodes, and the overall score is recalculated. n (t), and select nodes with a trust level greater than the threshold and sufficient resources according to the intelligent selection strategy;

[0073] The affected tasks will be migrated to a new node to continue execution, and the task status will be continuously monitored. If an anomaly occurs again, the rescheduling rule will be triggered again until the task is successfully completed.

[0074] Furthermore, the blockchain module specifically includes:

[0075] By leveraging the decentralized, immutable, and cryptographic protection features of blockchain, key data such as each task allocation, node selection, task completion status, and node behavior monitoring are written into the blockchain as transactions to ensure the integrity, transparency, and traceability of all scheduling-related data, and to provide a reliable data source for dynamic trust assessment.

[0076] Through smart contracts deployed on the blockchain, the smart selection strategy, rescheduling rules and dynamic trust evaluation algorithm of the smart selection module and rescheduling module are automatically executed; the smart contract automatically verifies the task execution results submitted by the node, and dynamically calculates and updates the node's trust value based on the predefined trust degree algorithm and the on-chain historical behavior records, and writes the tamper-proof update result into the blockchain as the core basis for subsequent scheduling.

[0077] When the smart contract detects through real-time monitoring that the node's trust level is below the threshold T, threshold When malicious behavior is detected, a preset emergency mechanism is automatically triggered, including automatic alarm, node isolation, or task reassignment.

[0078] Beneficial effects: Traditional distributed computing task scheduling methods rely on static resource allocation or task queue scheduling, lacking real-time assessment of node trust levels. This leads to unstable task allocation when nodes fail, maliciously attacked, or the environment changes, affecting system security and performance. In particular, in dynamic and complex cloud computing environments, the adaptive capabilities are insufficient.

[0079] This invention introduces a dynamic trust assessment mechanism, which analyzes the multi-dimensional performance of nodes in real time, including task history, load, and abnormal behavior, ensuring that tasks are assigned to reliable nodes. This avoids interference from malicious or unstable nodes, improving system security and performance. Dynamically updating node trust levels ensures the timeliness and accuracy of task scheduling decisions. Furthermore, this invention combines blockchain technology with the dynamic trust assessment mechanism to provide enhanced security measures. The decentralized and immutable nature of blockchain ensures transparency and data security during task execution. Blockchain protects task execution data through encryption, ensuring that data is not accessed or tampered with without authorization during transmission and storage. All key data during task execution, including task allocation, node selection, and task completion status, are recorded on the blockchain, and encryption algorithms ensure data confidentiality and integrity. This invention also features an adaptive scheduling and dynamic adjustment mechanism, which can flexibly adjust allocation strategies based on node status and task characteristics, avoiding resource overload and node failure, improving task execution efficiency and security, and reducing manual intervention. Attached Figure Description

[0080] Figure 1 This is a flowchart of the method of the present invention.

[0081] Figure 2 This is a flowchart of the task scheduling process of the present invention. Detailed Implementation

[0082] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. After reading the present invention, any modifications of the present invention in various equivalent forms by those skilled in the art will fall within the scope defined by the appended claims.

[0083] Example 1

[0084] Please see Figure 1 , Figure 2 As shown, this invention provides a distributed computing task scheduling security assurance method based on dynamic trust assessment, comprising:

[0085] S1. Collect and preprocess the running status data of each computing node in real time, and dynamically calculate and update the trust level of each node based on the running status data.

[0086] S2. Based on the trust level and combined with the node resource load, an intelligent selection strategy is adopted to select a node with a trust level not lower than a preset threshold and sufficient resources to perform the task.

[0087] S3. During task execution, continuously monitor the node status. When a node failure or node trust level is detected to be lower than the threshold, trigger the task rescheduling rule to migrate the affected task to the remaining high-trust node with sufficient resources for continued execution.

[0088] S4. Utilize the decentralized, immutable, and encrypted protection features of blockchain to write key data such as task allocation, node selection, and task completion status into the blockchain to ensure the transparency and data security of the task execution process.

[0089] In this embodiment, S1 proposes a dynamic trust evaluation mechanism to optimize task scheduling and ensure node security in the current distributed computing environment. By dynamically evaluating data from multiple dimensions, including real-time behavior of computing nodes, task execution history, load conditions, inter-node interactions, and behavior monitoring, it ensures that tasks are assigned to high-trust, reliable nodes, avoiding the impact of malicious and unstable nodes on the system. This mechanism, by comprehensively considering changes in various factors, can flexibly respond to node failures, performance fluctuations, and security threats in the distributed computing environment, ensuring the efficiency and security of task scheduling. The implementation of the dynamic trust evaluation mechanism includes the following key steps:

[0090] S11. Data Acquisition and Preprocessing

[0091] Data acquisition and preprocessing are fundamental steps in the dynamic trust assessment mechanism, ensuring the accuracy and reliability of trust level calculations. First, the system collects task execution history, load data, inter-node interaction information, and behavior monitoring data from each computing node in real time, providing a comprehensive foundation for node performance evaluation. To ensure data consistency and high quality, the system cleans, denoises, and standardizes the collected data, providing stable and high-quality input for subsequent trust level calculations and ensuring the reliability of subsequent steps.

[0092] Data cleaning and standardization are crucial steps in ensuring data consistency. The system employs methods such as moving averages to remove outliers and noise, and fills in missing data (e.g., using the mean or median). Simultaneously, data standardization transforms the data into a zero-mean, unit-variance form to avoid the impact of dimensional differences between features on subsequent analysis. These preprocessing steps ensure high-quality input data, providing accurate data support for trust level calculations.

[0093] S12, Trust Calculation

[0094] Based on data acquisition and preprocessing, this invention uses a trust level calculation formula to dynamically evaluate the current trust level of nodes in conjunction with real-time data. The trust level T of each node... n (t) reflects the reliability and security of the node in the current task scheduling. It is an indispensable part of the task scheduling decision and directly affects the effectiveness of task allocation and the overall system performance.

[0095] The dynamic trust assessment mechanism of this invention calculates trust levels based on multiple dimensions of node data, including: task execution success rate, resource load, node interaction quality, and behavior monitoring. Data from each dimension helps evaluate the node's performance in actual task execution and collectively influences the final trust level.

[0096] Trust level T of a node n (t) is determined by the task execution success rate S n (t), Node Load and Performance L n (t), Inter-node interaction quality C n (t) and node behavior anomaly A n (t) Comprehensive decision. Task execution success rate S n (t) The reliability of nodes is assessed by monitoring their success rate in historical tasks. Nodes with higher success rates have higher trust levels, while nodes with frequent failures will have lower trust levels. The formula is as follows:

[0097]

[0098] Node load and performance L n(t) represents the current resource utilization of the node, including the usage of CPU, memory, and network resources. Excessive load may cause task processing delays or failures, thereby affecting the trust level of the node. The formula is:

[0099]

[0100] Among them, β1, β2, and β3 are the weights of CPU utilization, memory utilization, and network bandwidth utilization; the maximum carrying capacity of a node is set based on the maximum values ​​of CPU, memory, and network bandwidth.

[0101] Inter-node interaction quality C n (t) represents the communication quality between nodes, including data transmission latency and bandwidth utilization. If frequent delays or packet loss occur in communication between nodes, the trust level will be negatively affected. The formula is:

[0102]

[0103] The effective data transmission rate is the amount of data successfully transmitted, excluding failed data due to delays or packet loss. The total data transmission rate refers to the amount of all data transmitted, including both successfully transmitted and failed data.

[0104] Node Behavior Anomaly A n (t) The behavior monitoring system detects whether nodes exhibit malicious behavior or abnormal operations, such as data tampering or resource abuse. If a node exhibits abnormal behavior, its trust level will significantly decrease. The formula is:

[0105]

[0106] Where m is the type of abnormal behavior, β 4i It represents the weight of each abnormal behavior, indicating its impact on the node's trust level. Abnormal behaviors can include data tampering, resource abuse, etc., and each behavior has a different weight.

[0107] The final trust level of a node is calculated by weighting the above evaluation factors, as shown in the following formula:

[0108] T n (t)=w1S n (t)+w2L n (t)+w3C n (t)+w4A n (t)

[0109] Among them, w1, w2, w3, and w4 are the weights of each evaluation factor, which are adjusted according to the actual application scenario and the importance of the task. n (t),L n(t),C n (t),A n (t) represent the task success rate, load, interaction quality, and abnormal behavior value of node n at time t, respectively.

[0110] S13, Trust Update

[0111] Trust level updates directly impact the security and efficiency of task scheduling. After each task execution, the system dynamically adjusts the node's trust level based on the task execution results and system feedback. If a node performs well in task execution, completing the task quickly and without anomalies, the system increases the node's trust level. Conversely, if a node fails or encounters an anomaly, the system decreases the node's trust level to reflect changes in the node's reliability under the current environment.

[0112] To ensure tasks are executed on high-trust nodes, the system dynamically adjusts the task allocation strategy based on the node's current trust level before each task scheduling. This strategy is based on node behavior, historical task execution results, and real-time feedback, ensuring tasks are always assigned to the most reliable and secure nodes. If a node's trust level falls below a set threshold T... threshold If a node fails to meet the trust requirement, it will be temporarily excluded from task allocation, and the task will be reassigned to a node with higher trust. In this way, the system ensures the stability and security of task execution, reducing the risk of task failure due to unreliable nodes. The trust update formula is as follows:

[0113] T n (t+1)=T n (t)+αΔT n (t)T n (t+1) represents the trust level of node n at time t+1. α is the learning rate, controlling the sensitivity of trust level updates. ΔT n (t) is the change in trust level of node n at time t, which depends on the node's task execution results, load status, and behavior monitoring feedback.

[0114] By dynamically updating node trust levels, the system can flexibly respond to changes in node status and ensure that nodes with high trust levels are prioritized for task execution during task scheduling. This mechanism not only improves the system's execution efficiency but also effectively enhances task execution security, avoiding interference from malicious or unstable nodes and ensuring the efficient and secure completion of tasks.

[0115] S2 introduces a task scheduling system. Before each task scheduling, the system employs an intelligent selection strategy. By monitoring and evaluating the trustworthiness and current resource load of each computing node in real time, it flexibly selects the execution node, avoiding computational failures or delays caused by unreliable nodes, thereby improving task execution efficiency and overall system performance. The intelligent selection strategy includes:

[0116] S21. Set the trust threshold T threshold .

[0117] S22. Before each task scheduling, obtain the current trust level T of each node in real time. n Given the load of CPU, memory, and network resources (t), calculate resource availability and the node resource availability R. n (t), and will simultaneously satisfy the trust level T. n (t)≥T threshold Nodes with sufficient but unsaturated resources are placed in the task scheduling pool;

[0118] S23. Calculate the overall score for the nodes in the task scheduling pool. n (t)=α1T n (t)+α2R n (t), where Score n (t) represents the comprehensive score of node n at time t, and α1 and α2 are the weights of trust and resource load on task scheduling decisions. Using this formula, the system performs a preliminary sorting of all nodes to confirm their priority.

[0119] S24. Based on the priority determined by the comprehensive score ranking, the system further selects the most suitable node for task execution from the top-ranked nodes based on the node's historical performance, current resource status, task type, and mutual influence between nodes. For nodes with low load and high trust level, the system will prioritize them to assign high-priority tasks to improve execution efficiency and ensure the stable completion of tasks. Among the top-ranked nodes, the system selects the most suitable node based on the task type.

[0120] For computationally intensive tasks, select the node with the highest CPU idle rate; for latency-sensitive tasks, select the node with the highest interaction quality; for ordinary tasks, select the node with the highest overall score. If multiple candidate nodes appear according to the above rules, select the node with the higher historical task success rate.

[0121] S25. Dynamic monitoring and rescheduling trigger: During task execution, the system monitors the trust level and resource load status of allocated nodes in real time; if the node trust level is found to drop below the preset threshold T... thresholdIf resources are overloaded, the rescheduling rules described in S3 will be triggered to provide an anomaly signal for subsequent task migration, ensuring that the system can respond to runtime faults in a timely manner and guarantee the continuity of tasks and the reliability of execution results.

[0122] For tasks requiring collaboration among multiple nodes, the system allocates tasks based on the nodes' collaborative capabilities (such as communication quality and data synchronization capabilities). Nodes with strong collaborative capabilities are prioritized for task execution, reducing task interruptions caused by collaboration failures. In this way, the system optimizes task scheduling in a distributed computing environment, improving computational performance and task processing stability.

[0123] The task scheduling system of this invention possesses strong adaptive capabilities, dynamically adjusting scheduling strategies based on the nature of different tasks, the real-time status of nodes, and changes in the computing environment. For computationally intensive tasks, the system prioritizes nodes with strong computing power and low load; for latency-sensitive tasks, the system selects nodes with fast response speed and low latency. This adaptive scheduling mechanism enables the system to optimize task allocation in complex and dynamic distributed environments, ensuring efficient and secure task execution.

[0124] In S3, to ensure the continuous and stable operation of the task scheduling system in a distributed computing environment, this invention introduces node fault detection and task rescheduling rules. The system can monitor the health status of nodes in real time and automatically reschedule tasks when a node fails, experiences performance abnormalities, or is attacked, transferring the task to a high-trust node for execution, thus ensuring task continuity and system stability. The rescheduling rules include:

[0125] S31. Monitor the trust level and task execution status of each node in real time;

[0126] S32. Implement the node fault detection mechanism. When the node trust level T... n (t) is lower than the preset threshold T threshold If the response time is abnormal, the node will be marked as an unreliable node, removed from the task scheduling pool, and task rescheduling will be triggered to prevent task execution from being affected.

[0127] S33. Dynamic trust assessment algorithm: reassesses the trust level and resource load of the remaining nodes, and recalculates the comprehensive score. n (t), and selects nodes with a trust level greater than the threshold and sufficient resources based on the intelligent selection strategy; if the affected task is computationally intensive, the system will select the node with the strongest CPU among the high-scoring nodes. If it is latency-sensitive, it will select the node with the best interaction quality;

[0128] S34. Migrate the affected task to a new node to continue execution, and continuously monitor the task status. If an anomaly occurs again, trigger rescheduling again until the task is successfully completed.

[0129] The task rescheduling mechanism of this invention supports adaptive scheduling and dynamic resource adjustment. The system automatically selects the most suitable node for processing based on task type (e.g., computationally intensive or latency-sensitive), avoiding assigning tasks to overloaded nodes to prevent task execution delays or failures due to resource bottlenecks. When node resources are nearing saturation, the system can automatically transfer tasks to less loaded nodes, thereby ensuring task efficiency and stability.

[0130] In S4, to enhance the security of the task scheduling system, this invention combines blockchain technology and a dynamic trust assessment mechanism to address the shortcomings of current technologies in the face of dynamic node changes and network attacks. Although existing blockchain technology can provide tamper-proof capabilities and a certain level of security during task execution, existing security measures often lack real-time performance and adaptability when facing node state changes, node failures, or malicious attacks in a distributed computing environment, making it difficult to effectively cope with complex and ever-changing environments.

[0131] While blockchain's consensus mechanism ensures the reliability of task execution to a certain extent, its real-time assessment of node dynamic trust levels and automated protection mechanisms remain insufficient. Therefore, this invention introduces a dynamic trust assessment mechanism. This mechanism not only leverages the immutability and transparency of blockchain to safeguard data security but also dynamically adjusts task allocation based on real-time assessment of node trust levels and resource status, preventing malicious or unstable nodes from affecting task execution.

[0132] To address this, the present invention provides an enhanced security measure by combining blockchain technology with a dynamic trust assessment mechanism. This mechanism not only ensures transparency and data security in task execution through the decentralized and immutable nature of blockchain, but also dynamically adjusts task allocation strategies based on the real-time assessment of node trust levels, resource status, and node behavior. This ensures that tasks are always assigned to reliable and resource-sufficient nodes, preventing task failures, delays, or data leaks caused by malicious or unstable nodes.

[0133] Furthermore, the introduction of smart contracts provides automated rule execution and emergency response capabilities for task scheduling. When the system detects abnormal behavior, smart contracts can automatically trigger preset emergency mechanisms, such as automatic alarms, node isolation, or task reassignment, to deal with abnormal situations. This not only improves the system's emergency response capabilities but also ensures that task execution always adheres to security protocols, avoiding security risks caused by node failures or malicious behavior.

[0134] S4 specifically includes:

[0135] By leveraging the decentralized, immutable, and cryptographic protection features of blockchain, key data such as each task allocation, node selection, task completion status, and node behavior monitoring are written into the blockchain as transactions to ensure the integrity, transparency, and traceability of all scheduling-related data, and to provide a reliable data source for dynamic trust assessment.

[0136] Through smart contracts deployed on the blockchain, the smart selection strategy, rescheduling rules and dynamic trust evaluation algorithm of S2 and S3 are automatically executed; the smart contract automatically verifies the task execution results submitted by the node, and dynamically calculates and updates the node's trust value based on the predefined trust degree algorithm and the on-chain historical behavior records, and writes the tamper-proof update result into the blockchain as the core basis for subsequent scheduling.

[0137] When the smart contract detects through real-time monitoring that the node's trust level is below the threshold T, threshold When malicious behavior is detected, a preset emergency mechanism is automatically triggered, including automatic alarm, node isolation, or task reassignment.

[0138] Blockchain technology ensures the security of task execution data during transmission and storage through encryption, preventing unauthorized access or tampering. All critical data during task execution, including task allocation, node selection, and task completion status, is recorded on the blockchain, and encryption algorithms ensure confidentiality and integrity. Through the decentralized and smart contract mechanisms of blockchain, the system can ensure the automatic execution of rules during task scheduling without centralized control, reducing human intervention and improving the efficiency and transparency of task scheduling.

[0139] Example 2

[0140] Please see Figure 1 , Figure 2 As shown in Embodiment 1, this invention provides a distributed computing task scheduling security system based on dynamic trust assessment, comprising:

[0141] The trust calculation module is used to collect and preprocess the running status data of each computing node in real time, and dynamically calculate and update the trust level of each node based on the running status data.

[0142] The intelligent selection module is used to select nodes with a trust level not lower than a preset threshold and sufficient resources to perform tasks based on the trust level and node resource load, using an intelligent selection strategy.

[0143] The rescheduling module is used to continuously monitor the node status during task execution. When a node failure or node trust level is lower than the threshold is detected, the task rescheduling rule is triggered to migrate the affected task to the remaining high-trust nodes with sufficient resources to continue execution.

[0144] The blockchain module is used to leverage the decentralized, immutable, and cryptographic protection features of blockchain to write key data such as task allocation, node selection, and task completion status into the blockchain, thereby ensuring the transparency and data security of the task execution process.

[0145] In this embodiment, the trust calculation module proposes a dynamic trust evaluation mechanism to optimize task scheduling and ensure node security in the current distributed computing environment. By dynamically evaluating data from multiple dimensions, including real-time behavior of computing nodes, task execution history, load conditions, inter-node interactions, and behavior monitoring, it ensures that tasks are assigned to high-trust, reliable nodes, avoiding the impact of malicious and unstable nodes on the system. This mechanism, by comprehensively considering changes in various factors, can flexibly respond to node failures, performance fluctuations, and security threats in the distributed computing environment, ensuring the efficiency and security of task scheduling. The implementation of the dynamic trust evaluation mechanism includes:

[0146] The data acquisition and preprocessing module ensures the accuracy and reliability of trust level calculations. First, the system collects task execution history, load data, inter-node interaction information, and behavior monitoring data from each computing node in real time, providing a comprehensive foundation for node performance evaluation. To ensure data consistency and high quality, the system cleans, denoises, and standardizes the collected data, providing stable and high-quality input for subsequent trust level calculations and ensuring the reliability of subsequent steps.

[0147] Data cleaning and standardization are crucial steps in ensuring data consistency. The system employs methods such as moving averages to remove outliers and noise, and fills in missing data (e.g., using the mean or median). Simultaneously, data standardization transforms the data into a zero-mean, unit-variance form to avoid the impact of dimensional differences between features on subsequent analysis. These preprocessing steps ensure high-quality input data, providing accurate data support for trust level calculations.

[0148] The trust level calculation module, building upon the data acquisition and preprocessing modules, uses a trust level calculation formula to dynamically evaluate the current trust level of nodes in conjunction with real-time data. The trust level T of each node... n (t) reflects the reliability and security of the node in the current task scheduling. It is an indispensable part of the task scheduling decision and directly affects the effectiveness of task allocation and the overall system performance.

[0149] The dynamic trust assessment mechanism of this invention calculates trust levels based on multiple dimensions of node data, including: task execution success rate, resource load, node interaction quality, and behavior monitoring. Data from each dimension helps evaluate the node's performance in actual task execution and collectively influences the final trust level.

[0150] Trust level T of a node n (t) is determined by the task execution success rate S n (t), Node Load and Performance L n (t), Inter-node interaction quality C n (t) and node behavior anomaly A n (t) Comprehensive decision. Task execution success rate S n (t) The reliability of nodes is assessed by monitoring their success rate in historical tasks. Nodes with higher success rates have higher trust levels, while nodes with frequent failures will have lower trust levels. The formula is as follows:

[0151]

[0152] Node load and performance L n (t) represents the current resource utilization of the node, including the usage of CPU, memory, and network resources. Excessive load may cause task processing delays or failures, thereby affecting the trust level of the node. The formula is:

[0153]

[0154] Among them, β1, β2, and β3 are the weights of CPU utilization, memory utilization, and network bandwidth utilization; the maximum carrying capacity of a node is set based on the maximum values ​​of CPU, memory, and network bandwidth.

[0155] Inter-node interaction quality C n (t) represents the communication quality between nodes, including data transmission latency and bandwidth utilization. If frequent delays or packet loss occur in communication between nodes, the trust level will be negatively affected. The formula is:

[0156]

[0157] The effective data transmission rate is the amount of data successfully transmitted, excluding failed data due to delays or packet loss. The total data transmission rate refers to the amount of all data transmitted, including both successfully transmitted and failed data.

[0158] Node Behavior Anomaly A n (t) The behavior monitoring system detects whether nodes exhibit malicious behavior or abnormal operations, such as data tampering or resource abuse. If a node exhibits abnormal behavior, its trust level will significantly decrease. The formula is:

[0159]

[0160] Where m is the type of abnormal behavior, β 4i It represents the weight of each abnormal behavior, indicating its impact on the node's trust level. Abnormal behaviors can include data tampering, resource abuse, etc., and each behavior has a different weight.

[0161] The final trust level of a node is calculated by weighting the above evaluation factors, as shown in the following formula:

[0162] T n (t)=w1S n (t)+w2L n (t)+w3C n (t)+w4A n (t)

[0163] Among them, w1, w2, w3, and w4 are the weights of each evaluation factor, which are adjusted according to the actual application scenario and the importance of the task. n (t),L n (t),C n (t),A n (t) represent the task success rate, load, interaction quality, and abnormal behavior value of node n at time t, respectively.

[0164] The trust level update module is used to update the trust level. Trust level updates directly affect the security and efficiency of task scheduling. After each task execution, the system dynamically adjusts the node's trust level based on the node's task execution results and system feedback. If a node performs well in task execution, completing the task quickly and without anomalies, the system will increase the node's trust level. Conversely, if a node fails or an anomaly occurs, the system will decrease the node's trust level to reflect changes in the node's reliability under the current environment.

[0165] To ensure tasks are executed on high-trust nodes, the system dynamically adjusts the task allocation strategy based on the node's current trust level before each task scheduling. This strategy is based on node behavior, historical task execution results, and real-time feedback, ensuring tasks are always assigned to the most reliable and secure nodes. If a node's trust level falls below a set threshold T... threshold If a node fails to meet the trust requirement, it will be temporarily excluded from task allocation, and the task will be reassigned to a node with higher trust. In this way, the system ensures the stability and security of task execution, reducing the risk of task failure due to unreliable nodes. The trust update formula is as follows:

[0166] T n (t+1)=T n(t)+αΔT n (t)T n (t+1) represents the trust level of node n at time t+1. α is the learning rate, controlling the sensitivity of trust level updates. ΔT n (t) is the change in trust level of node n at time t, which depends on the node's task execution results, load status, and behavior monitoring feedback.

[0167] By dynamically updating node trust levels, the system can flexibly respond to changes in node status and ensure that nodes with high trust levels are prioritized for task execution during task scheduling. This mechanism not only improves the system's execution efficiency but also effectively enhances task execution security, avoiding interference from malicious or unstable nodes and ensuring the efficient and secure completion of tasks.

[0168] The intelligent selection module incorporates a task scheduling system. Before each task scheduling, the system employs an intelligent selection strategy. By monitoring and evaluating the trust level and current resource load of each computing node in real time, it flexibly selects the execution node, avoiding computational failures or delays caused by unreliable nodes, thereby improving task execution efficiency and overall system performance. The intelligent selection strategy includes:

[0169] The threshold setting module is used to set the trust threshold T. threshold .

[0170] The calculation module is used to obtain the current trust level T of each node in real time before each task scheduling. n Given the load of CPU, memory, and network resources (t), calculate resource availability and the node resource availability R. n (t), and will simultaneously satisfy the trust level T. n (t)≥T threshold Nodes with sufficient but unsaturated resources are placed in the task scheduling pool;

[0171] The priority sorting module is used to calculate a comprehensive score for nodes in the task scheduling pool. n (t)=α1T n (t)+α2R n (t), where Score n (t) represents the comprehensive score of node n at time t, and α1 and α2 are the weights of trust and resource load on task scheduling decisions. Using this formula, the system performs a preliminary sorting of all nodes to confirm their priority.

[0172] The node selection module is used to intelligently select the most suitable node for task execution from the top-ranked nodes based on the priority determined by the comprehensive score ranking, and further considering the node's historical performance, current resource status, task type, and mutual influence between nodes. For nodes with low load and high trust level, the system will prioritize them to assign high-priority tasks to improve execution efficiency and ensure stable task completion; among the top-ranked nodes, the most suitable node is selected based on the task type.

[0173] For computationally intensive tasks, select the node with the highest CPU idle rate; for latency-sensitive tasks, select the node with the highest interaction quality; for ordinary tasks, select the node with the highest overall score. If multiple candidate nodes appear according to the above rules, select the node with the higher historical task success rate.

[0174] The dynamic monitoring and triggering module monitors the trust level and resource load status of assigned nodes in real time during task execution. Once this module detects any abnormal situation such as a node's trust level falling below a preset threshold or resource utilization overload, it immediately marks the node as abnormal and triggers a rescheduling rule. The rescheduling module then migrates the task to a node with high trust level and sufficient resources for continued execution. This module ensures that the system can promptly avoid the risks of task failure, delays, or data leakage caused by unreliable nodes, performance degradation, or resource saturation, thereby guaranteeing the continuity, security, and reliability of task scheduling.

[0175] For tasks requiring collaboration among multiple nodes, the system allocates tasks based on the nodes' collaborative capabilities (such as communication quality and data synchronization capabilities). Nodes with strong collaborative capabilities are prioritized for task execution, reducing task interruptions caused by collaboration failures. In this way, the system optimizes task scheduling in a distributed computing environment, improving computational performance and task processing stability.

[0176] The task scheduling system of this invention possesses strong adaptive capabilities, dynamically adjusting scheduling strategies based on the nature of different tasks, the real-time status of nodes, and changes in the computing environment. For computationally intensive tasks, the system prioritizes nodes with strong computing power and low load; for latency-sensitive tasks, the system selects nodes with fast response speed and low latency. This adaptive scheduling mechanism enables the system to optimize task allocation in complex and dynamic distributed environments, ensuring efficient and secure task execution.

[0177] In the rescheduling module, to ensure the continuous and stable operation of the task scheduling system in a distributed computing environment, this invention introduces node fault detection and task rescheduling rules. The system can monitor the health status of nodes in real time and automatically reschedule tasks when a node fails, experiences performance abnormalities, or is attacked, transferring the task to a high-trust node for execution, thus ensuring task continuity and system stability. The rescheduling rules include:

[0178] The real-time monitoring module is used to monitor the trust level and task execution status of each node in real time.

[0179] The fault detection module is used to execute the node fault detection mechanism. When the node trust level T... n (t) is lower than the preset threshold T threshold If the response time is abnormal, the node will be marked as an unreliable node, removed from the task scheduling pool, and task rescheduling will be triggered to prevent task execution from being affected.

[0180] The dynamic trust assessment module is used to reassess the trust level and resource load of the remaining nodes using a dynamic trust assessment algorithm, and recalculate the overall score. n (t), and selects nodes with a trust level greater than the threshold and sufficient resources based on the intelligent selection strategy; if the affected task is computationally intensive, the system will select the node with the strongest CPU among the high-scoring nodes. If it is latency-sensitive, it will select the node with the best interaction quality;

[0181] The continuous monitoring module is used to migrate affected tasks to a new node for continued execution and continuously monitor the task status. If an anomaly occurs again, rescheduling will be triggered again until the task is successfully completed.

[0182] The task rescheduling mechanism of this invention supports adaptive scheduling and dynamic resource adjustment. The system automatically selects the most suitable node for processing based on task type (e.g., computationally intensive or latency-sensitive), avoiding assigning tasks to overloaded nodes to prevent task execution delays or failures due to resource bottlenecks. When node resources are nearing saturation, the system can automatically transfer tasks to less loaded nodes, thereby ensuring task efficiency and stability.

[0183] In the blockchain module, to enhance the security of the task scheduling system, this invention combines blockchain technology with a dynamic trust assessment mechanism to address the shortcomings of current technologies in the face of dynamic node changes and network attacks. Although existing blockchain technology can provide tamper-proof capabilities and a certain level of security during task execution, existing security measures often lack real-time performance and adaptability when facing node state changes, node failures, or malicious attacks in a distributed computing environment, making it difficult to effectively cope with complex and ever-changing environments.

[0184] While blockchain's consensus mechanism ensures the reliability of task execution to a certain extent, its real-time assessment of node dynamic trust levels and automated protection mechanisms remain insufficient. Therefore, this invention introduces a dynamic trust assessment mechanism. This mechanism not only leverages the immutability and transparency of blockchain to safeguard data security but also dynamically adjusts task allocation based on real-time assessment of node trust levels and resource status, preventing malicious or unstable nodes from affecting task execution.

[0185] To address this, the present invention provides an enhanced security measure by combining blockchain technology with a dynamic trust assessment mechanism. This mechanism not only ensures transparency and data security in task execution through the decentralized and immutable nature of blockchain, but also dynamically adjusts task allocation strategies based on the real-time assessment of node trust levels, resource status, and node behavior. This ensures that tasks are always assigned to reliable and resource-sufficient nodes, preventing task failures, delays, or data leaks caused by malicious or unstable nodes.

[0186] Furthermore, the introduction of smart contracts provides automated rule execution and emergency response capabilities for task scheduling. When the system detects abnormal behavior, smart contracts can automatically trigger preset emergency mechanisms, such as automatic alarms, node isolation, or task reassignment, to deal with abnormal situations. This not only improves the system's emergency response capabilities but also ensures that task execution always adheres to security protocols, avoiding security risks caused by node failures or malicious behavior.

[0187] The blockchain module includes: leveraging the decentralized, immutable, and cryptographic protection characteristics of blockchain, writing key data such as each task allocation, node selection, task completion status, and node behavior monitoring as transactions into the blockchain to ensure the integrity, transparency, and traceability of all scheduling-related data, and to provide a reliable data source for dynamic trust assessment;

[0188] Through smart contracts deployed on the blockchain, the smart selection strategy, rescheduling rules and dynamic trust evaluation algorithm of the smart selection module and rescheduling module are automatically executed; the smart contract automatically verifies the task execution results submitted by the node, and dynamically calculates and updates the node's trust value based on the predefined trust degree algorithm and the on-chain historical behavior records, and writes the tamper-proof update result into the blockchain as the core basis for subsequent scheduling.

[0189] When the smart contract detects through real-time monitoring that the node's trust level is below the threshold T, threshold When malicious behavior is detected, a preset emergency mechanism is automatically triggered, including automatic alarm, node isolation, or task reassignment.

[0190] Blockchain technology ensures the security of task execution data during transmission and storage through encryption, preventing unauthorized access or tampering. All critical data during task execution, including task allocation, node selection, and task completion status, is recorded on the blockchain, and encryption algorithms ensure confidentiality and integrity. Through the decentralized and smart contract mechanisms of blockchain, the system can ensure the automatic execution of rules during task scheduling without centralized control, reducing human intervention and improving the efficiency and transparency of task scheduling.

Claims

1. A method for ensuring the security of distributed computing task scheduling based on dynamic trust assessment, characterized in that, include: S1. Collect and preprocess the running status data of each computing node in real time, and dynamically calculate and update the trust level of each node based on the running status data. S2. Based on the trust level and combined with the node resource load, an intelligent selection strategy is adopted to select a node with a trust level not lower than a preset threshold and sufficient resources to perform the task. S3. During task execution, continuously monitor the node status. When a node failure or node trust level is detected to be lower than the threshold, trigger the task rescheduling rule to migrate the affected task to the remaining high-trust node with sufficient resources for continued execution. S4. Utilize the decentralized, immutable, and cryptographically protected characteristics of blockchain to write key data such as task allocation, node selection, and task completion status into the blockchain.

2. The distributed computing task scheduling security assurance method based on dynamic trust assessment according to claim 1, characterized in that, The calculation and updating of the trust level of each node as described in S1 includes: Collect task execution history, load data, inter-node interaction information, and behavior monitoring data; calculate task execution success rate, node load and performance, inter-node interaction quality, and abnormal node behavior. The formula for the success rate of task execution is: The formula for calculating node load and performance is as follows: Among them, β1, β2, and β3 are the weights for CPU utilization, memory utilization, and network bandwidth utilization. The formula for calculating the interaction quality between nodes is: The effective data transmission rate is the amount of data successfully transmitted, excluding failed data due to delays or packet loss. The total data transmission rate refers to the amount of all data transmitted, including both successfully transmitted and failed data. The formula for calculating abnormal node behavior is as follows: Where m is the type of abnormal behavior, β 4i It represents the weight of each abnormal behavior, indicating its degree of impact on the node's trust level; Using a weighted model T n (t)=w1S n (t)+w2L n (t)+w3C n (t)+w4A n (t) Calculate the node trust level, where w1, w2, w3, w4 are the weights of each evaluation factor, and S n (t) represents the task execution success rate, L n (t) represents node load and performance, C n (t) represents the interaction quality between nodes, A n (t) indicates abnormal node behavior; T n (t) represents the current trust level of the node; After each task is completed, based on the task execution results and node status feedback, press T. n (t+1)=T n (t)+αΔT n (t) Update node trust, where T n (t+1) represents the trust level of node n at time t+1, α is the learning rate, and ΔT n (t) is the change in trust level of node n at time t.

3. The distributed computing task scheduling security assurance method based on dynamic trust assessment according to claim 1, characterized in that, The intelligent selection strategy described in S2 includes: Set a trust threshold T threshold ; Before each task scheduling, the current trust level T of each node is obtained in real time. n Given the load of CPU, memory, and network resources (t), calculate resource availability and the node resource availability R. n (t); Calculate the overall score. n (t)=α1T n (t)+α2R n (t), determines the task priority, where Score n (t) represents the comprehensive score of node n at time t, and α1 and α2 are the weights of the influence of trust and resource load on task scheduling decisions in task scheduling. If the trust level T n (t) <T threshold If node resources are nearing saturation, the task rescheduling rule will be triggered. In meeting the trust level T n (t)≥T threshold In the task scheduling pool, nodes are selected to execute tasks based on task priority and task type.

4. The distributed computing task scheduling security assurance method based on dynamic trust assessment according to claim 3, characterized in that, S3 rescheduling rules include: Real-time monitoring of the trust level and task execution status of each node; When the node trust level T n (t) is lower than the preset threshold T threshold If a node fails, that node will be removed from the task scheduling pool. The dynamic trust assessment algorithm is executed to reassess the trust level and resource load of the remaining nodes, and the overall score is recalculated. n (t), and select nodes with a trust level greater than the threshold and sufficient resources according to the intelligent selection strategy; The affected tasks will be migrated to a new node to continue execution, and the task status will be continuously monitored. If an anomaly occurs again, the rescheduling rule will be triggered again until the task is successfully completed.

5. The distributed computing task scheduling security assurance method based on dynamic trust assessment according to claim 4, characterized in that, S4 specifically includes: By leveraging the decentralized, immutable, and cryptographic protection features of blockchain, key data such as each task allocation, node selection, task completion status, and node behavior monitoring are written into the blockchain as transactions to ensure the integrity, transparency, and traceability of all scheduling-related data, and to provide a reliable data source for dynamic trust assessment. Through smart contracts deployed on the blockchain, the smart selection strategy, rescheduling rules and dynamic trust evaluation algorithm of S2 and S3 are automatically executed; the smart contract automatically verifies the task execution results submitted by the node, and dynamically calculates and updates the node's trust value based on the predefined trust degree algorithm and the on-chain historical behavior records, and writes the tamper-proof update result into the blockchain as the core basis for subsequent scheduling. When the smart contract detects through real-time monitoring that the node's trust level is below the threshold T, threshold When malicious behavior is detected, a preset emergency mechanism is automatically triggered, including automatic alarm, node isolation, or task reassignment.

6. A distributed computing task scheduling security system based on dynamic trust assessment, characterized in that, include: The trust calculation module is used to collect and preprocess the running status data of each computing node in real time, and dynamically calculate and update the trust level of each node based on the running status data. The intelligent selection module is used to select nodes with a trust level not lower than a preset threshold and sufficient resources to perform tasks based on the trust level and node resource load, using an intelligent selection strategy. The rescheduling module is used to continuously monitor the node status during task execution. When a node failure or node trust level is lower than the threshold is detected, the task rescheduling rule is triggered to migrate the affected task to the remaining high-trust nodes with sufficient resources to continue execution. The blockchain module is used to leverage the decentralized, immutable, and cryptographically protected characteristics of blockchain to write key data such as task allocation, node selection, and task completion status into the blockchain.

7. The distributed computing task scheduling security system based on dynamic trust assessment according to claim 6, characterized in that, The trust calculation module calculates and updates the trust level of each node, including: Collect task execution history, load data, inter-node interaction information, and behavior monitoring data; calculate task execution success rate, node load and performance, inter-node interaction quality, and abnormal node behavior. The formula for the success rate of task execution is: The formula for calculating node load and performance is as follows: Among them, β1, β2, and β3 are the weights for CPU utilization, memory utilization, and network bandwidth utilization. The formula for calculating the interaction quality between nodes is: The effective data transmission rate is the amount of data successfully transmitted, excluding failed data due to delays or packet loss. The total data transmission rate refers to the amount of all data transmitted, including both successfully transmitted and failed data. The formula for calculating abnormal node behavior is as follows: Where m is the type of abnormal behavior, β 4i It represents the weight of each abnormal behavior, indicating its degree of impact on the node's trust level; Using a weighted model T n (t)=w1S n (t)+w2L n (t)+w3C n (t)+w4A n (t) Calculate the node trust level, where w1, w2, w3, w4 are the weights of each evaluation factor, and S n (t) represents the task execution success rate, L n (t) represents node load and performance, C n (t) represents the interaction quality between nodes, A n (t) indicates abnormal node behavior; T n (t) represents the current trust level of the node; After each task is completed, based on the task execution results and node status feedback, press T. n (t+1)=T n (t)+αΔT n (t) Update node trust, where T n (t+1) represents the trust level of node n at time t+1, α is the learning rate, and ΔT n (t) is the change in trust level of node n at time t.

8. The distributed computing task scheduling security system based on dynamic trust assessment according to claim 6, characterized in that, The intelligent selection strategy described in the intelligent selection module includes: Set a trust threshold T threshold ; Before each task scheduling, the current trust level T of each node is obtained in real time. n Given the load of CPU, memory, and network resources (t), calculate resource availability and the node resource availability R. n (t); Calculate the overall score. n (t)=α1T n (t)+α2R n (t), determines the task priority, where Score n (t) represents the comprehensive score of node n at time t, and α1 and α2 are the weights of the influence of trust and resource load on task scheduling decisions in task scheduling. If the trust level T n (t) <T threshold If node resources are nearing saturation, the task rescheduling rule will be triggered. In meeting the trust level T n (t)≥T threshold In the task scheduling pool, nodes are selected to execute tasks based on task priority and task type.

9. The distributed computing task scheduling security system based on dynamic trust assessment according to claim 8, characterized in that, The rescheduling rules in the rescheduling module include: Real-time monitoring of the trust level and task execution status of each node; When the node trust level T n (t) is lower than the preset threshold T threshold If a node fails, that node will be removed from the task scheduling pool. The dynamic trust assessment algorithm is executed to reassess the trust level and resource load of the remaining nodes, and the overall score is recalculated. n (t), and select nodes with a trust level greater than the threshold and sufficient resources according to the intelligent selection strategy; The affected tasks will be migrated to a new node to continue execution, and the task status will be continuously monitored. If an anomaly occurs again, the rescheduling rule will be triggered again until the task is successfully completed.

10. The distributed computing task scheduling security system based on dynamic trust assessment according to claim 9, characterized in that, The blockchain module specifically includes: By leveraging the decentralized, immutable, and cryptographic protection features of blockchain, key data such as each task allocation, node selection, task completion status, and node behavior monitoring are written into the blockchain as transactions to ensure the integrity, transparency, and traceability of all scheduling-related data, and to provide a reliable data source for dynamic trust assessment. Through smart contracts deployed on the blockchain, the smart selection strategy, rescheduling rules and dynamic trust evaluation algorithm of the smart selection module and rescheduling module are automatically executed; the smart contract automatically verifies the task execution results submitted by the node, and dynamically calculates and updates the node's trust value based on the predefined trust degree algorithm and the on-chain historical behavior records, and writes the tamper-proof update result into the blockchain as the core basis for subsequent scheduling. When the smart contract detects through real-time monitoring that the node's trust level is below the threshold T, threshold When malicious behavior is detected, a preset emergency mechanism is automatically triggered, including automatic alarm, node isolation, or task reassignment.

Images