A service migration method and system considering reputation and fault tolerance mechanism

Abstract

The application discloses a service migration method and system considering reputation and fault-tolerant mechanism, and relates to the field of service migration in a mobile edge environment.The method comprises the following steps: obtaining a starting edge node and a target migration node corresponding to a service, and calculating the total reputation score of the edge nodes in the edge node set; screening the edge nodes based on the total reputation score and the edge node reputation threshold, constructing a path network topology graph, obtaining a migration path set by using a depth-first traversal algorithm, and screening a candidate path set; evaluating the resource redundancy performance of each candidate path and screening a candidate reliable path set; implementing a fault-tolerant redundancy strategy on each candidate reliable path, establishing a fault-tolerant feedback adjustment mechanism, and adjusting the fault-tolerant strategy according to the feedback result; calculating the transmission delay of each candidate reliable path, and selecting the candidate reliable path with the highest comprehensive score as the service migration path.The application effectively improves the stability, reliability and efficiency of the service migration process.

Description

Technical Field

[0001] This invention relates to the field of edge computing service migration, and in particular to a service migration method and system that takes into account both reputation and fault tolerance mechanisms. Background Technology

[0002] With the rapid development of edge computing, the demand for service migration in mobile edge environments is increasing. Edge computing improves response speed and bandwidth efficiency while reducing network load by shifting computing tasks from data centers to edge nodes closer to users. However, in dynamic and complex mobile edge environments, efficiently migrating services while ensuring service quality and system reliability remains a significant challenge. Traditional service migration methods typically rely on static resource allocation or predetermined migration strategies, lacking real-time response to dynamic environmental changes. In practical applications, user movement trajectories and edge node resource status are constantly changing, necessitating more flexible and intelligent migration mechanisms. Furthermore, the reliability of service migration is also affected by the reputation of edge nodes, such as their computing power, energy status, and historical performance. How to dynamically optimize the migration process while ensuring service quality, taking into account node reputation, has become a critical issue that urgently needs to be addressed.

[0003] Patent document CN109379739A, entitled "A Trusted Collaborative Service Method for Maritime Wireless Mesh Networks," describes a method that incentivizes node collaboration by introducing a node reputation evaluation mechanism. This constructs a virtual collaborative service system based on the mesh network and trust evaluation to ensure reliable transmission of network tasks. Specifically, it utilizes node reputation, contribution, available bandwidth, computing power, and link quality as characteristic attributes of nodes. Driven by a task, MRC nodes select nodes for collaborative services based on node service capability evaluation and collaborative service migration priorities, thus building a virtual collaborative service system that enables rapid resource integration and provides reliable services to users. This method effectively addresses congestion, single points of failure, and low data distribution efficiency in wireless collaborative communication networks in marine and island environments, improving the service quality of mesh networks. However, the reputation and contribution calculations in this method are statically obtained, leading to inaccuracies. Furthermore, the lack of explicit fault-tolerance strategies, relying heavily on node service capabilities, results in link instability. Summary of the Invention

[0004] The summary section introduces a series of simplified concepts, which will be further explained in detail in the detailed description section. The summary section of this invention is not intended to limit the key features and essential technical features of the claimed technical solution, nor is it intended to determine the scope of protection of the claimed technical solution.

[0005] To address the technical problems of existing service migration methods that do not fully consider the reputation of edge nodes, leading to unstable migration paths and a lack of effective fault tolerance mechanisms to cope with node failures and dynamic environmental changes, this invention provides a service migration method and system that takes into account both reputation and fault tolerance mechanisms. By introducing a multi-dimensional reputation evaluation model to screen reliable nodes, and combining horizontal and vertical redundancy strategies and dynamic feedback adjustment mechanisms, efficient, stable and reliable service migration is achieved.

[0006] Firstly, a service migration method that balances reputation and fault tolerance is provided, including:

[0007] Obtain the starting edge node and target migration node corresponding to the service used by the user, and calculate the total reputation score of the edge nodes in the edge node set;

[0008] Set an edge node reputation threshold, filter the edge nodes based on the total reputation score, construct a migration path set, and filter out a candidate path set;

[0009] Evaluate the resource redundancy performance of each candidate path in the candidate path set, and select candidate reliable paths to form a candidate reliable path set;

[0010] A fault-tolerant redundancy strategy is implemented on the candidate reliable paths in the candidate reliable path set to maintain service stability, a fault-tolerant feedback adjustment mechanism is established, and the fault-tolerant strategy is adjusted according to the feedback results.

[0011] Calculate the transmission delay of each candidate reliable path in the candidate reliable path set, and select the candidate reliable path with the highest comprehensive score as the service migration path.

[0012] Furthermore, the calculation of the total reputation score of the edge nodes in the edge node set refers to: calculating the total reputation score by weighting the interaction reputation, energy reputation, and recommendation reputation of each edge node;

[0013] The set of edge nodes refers to the set of all edge nodes within the region from the starting edge node to the target migration node, expressed as:

[0014] ;

[0015] in, Represents the set of edge nodes. Indicates the starting edge node, Indicates the target migration node. This indicates the number of edge nodes in the edge node set.

[0016] Furthermore, the aforementioned interactive reputation refers to a quantitative indicator constructed based on the number of successful and failed task deliveries in the history of edge nodes, calculated using the following formula:

[0017] ;

[0018] in, Represents edge nodes Interaction reputation value, , These represent the certainty factor and uncertainty factor of the interactive reputation, respectively. Represents edge nodes Number of successful historical mission deliveries Represents edge nodes Number of historical mission delivery failures;

[0019] The energy reputation refers to a quantitative indicator built upon the energy consumed by edge nodes in executing tasks, and its calculation formula is as follows:

[0020] ;

[0021] in, Represents edge nodes exist Energy reputation value at any moment Represents edge nodes exist Energy reputation value at any moment This represents the memory factor, used to control the historical energy reputation weight. Represents edge nodes exist The confidence level of anomalies at any given moment;

[0022] The anomaly confidence level is calculated using the following formula:

[0023] ;

[0024] in, Represents edge nodes exist The confidence level of anomalies at any given moment. Indicates the tolerance coefficient. Represents edge nodes exist Power deviation at any given time;

[0025] The power deviation is calculated using the following formula:

[0026] ;

[0027] in, Represents edge nodes exist Power deviation at time, Represents edge nodes exist Instantaneous power at a given moment Represents edge nodes exist Normal reference power at time, Represents edge nodes exist Standard deviation of power fluctuation at time t. Indicates the smoothing factor;

[0028] The recommended reputation refers to the reputation of other edge nodes towards this edge node, calculated using the following formula:

[0029] , , ;

[0030] in, Represents edge nodes Recommended reputation value, Represents the set of edge nodes. Represents edge nodes For edge nodes Recommendation reputation score Represents edge nodes Recommendation factor, Represents the set of edge nodes The number of edge nodes;

[0031] The total reputation score is calculated using the following formula:

[0032] ;

[0033] in, Represents edge nodes Total reputation score These represent the weight values ​​of interaction reputation, energy reputation, and recommendation reputation, respectively. Representing edge nodes respectively The interaction reputation value, energy reputation value, and recommendation reputation value.

[0034] Further, the step of setting an edge node reputation threshold, filtering edge nodes based on the total reputation score, constructing a migration path set, and filtering out a candidate path set includes:

[0035] Set an edge node reputation threshold to filter out edge nodes whose total reputation score is not lower than the edge node reputation threshold;

[0036] A path network topology graph is constructed based on the edge nodes retained after filtering. A depth-first traversal algorithm is used to obtain all migration paths from the starting edge node to the target migration node, forming a migration path set. The path quality of each migration path in the migration path set is evaluated and sorted in descending order. The top 30% of the migration paths are selected as candidate paths, forming a candidate path set.

[0037] Furthermore, the formula for selecting edge nodes whose total reputation score is not lower than the edge node reputation threshold is as follows:

[0038] , ;

[0039] in, This represents the set of reputation edge nodes, consisting of edge nodes whose total reputation score is not lower than the reputation threshold of edge nodes. Represents the set of edge nodes. Represents edge nodes Total reputation score This represents the reputation threshold of the edge node. This represents the average reputation of all edge nodes. This represents the standard deviation of the reputation of all edge nodes. This represents the sensitivity coefficient.

[0040] Furthermore, the construction of the path network topology graph based on the retained edge nodes after filtering includes:

[0041] Based on the connection relationships of the edge nodes retained after filtering, the edge nodes are mapped to vertices of the path network topology graph, and the physical links between edge nodes are mapped to edges of the path network topology graph, thereby constructing a complete path network topology graph.

[0042] The path network topology diagram is expressed as follows:

[0043] ;

[0044] in, Represents the path network topology graph. Represents the set of reputation edge nodes. Represents the set of edges in a graph;

[0045] The graph edge set is expressed as:

[0046] ;

[0047] in, Represents the set of graph edges. Represents edge nodes The logical connection between them This represents the set of reputation edge nodes.

[0048] Furthermore, the depth-first traversal algorithm is used to obtain all migration paths from the starting edge node to the target migration node, forming a migration path set. The path quality of each migration path in the migration path set is evaluated and sorted in descending order. The top 30% of the migration paths are selected as candidate paths, forming a candidate path set, including:

[0049] S11: Initialize the migration path set The expression is: ,in Represents the empty set;

[0050] S12: Using a depth-first traversal algorithm, obtain the migration path from the starting edge node to the target migration node. , migration path Add to migration path set The expression is: ;

[0051] S13: Calculate the path quality score for each migration path in the migration path set;

[0052] S14: Sort the migration paths in the migration path set in descending order based on the path quality score, and select the top 30% of the migration paths as candidate paths to form a candidate path set.

[0053] The path quality score is calculated using the following formula:

[0054] ;

[0055] in, Indicates migration path Path quality score, This represents the total reputation value of the migration path. This indicates the number of hops in the migration path. Indicates migration path Minimum bandwidth, These represent the weighting coefficients for total reputation score, hop count, and minimum bandwidth, respectively. ;

[0056] The total reputation value of the migration path refers to the sum of the total reputation scores of all edge nodes on the migration path, calculated using the following formula:

[0057] ;

[0058] in, This represents the total reputation value of the migration path. Represents edge nodes Total reputation score Indicates the number of edge nodes in the migration path;

[0059] The candidate path set is expressed as:

[0060] ;

[0061] in, Represents the set of candidate paths. Indicates the first Candidate paths, This represents the total number of candidate paths in the candidate path set.

[0062] Further, evaluating the resource redundancy performance of each candidate path in the candidate path set includes:

[0063] Calculate the remaining computing resources of each edge node in the candidate path, and evaluate the horizontal redundancy, vertical redundancy and resource redundancy performance of each candidate path.

[0064] The remaining computing resources are calculated using the following formula:

[0065] ;

[0066] in, Represents edge nodes The remaining computing resources Representing edge nodes respectively The current available CPU, memory, and bandwidth. Representing edge nodes respectively The total amount of CPU, memory, and bandwidth resources. Indicates the weighting coefficient;

[0067] The aforementioned horizontal redundancy capability refers to the number of nodes that create backup tasks in the candidate paths, calculated using the following formula:

[0068] ;

[0069] ;

[0070] in, Indicate candidate path Horizontal redundancy capability Indicate candidate path The number of edge nodes that can be used as backups Indicate candidate path The number of edge nodes in the data. As a constraint, it means that the remaining computing resources of the edge node must be greater than or equal to the minimum resources required for the backup task multiplied by the redundancy factor. This indicates the minimum computing resource requirements for the backup task. Indicates the redundancy coefficient;

[0071] The vertical redundancy capability refers to: calculating the vertical expansion potential of each key edge node in the candidate path, and determining the average vertical expansion potential of all key edge nodes in the candidate path as the vertical redundancy capability of the candidate path. The calculation formula is as follows:

[0072] ;

[0073] ;

[0074] in, Indicate candidate path Longitudinal redundancy capability Indicates the number of critical edge nodes. Represents the set of key edge nodes. Represents key edge nodes Vertical expansion potential Represents key edge nodes Current resource usage Represents key edge nodes Maximum resource that can be dynamically expanded;

[0075] The key edge nodes include: the edge node with the lowest total reputation score in the candidate path, the edge node with a node load exceeding 80%, and the edge node directly adjacent to the starting edge node or the target migration node.

[0076] The resource redundancy performance is calculated using the following formula:

[0077] ;

[0078] in, Indicate candidate path Resource redundancy performance, Indicate candidate path Horizontal redundancy capability Indicate candidate path Longitudinal redundancy capability Indicate candidate path Path quality score, Indicates the weighting coefficient;

[0079] The selected candidate reliable paths constitute a candidate reliable path set, including:

[0080] Based on the resource redundancy performance of the candidate paths, the candidate paths are sorted in descending order, and the top 30% of the candidate paths are selected as candidate reliable paths to form a set of candidate reliable paths.

[0081] The set of candidate reliable paths is expressed as:

[0082] ;

[0083] in, Represents the set of candidate reliable paths. This represents the candidate reliable path with the highest resource redundancy performance in the candidate reliable path set. Indicates the resource redundancy performance of the candidate reliable path set. High candidate reliable path, This indicates the number of candidate reliable paths in the candidate reliable path set.

[0084] Furthermore, the implementation of a fault-tolerant redundancy strategy on the candidate reliable paths in the candidate reliable path set includes: a horizontal redundancy strategy and a vertical redundancy strategy;

[0085] The aforementioned horizontal redundancy strategy refers to:

[0086] The remaining computing resources are selected from the candidate reliable paths to satisfy Constraints are imposed such that the total reputation score is greater than the reputation threshold of the edge node, and the edge node with the fewest hops from the starting edge node is deployed with a replica to ensure the state synchronization of the replica, thereby achieving single-node fault tolerance.

[0087] The vertical redundancy strategy refers to:

[0088] By adjusting task priorities and dynamically scheduling resources, the fault tolerance of the key edge nodes is improved, ensuring that they can maintain service stability under high load or failure.

[0089] The deployed copy refers to:

[0090] To prevent single-node failures from impacting services, task backups are performed on other edge nodes.

[0091] Furthermore, the improvement of the fault tolerance of the critical edge nodes through task priority adjustment and dynamic resource scheduling, ensuring that they can maintain service stability under high load or failure, includes:

[0092] Resource allocation is controlled by priority tags, tasks are divided into critical tasks and ordinary tasks, and task priorities are adjusted and resources are scheduled according to resource preemption rules.

[0093] The key tasks mentioned above refer to: state synchronization task and real-time migration task;

[0094] The common tasks mentioned above refer to: computing tasks and data backup;

[0095] The resource preemption rules include:

[0096] The resources of the key edge nodes are divided into reserved resources, key task resources, and ordinary task resources;

[0097] When available resources are insufficient, terminate the normal task to free up resources for the critical task; when available resources are sufficient, maintain the current task.

[0098] The reserved resources, accounting for 10%, are used to configure resource reservation strategies for the key edge nodes.

[0099] The critical task resources, accounting for 60%, are used to execute critical tasks.

[0100] The resources for ordinary tasks are used to perform ordinary tasks and account for 30%.

[0101] Furthermore, the fault-tolerant feedback adjustment mechanism includes: fault-tolerant positive feedback adjustment and fault-tolerant negative feedback adjustment;

[0102] The aforementioned fault-tolerant positive feedback adjustment refers to:

[0103] If an edge node successfully completes its migration task without experiencing any failures during the task, its overall reputation score will be increased. The calculation formula is as follows:

[0104] ;

[0105] in, express Time edge node Total reputation score express Time edge node Total reputation score This represents the maximum gain coefficient. Indicates the growth rate factor. Represents edge nodes Number of times historical tasks were successfully delivered;

[0106] The fault-tolerant negative feedback adjustment refers to:

[0107] If an edge node fails or malfunctions during a migration task, its usage frequency in future migrations will be reduced, and the node's overall reputation score will be lowered. The calculation formula is as follows:

[0108] ;

[0109] ;

[0110] in, Represents edge nodes Total reputation score express Time edge node Total reputation score express Time edge node Total reputation score Represents the decay rate factor. Represents the residual coefficient. Indicates the lower limit of the benchmark credit rating. Represents the set of edge nodes. Represents the edge set The number of edge nodes.

[0111] Furthermore, the transmission delay is calculated using the following formula:

[0112] ;

[0113] in, Indicates candidate reliable paths Transmission delay, Indicates the edges in the candidate reliable path physical distance, Indicates the speed of signal propagation. Indicates the amount of work. Representing an edge Available bandwidth;

[0114] The comprehensive score is calculated using the following formula:

[0115] ;

[0116] in, Indicates candidate reliable paths Overall score Indicates candidate reliable paths Transmission delay, Indicates candidate reliable paths Total reputation score Indicates candidate reliable paths Number of jumps Indicates candidate reliable paths Resource redundancy performance, These represent the weighting coefficients for the total reputation score and resource redundancy, respectively.

[0117] The service migration path is expressed as:

[0118] ;

[0119] in, Indicates the service migration path. Indicates candidate reliable paths Overall score This represents the set of candidate reliable paths.

[0120] Secondly, a service migration system that balances reputation and fault tolerance mechanisms is provided, including: an edge node reputation assessment module, a candidate reliable path construction module, a fault tolerance redundancy and fault tolerance adjustment module, and a service migration path selection module;

[0121] The edge node reputation assessment module is used to obtain the starting edge node and the target migration node where the service is located, and to assess the reputation of edge nodes within the region; the total reputation score is obtained by weighting the interaction reputation, energy reputation and recommendation reputation of each edge node.

[0122] The candidate reliable path construction module is used to set the reputation threshold of edge nodes based on the reputation assessment results, filter edge nodes to construct a path network topology map; use a depth-first traversal algorithm to obtain all migration paths, and filter out a set of candidate paths based on the path quality score; further evaluate the resource redundancy performance of each candidate path, filter out candidate reliable paths and form a set of candidate reliable paths.

[0123] The fault-tolerant redundancy and fault-tolerant adjustment module is used to implement horizontal and vertical redundancy strategies on candidate reliable paths to maintain service stability; and based on the fault-tolerant feedback adjustment mechanism, it continuously monitors the operating status of edge nodes on the path and dynamically adjusts the total reputation score of edge nodes according to the success or failure of the migration task.

[0124] The service migration path selection module is used to calculate the transmission latency of each path in the candidate reliable path set, and calculate a comprehensive score by combining the path's total reputation value, path hop count, and resource redundancy performance; finally, the candidate reliable path with the highest comprehensive score is selected as the service migration path.

[0125] The beneficial effects of the invention are:

[0126] This invention provides a service migration method that balances reputation and fault tolerance. By introducing a multi-dimensional reputation evaluation mechanism that includes interactive reputation, energy reputation, and recommendation reputation, the service migration path selection becomes more accurate and reliable.

[0127] This invention provides a service migration method that balances reputation and fault tolerance. By constructing a multi-stage screening mechanism, it comprehensively considers path quality and deeply evaluates its redundancy capabilities, ensuring that the final selected migration path has both high performance and high reliability.

[0128] This invention provides a service migration method that balances reputation and fault tolerance mechanisms. By combining a fault-tolerant feedback mechanism, the system can make dynamic adjustments based on real-time changes in the environment, ensuring the optimality and adaptability of the service migration path selection. Attached Figure Description

[0129] Figure 1 This is a schematic diagram of a service migration method that balances reputation and fault tolerance mechanisms, provided by an embodiment of the present invention.

[0130] Figure 2 This is a schematic diagram of a path network topology provided in one embodiment of the present invention.

[0131] Figure 3 This is a flowchart of constructing a candidate path set provided in one embodiment of the present invention.

[0132] Figure 4 This is a schematic diagram of a service migration system that balances reputation and fault tolerance mechanisms, provided by an embodiment of the present invention. Detailed Implementation

[0133] The preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby providing a clearer and more explicit definition of the scope of protection of the present invention.

[0134] Many specific details are set forth in the following description in order to provide a full understanding of the invention, but the invention may also be practiced in other ways different from those described herein; obviously, the embodiments in the specification are only some embodiments of the invention, and not all embodiments.

[0135] The specific implementation of the technical solution of the present invention includes the following:

[0136] Example 1

[0137] The service migration method that balances reputation and fault tolerance mechanisms according to Embodiment 1 of the present invention includes:

[0138] The system obtains the starting edge node and target migration node corresponding to the service used by the user, and performs a reputation assessment on the edge nodes in the edge node set. Based on the reputation assessment results, the edge nodes are filtered to construct a migration path set and select a candidate path set. The resource redundancy performance of each candidate path in the candidate path set is evaluated, and candidate reliable paths are selected to form a candidate reliable path set. A fault-tolerant redundancy strategy is implemented on the candidate reliable paths in the candidate reliable path set to maintain service stability, a fault-tolerant feedback adjustment mechanism is established, and the fault-tolerant strategy is adjusted according to the feedback results. The transmission latency of each candidate reliable path in the candidate reliable path set is calculated, and the candidate reliable path with the highest comprehensive score is selected as the service migration path.

[0139] Specifically, Figure 1 A flowchart illustrating a service migration method that balances reputation and fault tolerance mechanisms, as shown in Embodiment 1 of the application, is presented, including:

[0140] S1. Obtain the starting edge node and target migration node corresponding to the service used by the user, and calculate the total reputation score of the edge nodes in the edge node set.

[0141] The calculation of the total reputation score of the edge nodes in the edge node set in step S1 refers to: calculating the total reputation score by weighting the interaction reputation, energy reputation and recommendation reputation of each edge node.

[0142] The set of edge nodes refers to the set of all edge nodes within the region from the starting edge node to the target migration node, expressed as:

[0143] ;

[0144] in, Represents the set of edge nodes. Indicates the starting edge node, Indicates the target migration node. This indicates the number of edge nodes in the edge node set.

[0145] For example, a user is using a high-definition video conferencing service and traveling by vehicle from a starting area "Location A" to a destination area "Location B". Their service is currently connected to the starting edge node. "Location A Base Station" ( (and is expected to migrate to the target migration node after reaching its destination). "Location B base station" ( In this case, the set of edge nodes. This refers to users from Move to The physical path includes all available edge servers, including the originating node, the target node, and the relay nodes along the way, for example: The method of this invention calculates the set of edge nodes. The total reputation score of each edge node is used for subsequent analysis and processing.

[0146] The aforementioned interactive reputation refers to a quantitative indicator constructed based on the number of successful and failed task deliveries in the history of edge nodes. The calculation formula is as follows:

[0147] ;

[0148] in, Represents edge nodes Interaction reputation value, , These represent the certainty factor and uncertainty factor of interactive reputation, respectively. Represents edge nodes Number of successful historical mission deliveries Represents edge nodes Number of historical mission delivery failures;

[0149] The energy reputation refers to a quantitative indicator built upon the energy consumed by edge nodes in executing tasks, and its calculation formula is as follows:

[0150] ;

[0151] in, Represents edge nodes exist Energy reputation value at any time Represents edge nodes exist Energy reputation value at any time This represents the memory factor, used to control the historical energy reputation weight. Represents edge nodes exist The confidence level of anomalies at any given moment;

[0152] The anomaly confidence level is calculated using the following formula:

[0153] ;

[0154] in, Represents edge nodes exist The confidence level of anomalies at any given moment. Indicates the tolerance coefficient. Represents edge nodes exist Power deviation at any given time;

[0155] The power deviation is calculated using the following formula:

[0156] ;

[0157] in, Represents edge nodes exist Power deviation at time, Represents edge nodes exist Instantaneous power at a given moment Represents edge nodes exist Normal reference power at time, Represents edge nodes exist Standard deviation of power fluctuation at time t. Indicates the smoothing factor;

[0158] The recommended reputation refers to the reputation of other edge nodes towards this edge node, calculated using the following formula:

[0159] , , ;

[0160] in, Represents edge nodes Recommended reputation value, Represents the set of edge nodes. Represents edge nodes For edge nodes Recommendation reputation score Represents edge nodes Recommendation factor, Represents the set of edge nodes The number of edge nodes;

[0161] The total reputation score is calculated using the following formula:

[0162] ;

[0163] in, Represents edge nodes Total reputation score These represent the weight values ​​of interaction reputation, energy reputation, and recommendation reputation, respectively, with a default value. , Representing edge nodes respectively The interaction reputation value, energy reputation value, and recommendation reputation value.

[0164] For example, Node historical task delivery successful Second, failure Next, calculated according to the interactive reputation formula .exist time Instantaneous power of nodes Reference power Standard deviation of power fluctuation Memory factors , Calculations show that time Node Energy Reputation . Received weighted recommendation reputation Based on default weights , and thus The node's total reputation score Calculate the set of edge nodes sequentially. The total reputation score of each edge node is used in subsequent screening steps.

[0165] S2. Set the reputation threshold for edge nodes, filter edge nodes based on the total reputation score, construct a migration path set, and filter out a candidate path set.

[0166] Step S2, which involves setting an edge node reputation threshold and filtering edge nodes based on the total reputation score, includes:

[0167] Set an edge node reputation threshold to filter out edge nodes whose total reputation score is not lower than the edge node reputation threshold.

[0168] Specifically, the formula for selecting edge nodes whose total reputation score is not lower than the edge node reputation threshold is as follows:

[0169] ;

[0170] ;

[0171] in, This represents the set of reputation edge nodes, consisting of edge nodes whose total reputation score is not lower than the reputation threshold of edge nodes. Represents the set of edge nodes. Represents edge nodes Total reputation score This represents the reputation threshold of the edge node. This represents the average reputation of all edge nodes. This represents the standard deviation of the reputation of all edge nodes. This represents the sensitivity coefficient.

[0172] Step S2, which involves constructing a migration path set and filtering out a candidate path set, includes:

[0173] A path network topology graph is constructed based on the edge nodes retained after filtering. A depth-first traversal algorithm is used to obtain all migration paths from the starting edge node to the target migration node, forming a migration path set. The path quality of each migration path in the migration path set is evaluated and sorted in descending order. The top 30% of the migration paths are selected as candidate paths, forming a candidate path set.

[0174] Specifically, such as Figure 2 As shown, the path network topology graph constructed based on edge nodes involved in Embodiment 1 of this application, specifically includes the following:

[0175] a1: Define network topology elements, including: based on the connection relationships of edge nodes retained after filtering, abstractly represent edge nodes as graph nodes, and abstractly represent the connections between edge nodes as graph edges;

[0176] a2: Construct a path network topology graph, including: integrating graph nodes and edges, and establishing a mathematical model of the path network topology graph, the expression of which is:

[0177] ;

[0178] in, Represents the path network topology graph. Represents the set of reputation edge nodes. Represents the set of edges in a graph;

[0179] a3: Define the set of graph edges, including: specifying the rules for constructing the set of graph edges, the expression of which is:

[0180] ;

[0181] in, Represents the set of graph edges. Represents edge nodes The logical connection between them This represents the set of reputation edge nodes.

[0182] Specifically, such as Figure 3 As shown in the flowchart of the construction of the candidate path set involved in Embodiment 1 of this application, the step of using a depth-first traversal algorithm to obtain all migration paths from the starting edge node to the target migration node constitutes a migration path set. The path quality of each migration path in the migration path set is evaluated and sorted in descending order. The top 30% of the migration paths are selected as candidate paths to form the candidate path set. Specifically, this includes:

[0183] b1: Initialize the migration path set This includes: the initial set of migration paths. Set to an empty set;

[0184] b2: Use a depth-first traversal algorithm to find the migration path. This includes: using a depth-first traversal algorithm to obtain the migration path from the starting edge node to the target migration node in the network topology graph. ;

[0185] b3: Iteratively update the path set and determine the termination condition, including: determining whether a migration path has been successfully found. If a path is successfully found, it will be added to the migration path set. Then proceed to step b2 to continue searching for the next path; if no path is found and the set If the value is empty, it is determined that no migration path was found and the service migration process is terminated; if no path was found but the set is empty, it is determined that no migration path was found and the service migration process is terminated. If not empty, stop the search and proceed to step b4;

[0186] b4: Calculate the set of migration paths The quality score of each migration path includes: traversing the set of migration paths. For all paths in the migration, calculate the path quality score for each migration path.

[0187] b5: Sort the migration path set, including: based on the path quality scores obtained in step b4, sort all migration paths in the migration path set in descending order of their scores;

[0188] b6: Select a candidate path set, including: based on the ranking results, select the top 30% of migration paths as candidate paths and form a candidate path set.

[0189] The path quality score is calculated using the following formula:

[0190] ;

[0191] in, Indicates migration path Path quality score, This represents the total reputation value of the migration path. This indicates the number of hops in the migration path. Indicates migration path Minimum bandwidth, These represent the weighting coefficients for total reputation score, hop count, and minimum bandwidth, respectively. ;

[0192] The total reputation value of the migration path refers to the sum of the total reputation scores of all edge nodes on the migration path, calculated using the following formula:

[0193] ;

[0194] in, This represents the total reputation value of the migration path. Represents edge nodes Total reputation score This indicates the number of edge nodes in the migration path;

[0195] The candidate path set is expressed as:

[0196] ;

[0197] in, Represents the set of candidate paths. Indicates the first Candidate paths, This represents the total number of candidate paths in the candidate path set.

[0198] For example, based on the set of reputation edge nodes generated in S2 Construct a path network topology graph Then, S12 is executed, using a depth-first traversal algorithm, starting from the initial node " "To the target node" "The depth-first traversal algorithm found 10 different migration paths from the starting edge node to the target migration node and stored them in the migration path set." Next, execute S13 to calculate... Each path in Path quality score Finally, execute S14 to sort the 10 paths in the migration path set according to their... Sort the scores in descending order, for example, the sorted result is: The top 30% of paths, i.e., the three paths with the highest scores, are selected to form a candidate path set. The subsequent analysis of the candidate path set Resource redundancy assessment is performed on the candidate paths.

[0199] S3. Evaluate the resource redundancy performance of each candidate path in the candidate path set, and select candidate reliable paths to form a candidate reliable path set.

[0200] The evaluation of the resource redundancy performance of each candidate path in the candidate path set in step S3 includes: calculating the remaining computing resources of each edge node in the candidate path, and evaluating the horizontal redundancy capability, vertical redundancy capability, and resource redundancy performance of each candidate path.

[0201] The remaining computing resources are calculated using the following formula:

[0202] ;

[0203] in, Represents edge nodes The remaining computing resources Representing edge nodes respectively The current available CPU, memory, and bandwidth. Representing edge nodes respectively The total amount of CPU, memory, and bandwidth resources. These are the weighting coefficients, default. ;

[0204] The aforementioned horizontal redundancy capability refers to the number of nodes that create backup tasks in the candidate paths, calculated using the following formula:

[0205] ;

[0206] ;

[0207] in, Indicate candidate path Horizontal redundancy capability Indicate candidate path The number of edge nodes that can be used as backups Indicate candidate path The number of edge nodes in the data. As a constraint, it means that the remaining computing resources of the edge node must be greater than or equal to the minimum resources required for the backup task multiplied by the redundancy factor. This indicates the minimum computing resource requirements for the backup task. Indicates the redundancy coefficient;

[0208] The vertical redundancy capability refers to: calculating the vertical expansion potential of each key edge node in the candidate path, and determining the average vertical expansion potential of all key edge nodes in the candidate path as the vertical redundancy capability of the candidate path. The calculation formula is as follows:

[0209] ;

[0210] ;

[0211] in, Indicate candidate path Longitudinal redundancy capability Indicates the number of critical edge nodes. Represents the set of key edge nodes. Represents key edge nodes Vertical expansion potential Represents key edge nodes Current resource usage Represents key edge nodes Maximum resource that can be dynamically expanded;

[0212] The key edge nodes include: the edge node with the lowest total reputation score in the candidate path, the edge node with a node load exceeding 80%, and the edge node directly adjacent to the starting edge node or the target migration node.

[0213] The resource redundancy performance is calculated using the following formula:

[0214] ;

[0215] in, Indicate candidate path Resource redundancy performance, Indicate candidate path Horizontal redundancy capability Indicate candidate path Longitudinal redundancy capability Indicate candidate path Path quality score, These are the weighting coefficients, default. .

[0216] Step S3, which involves selecting candidate reliable paths to form a candidate reliable path set, includes:

[0217] Based on the resource redundancy performance of the candidate paths, the candidate paths are sorted in descending order, and the top 30% of the candidate paths are selected as candidate reliable paths to form a set of candidate reliable paths.

[0218] The set of candidate reliable paths is expressed as:

[0219] ;

[0220] in, Represents the set of candidate reliable paths. This represents the candidate reliable path with the highest resource redundancy performance in the candidate reliable path set. Indicates the resource redundancy performance of the candidate reliable path set. High candidate reliable path, This indicates the number of candidate reliable paths in the candidate reliable path set.

[0221] S4. Implement a fault-tolerant redundancy strategy for the candidate reliable paths in the candidate reliable path set to maintain service stability, establish a fault-tolerant feedback adjustment mechanism, and adjust the fault-tolerant strategy according to the feedback results.

[0222] The fault-tolerant redundancy strategy implemented in step S4 for the candidate reliable paths in the candidate reliable path set includes: a horizontal redundancy strategy and a vertical redundancy strategy.

[0223] The aforementioned horizontal redundancy strategy refers to:

[0224] The remaining computing resources are selected from the candidate reliable paths to satisfy Constraints are imposed such that the total reputation score is greater than the reputation threshold of the edge node, and the edge node with the fewest hops from the starting edge node is deployed with a replica to ensure the state synchronization of the replica, thereby achieving single-node fault tolerance.

[0225] The vertical redundancy strategy refers to:

[0226] By adjusting task priorities and dynamically scheduling resources, the fault tolerance of the critical edge nodes is improved, ensuring that they can maintain service stability under high load or failure.

[0227] Specifically, the improvement of the fault tolerance of key edge nodes through task priority adjustment and dynamic resource scheduling, ensuring that they can maintain service stability under high load or failure, includes:

[0228] Resource allocation is controlled by priority tags, tasks are divided into critical tasks and ordinary tasks, and task priorities are adjusted and resources are scheduled according to resource preemption rules.

[0229] The key tasks mentioned above refer to: state synchronization task and real-time migration task;

[0230] The common tasks mentioned above refer to: computing tasks and data backup;

[0231] The resource preemption rules include:

[0232] The resources of the key edge nodes are divided into reserved resources, key task resources, and ordinary task resources;

[0233] When available resources are insufficient, terminate the normal task to free up resources for the critical task; when available resources are sufficient, maintain the current task.

[0234] The reserved resources, accounting for 10%, are used to configure resource reservation strategies for the key edge nodes.

[0235] The critical task resources, accounting for 60%, are used to execute critical tasks.

[0236] The resources for ordinary tasks are used to perform ordinary tasks and account for 30%.

[0237] The deployed copy refers to:

[0238] To prevent single-node failures from impacting services, task backups are performed on other edge nodes.

[0239] The fault-tolerant feedback adjustment mechanism described in step S4 includes: fault-tolerant positive feedback adjustment and fault-tolerant negative feedback adjustment;

[0240] The aforementioned fault-tolerant positive feedback adjustment refers to:

[0241] If an edge node successfully completes its migration task without experiencing any failures during the task, its overall reputation score will be increased. The calculation formula is as follows:

[0242] ;

[0243] in, express Time edge node Total reputation score express Time edge node Total reputation score This represents the maximum gain coefficient. Indicates the growth rate factor. Represents edge nodes Number of times historical tasks were successfully delivered;

[0244] The fault-tolerant negative feedback adjustment refers to:

[0245] If an edge node fails or malfunctions during a migration task, its usage frequency in future migrations will be reduced, and the node's overall reputation score will be lowered. The calculation formula is as follows:

[0246] ;

[0247] ;

[0248] in, Represents edge nodes Total reputation score express Time edge node Total reputation score express Time edge node Total reputation score Represents the decay rate factor. Represents the residual coefficient. Indicates the lower limit of the benchmark credit rating. Represents the set of edge nodes. Represents the edge set The number of edge nodes.

[0249] S5. Calculate the transmission delay of each candidate reliable path in the candidate reliable path set, and select the candidate reliable path with the highest comprehensive score as the service migration path.

[0250] The transmission delay mentioned in step S5 is calculated using the following formula:

[0251] ;

[0252] in, Indicates candidate reliable paths Transmission delay, Indicates the edges in the candidate reliable path physical distance, Indicates the speed of signal propagation. Indicates the amount of work. Representing an edge Available bandwidth;

[0253] The comprehensive score is calculated using the following formula:

[0254] ;

[0255] in, Indicates candidate reliable paths Overall score Indicates candidate reliable paths Transmission delay, Indicates candidate reliable paths Total reputation score Indicates candidate reliable paths Number of jumps Indicates candidate reliable paths Resource redundancy performance, These represent the weighting coefficients for the total reputation score and resource redundancy, respectively.

[0256] The service migration path is expressed as:

[0257] ;

[0258] in, Indicates the service migration path. Indicates candidate reliable paths Overall score This represents the set of candidate reliable paths.

[0259] Example 2

[0260] like Figure 4 As shown in Embodiment 2 of this application, a service migration system that balances reputation and fault tolerance mechanisms includes: an edge node reputation assessment module, a candidate reliable path construction module, a fault tolerance redundancy and fault tolerance adjustment module, and a service migration path selection module.

[0261] Specifically, the edge node reputation assessment module is used to obtain the starting edge node and the target migration node where the service is located, and to perform reputation assessment on the edge nodes within the region; by calculating the interaction reputation, energy reputation and recommendation reputation of each edge node, the total reputation score is obtained;

[0262] The candidate reliable path construction module is used to set the reputation threshold of edge nodes based on the reputation assessment results, filter edge nodes to construct a path network topology map; use a depth-first traversal algorithm to obtain all migration paths, and filter out a set of candidate paths based on the path quality score; further evaluate the resource redundancy performance of each candidate path, filter out candidate reliable paths and form a set of candidate reliable paths.

[0263] The fault-tolerant redundancy and fault-tolerant adjustment module is used to implement horizontal and vertical redundancy strategies on candidate reliable paths to maintain service stability; and based on the fault-tolerant feedback adjustment mechanism, it continuously monitors the operating status of edge nodes on the path and dynamically adjusts the total reputation score of edge nodes according to the success or failure of the migration task.

[0264] The service migration path selection module is used to calculate the transmission latency of each path in the candidate reliable path set, and calculate a comprehensive score by combining the path's total reputation value, path hop count, and resource redundancy performance; finally, the candidate reliable path with the highest comprehensive score is selected as the service migration path.

[0265] The specific implementation method of this embodiment is the same as that of Embodiment 1, and will not be repeated here. Please refer to the description of Embodiment 1 for details.

[0266] Those skilled in the art will understand that although some embodiments herein include certain features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of the invention and form different embodiments.

[0267] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A service migration method that balances reputation and fault tolerance mechanisms, characterized in that, include: Obtain the starting edge node and target migration node corresponding to the service used by the user, and calculate the total reputation score by weighting the interaction reputation, energy reputation and recommendation reputation of the edge nodes in the edge node set; The interactive reputation refers to a quantitative indicator constructed based on the number of successful and failed task deliveries in the history of edge nodes. The energy reputation refers to a quantitative indicator built upon the energy consumed by edge nodes in performing tasks. The recommended reputation refers to the recommended reputation of this edge node by other edge nodes. Set a reputation threshold for edge nodes, filter edge nodes based on the total reputation score, construct a migration path set, and filter out a candidate path set; Evaluate the resource redundancy performance of each candidate path in the candidate path set, and select candidate reliable paths to form a candidate reliable path set; The evaluation of the resource redundancy performance of each candidate path in the candidate path set includes: calculating the remaining computing resources of each edge node in the candidate path, and evaluating the horizontal redundancy capability, vertical redundancy capability, and resource redundancy performance of each candidate path. The aforementioned horizontal redundancy capability refers to the number of nodes that create backup tasks in the candidate paths, calculated using the following formula: ； ； in, Indicate candidate path Horizontal redundancy capability Indicate candidate path The number of edge nodes that can be used as backups Indicate candidate path The number of edge nodes in the data. As a constraint, it means that the remaining computing resources of the edge node must be greater than or equal to the minimum resources required for the backup task multiplied by the redundancy factor. This indicates the minimum computing resource requirements for the backup task. Indicates the redundancy coefficient; The vertical redundancy capability refers to: calculating the vertical expansion potential of each key edge node in the candidate path, and determining the average vertical expansion potential of all key edge nodes in the candidate path as the vertical redundancy capability of the candidate path. The calculation formula is as follows: ； ； in, Indicate candidate path Longitudinal redundancy capability Indicates the number of critical edge nodes. Represents the set of key edge nodes. Represents key edge nodes Vertical expansion potential Represents key edge nodes Current resource usage Represents key edge nodes Maximum resource that can be dynamically expanded; The key edge nodes include: the edge node with the lowest total reputation score in the candidate path, the edge node with a node load exceeding 80%, and the edge node directly adjacent to the starting edge node or the target migration node. The resource redundancy performance is calculated using the following formula: ； in, Indicate candidate path Resource redundancy performance, Indicate candidate path Horizontal redundancy capability Indicate candidate path Longitudinal redundancy capability Indicate candidate path Path quality score, Indicates the weighting coefficient; A fault-tolerant redundancy strategy is implemented on the candidate reliable paths in the candidate reliable path set to maintain service stability, a fault-tolerant feedback adjustment mechanism is established, and the fault-tolerant strategy is adjusted according to the feedback results. Calculate the transmission delay of each candidate reliable path in the candidate reliable path set, and select the candidate reliable path with the highest comprehensive score as the service migration path.

2. The service migration method that balances reputation and fault tolerance mechanisms as described in claim 1, characterized in that, The set of edge nodes refers to the set of all edge nodes within the region from the starting edge node to the target migration node, expressed as: ； in, Represents the set of edge nodes. Indicates the starting edge node, Indicates the target migration node. Represents the set of edge nodes The number of edge nodes.

3. A service migration method that balances reputation and fault tolerance mechanisms as described in claim 2, characterized in that, The formula for calculating the interactive reputation is as follows: ； in, Represents edge nodes Interaction reputation value, , These represent the certainty factor and uncertainty factor of the interactive reputation, respectively. Represents edge nodes Number of successful historical mission deliveries Represents edge nodes Number of historical mission delivery failures; The energy credit is calculated using the following formula: ； in, Represents edge nodes exist Energy reputation value at any moment Represents edge nodes exist Energy reputation value at any moment This represents the memory factor, used to control the historical energy reputation weight. Represents edge nodes exist The confidence level of anomalies at any given moment; The anomaly confidence level is calculated using the following formula: ； in, Represents edge nodes exist The confidence level of anomalies at any given moment. Indicates the tolerance coefficient. Represents edge nodes exist Power deviation at any given time; The power deviation is calculated using the following formula: ； in, Represents edge nodes exist Power deviation at time, Represents edge nodes exist Instantaneous power at a given moment Represents edge nodes exist Normal reference power at time, Represents edge nodes exist Standard deviation of power fluctuation at time t. Indicates the smoothing factor; The recommendation reputation is calculated using the following formula: ， ， ； in, Represents edge nodes Recommended reputation value, Represents the set of edge nodes. Represents edge nodes For edge nodes Recommendation reputation score Represents edge nodes Recommendation factor, Represents the set of edge nodes The number of edge nodes; The total reputation score is calculated using the following formula: ； in, Represents edge nodes Total reputation score These represent the weight values ​​of interaction reputation, energy reputation, and recommendation reputation, respectively. Representing edge nodes respectively The interaction reputation value, energy reputation value, and recommendation reputation value.

4. A service migration method that balances reputation and fault tolerance mechanisms as described in claim 3, characterized in that, The process of setting an edge node reputation threshold, filtering edge nodes based on the total reputation score, constructing a migration path set, and selecting a candidate path set includes: Set an edge node reputation threshold to filter out edge nodes whose total reputation score is not lower than the edge node reputation threshold; A path network topology graph is constructed based on the edge nodes retained after filtering. A depth-first traversal algorithm is used to obtain all migration paths from the starting edge node to the target migration node, forming a migration path set. The path quality of each migration path in the migration path set is evaluated and sorted in descending order. The top 30% of the migration paths are selected as candidate paths, forming a candidate path set.

5. A service migration method that balances reputation and fault tolerance mechanisms as described in claim 4, characterized in that, The formula for selecting edge nodes whose total reputation score is not lower than the edge node reputation threshold is as follows: , ； in, This represents the set of reputation edge nodes, consisting of edge nodes whose total reputation score is not lower than the reputation threshold of edge nodes. Represents the set of edge nodes. Represents edge nodes Total reputation score This represents the reputation threshold of the edge node. This represents the average reputation of all edge nodes. This represents the standard deviation of the reputation of all edge nodes. This represents the sensitivity coefficient.

6. A service migration method that balances reputation and fault tolerance mechanisms as described in claim 5, characterized in that, The construction of the path network topology map based on the retained edge nodes after filtering includes: Based on the connection relationships of the edge nodes retained after filtering, the edge nodes are mapped to vertices of the path network topology graph, and the physical links between edge nodes are mapped to edges of the path network topology graph, thereby constructing a complete path network topology graph. The path network topology diagram is expressed as follows: ； in, Represents the path network topology graph. Represents the set of reputation edge nodes. Represents the set of edges in a graph; The graph edge set is expressed as: ； in, Represents the set of graph edges. Represents edge nodes The logical connection between them This represents the set of reputation edge nodes.

7. A service migration method that balances reputation and fault tolerance mechanisms as described in claim 6, characterized in that, The process employs a depth-first traversal algorithm to obtain all migration paths from the starting edge node to the target migration node, forming a migration path set. The path quality of each migration path in the set is evaluated and sorted in descending order. The top 30% of migration paths are selected as candidate paths, forming a candidate path set, including: S11: Initialize the migration path set The expression is: ,in Represents the empty set; S12: Using a depth-first traversal algorithm, obtain the migration path from the starting edge node to the target migration node. , migration path Add to migration path set The expression is: ; S13: Calculate the path quality score for each migration path in the migration path set; S14: Sort the migration paths in the migration path set in descending order based on the path quality score, and select the top 30% of the migration paths as candidate paths to form a candidate path set. The path quality score is calculated using the following formula: ； in, Indicates migration path Path quality score, This represents the total reputation value of the migration path. This indicates the number of hops in the migration path. Indicates migration path Minimum bandwidth, These represent the weighting coefficients for total reputation score, hop count, and minimum bandwidth, respectively. ; The total reputation value of the migration path refers to the sum of the total reputation scores of all edge nodes on the migration path, calculated using the following formula: ； in, This represents the total reputation value of the migration path. Represents edge nodes Total reputation score Indicates the number of edge nodes in the migration path; The candidate path set is expressed as: ； in, Represents the set of candidate paths. Indicates the first Candidate paths, This represents the total number of candidate paths in the candidate path set.

8. A service migration method that balances reputation and fault tolerance mechanisms as described in claim 7, characterized in that, The remaining computing resources are calculated using the following formula: ； in, Represents edge nodes The remaining computing resources Representing edge nodes respectively The current available CPU, memory, and bandwidth. Representing edge nodes respectively The total amount of CPU, memory, and bandwidth resources. Indicates the weighting coefficient; The selected candidate reliable paths constitute a candidate reliable path set, including: Based on the resource redundancy performance of the candidate paths, the candidate paths are sorted in descending order, and the top 30% of the candidate paths are selected as candidate reliable paths to form a set of candidate reliable paths. The set of candidate reliable paths is expressed as: ； in, Represents the set of candidate reliable paths. This represents the candidate reliable path with the highest resource redundancy performance in the candidate reliable path set. Indicates the resource redundancy performance of the candidate reliable path set. High candidate reliable path, This indicates the number of candidate reliable paths in the candidate reliable path set.

9. A service migration method that balances reputation and fault tolerance mechanisms as described in claim 8, characterized in that, The implementation of a fault-tolerant redundancy strategy on the candidate reliable paths in the candidate reliable path set includes: a horizontal redundancy strategy and a vertical redundancy strategy. The aforementioned horizontal redundancy strategy refers to: The remaining computing resources are selected from the candidate reliable paths to satisfy Constraints are imposed such that the total reputation score is greater than the reputation threshold of the edge node, and the edge node with the fewest hops from the starting edge node is deployed with a replica to ensure the state synchronization of the replica, thereby achieving single-node fault tolerance. The vertical redundancy strategy refers to: By adjusting task priorities and dynamically scheduling resources, the fault tolerance of the key edge nodes is improved, ensuring that they can maintain service stability under high load or failure. The deployed copy refers to: To prevent single-node failures from impacting services, task backups are performed on other edge nodes.

10. A service migration method that balances reputation and fault tolerance mechanisms as described in claim 9, characterized in that, The improvement of the fault tolerance of key edge nodes through task priority adjustment and dynamic resource scheduling, ensuring that they can maintain service stability under high load or failure, includes: Resource allocation is controlled by priority tags, tasks are divided into critical tasks and ordinary tasks, and task priorities are adjusted and resources are scheduled according to resource preemption rules. The key tasks mentioned above refer to: state synchronization task and real-time migration task; The common tasks mentioned above refer to: computing tasks and data backup; The resource preemption rules include: The resources of the key edge nodes are divided into reserved resources, key task resources, and ordinary task resources; When available resources are insufficient, terminate the normal task to free up resources for the critical task; when available resources are sufficient, maintain the current task. The reserved resources are used to configure resource reservation strategies for the key edge nodes; The critical task resources are used to execute critical tasks; The general task resources are used to perform general tasks.

11. A service migration method that balances reputation and fault tolerance mechanisms as described in claim 10, characterized in that, The fault-tolerant feedback adjustment mechanism includes: fault-tolerant positive feedback adjustment and fault-tolerant negative feedback adjustment; The aforementioned fault-tolerant positive feedback adjustment refers to: If an edge node successfully completes its migration task without experiencing any failures during the task, its overall reputation score will be increased. The calculation formula is as follows: ； in, express Time edge node Total reputation score express Time edge node Total reputation score This represents the maximum gain coefficient. Indicates the growth rate factor. Represents edge nodes Number of times historical tasks were successfully delivered; The fault-tolerant negative feedback adjustment refers to: If an edge node fails or malfunctions during a migration task, its usage frequency in future migrations will be reduced, and the node's overall reputation score will be lowered. The calculation formula is as follows: ； ； in, Represents edge nodes Total reputation score express Time edge node Total reputation score express Time edge node Total reputation score Represents the decay rate factor. Represents the residual coefficient. Indicates the lower limit of the benchmark credit rating. Represents the set of edge nodes. Represents the edge set The number of edge nodes.

12. A service migration method that balances reputation and fault tolerance mechanisms as described in claim 11, characterized in that, The transmission delay is calculated using the following formula: ； in, Indicates candidate reliable paths Transmission delay, Indicates the edges in the candidate reliable path physical distance, Indicates the speed of signal propagation. Indicates the amount of work. Representing an edge Available bandwidth; The comprehensive score is calculated using the following formula: ； in, Indicates candidate reliable paths Overall score Indicates candidate reliable paths Transmission delay, Indicates candidate reliable paths Total reputation score Indicates candidate reliable paths Number of jumps Indicates candidate reliable paths Resource redundancy performance, These represent the weighting coefficients for the total reputation score and resource redundancy, respectively. The service migration path is expressed as: ； in, Indicates the service migration path. Indicates candidate reliable paths Overall score This represents the set of candidate reliable paths.

13. A service migration system that balances reputation and fault tolerance mechanisms, characterized in that, The service migration method that balances reputation and fault tolerance according to any one of claims 1-12 specifically includes: The edge node reputation assessment module is used to obtain the starting edge node and target migration node where the service is located, and to assess the reputation of edge nodes within the region; the total reputation score is obtained by weighting the interaction reputation, energy reputation and recommendation reputation of each edge node. The candidate reliable path construction module is used to set the reputation threshold of edge nodes based on the reputation assessment results, and to filter edge nodes to construct a path network topology. It uses a depth-first traversal algorithm to obtain all migration paths and filters out the candidate path set based on the path quality score. It further evaluates the resource redundancy performance of each candidate path, filters out candidate reliable paths, and forms a candidate reliable path set. The fault-tolerant redundancy and fault-tolerant adjustment module is used to implement horizontal and vertical redundancy strategies on candidate reliable paths to maintain service stability; and based on the fault-tolerant feedback adjustment mechanism, it continuously monitors the running status of edge nodes on the path and dynamically adjusts the total reputation score of edge nodes according to the success or failure of the migration task. The service migration path selection module is used to calculate the transmission latency of each path in the candidate reliable path set, and to calculate a comprehensive score by combining the path's total reputation value, path hop count, and resource redundancy performance; finally, the candidate reliable path with the highest comprehensive score is selected as the service migration path.

Images