Multi-agent cluster-oriented data backup method and device, equipment and medium

Abstract

This application discloses a data backup method, apparatus, device, and medium for multi-agent clusters, relating to the field of cloud computing data backup technology. The method includes: obtaining the data generation rate, remaining storage capacity, and transmission capacity of the communication links between nodes in the multi-agent cluster; dynamically allocating the amount of backup data using a data integrity priority or traffic balancing strategy; and inputting the results into a backup configuration matrix to perform data backup. This improves the efficiency and reliability of data backup in complex environments, ensures that data can be recovered when nodes fail, and enhances the system's adaptability and robustness.

Description

Technical Field

[0001] This invention relates to the field of cloud computing data backup technology, and in particular to a data backup method, apparatus, device, and medium for multi-agent clusters. Background Technology

[0002] Currently, centralized backup schemes are commonly used in data center networks, backing up data to dedicated storage devices via high-speed, stable networks. While these schemes achieve efficient data backup and rapid recovery, they have limitations when dealing with distributed systems. In wireless sensor networks, optimization techniques such as the alternating direction multiplier method and mixed-integer linear programming, by establishing multi-objective optimization models and comprehensively considering multi-dimensional constraints on data backup performance and communication efficiency, can meet data backup requirements to a certain extent. Furthermore, there are learning-based algorithms that predict data backup needs and optimize backup strategies using machine learning models. However, most existing data backup algorithms are designed for static or relatively stable network environments. In the dynamically changing environment of multi-agent clusters, they struggle to adapt in real-time to the dynamic addition and removal of nodes, changes in communication topology, and the heterogeneity of node computing and storage capabilities. This can lead to backup strategy failures or a significant decrease in backup efficiency. Therefore, there is an urgent need for a more efficient and accurate data backup method to achieve data stability in multi-agent clusters under complex environments. Summary of the Invention

[0003] The main objective of this application is to provide a data backup method, apparatus, device, and medium for multi-agent clusters, aiming to solve the technical problem of how to improve the stability of multi-agent cluster data in complex environments.

[0004] To achieve the above objectives, this application proposes a data backup method for multi-agent clusters, comprising:

[0005] Obtain the data generation rate, remaining storage capacity, and transmission capacity of inter-node communication links for all nodes in a multi-agent cluster;

[0006] Based on the data generation rate, the remaining storage capacity, and the transmission capacity of the inter-node communication link, the amount of backup data is allocated using a data integrity priority strategy or a traffic balancing strategy to obtain the allocation result;

[0007] The allocation result is input into the backup configuration matrix to perform data backup operations between the source node and the target node. The rows of the backup configuration matrix represent the source node, the columns of the backup configuration matrix represent the target node, and the elements in the matrix represent the amount of backup data from the source node to the target node.

[0008] In one embodiment, the step of allocating the backup data volume according to the data generation rate, the remaining storage capacity, and the transmission capacity of the inter-node communication link using a data integrity priority strategy or a traffic balancing strategy to obtain the allocation result includes:

[0009] Initialize the remaining amount of data to be backed up to the total amount of original backup data for the corresponding node;

[0010] All nodes are sorted in descending order according to the data generation rate to obtain a sorted list of nodes.

[0011] Traverse the sorted list of nodes and sort the transmission capacity of the communication links of all nodes in descending order of channel capacity to obtain the sorted set of edges.

[0012] Traverse the sorted set of edges and calculate the first backup data volume of the corresponding node. The backup data volume is obtained by comparing the total original backup data volume of the current node, the remaining storage capacity of the target node, and the transmission capacity of the communication link between the current node and the target node. The specific formula is as follows:

[0013]

[0014] in, This represents the remaining storage capacity of the target node. The transmission capacity of the communication link between the current node and the target node. This represents the total amount of the original backup data. For the current node, For the target node, For time;

[0015] The first backup data volume is used as the first allocation result, which is obtained by processing with a data integrity priority strategy.

[0016] In one embodiment, the step of allocating the backup data volume according to the data generation rate, the remaining storage capacity, and the transmission capacity of the inter-node communication link using a data integrity priority strategy or a traffic balancing strategy to obtain the allocation result further includes:

[0017] Initialize the data generation rate, remaining storage capacity, and transmission capacity of the inter-node communication links for all nodes;

[0018] Sort all nodes in descending order according to the data generation rate to obtain a sorted list of nodes.

[0019] Traverse the sorted list of nodes and process all nodes in turn to obtain the set of all outgoing edge endpoint nodes;

[0020] Traverse the set of endpoint nodes to obtain the second backup data volume of the corresponding node, using the following formula:

[0021]

[0022] in, For the set of endpoint nodes, This represents the remaining storage capacity of the target node. The transmission capacity of the communication link between the current node and the target node. This represents the total amount of the original backup data. For the current node, For the target node, For time;

[0023] The second backup data volume is used as the second allocation result, which is obtained by processing with a traffic balancing strategy.

[0024] In one embodiment, the step of obtaining the data generation rate, remaining storage capacity, and transmission capacity of the inter-node communication links of all nodes in the multi-agent cluster includes:

[0025] The node's operational status data is acquired through distributed sensors, and the operational status data includes data generation rate and remaining storage capacity;

[0026] The transmission capacity of the inter-node communication link is obtained through a cooperative communication mechanism among multiple agents.

[0027] The target neural network model is used to analyze and adjust the running status data and the transmission capacity of the inter-node communication links to obtain the data generation rate, remaining storage capacity and transmission capacity of the inter-node communication links of all nodes in the final multi-agent cluster.

[0028] In one embodiment, before the step of analyzing and adjusting the running status data and the transmission capacity of the inter-node communication links using a target neural network model to obtain the data generation rate, remaining storage capacity, and transmission capacity of the inter-node communication links for all nodes in the final multi-agent cluster, the method further includes:

[0029] Obtain the historical operating status data of the node and construct an initial target neural network model;

[0030] The historical operating status data is input into the initial target neural network model for training to obtain the target neural network model;

[0031] After obtaining the target neural network model, the process further includes:

[0032] The running status data and the transmission capacity of the inter-node communication link are input into the target neural network model to obtain the predicted data value, which includes the predicted data generation rate of the node, the predicted remaining storage capacity, and the predicted transmission capacity of the inter-node communication link in the future time period.

[0033] Based on the predicted data values, the data backup priority and target node selection of the nodes are dynamically adjusted to obtain the data generation rate, remaining storage capacity, and transmission capacity of the inter-node communication links of all nodes in the multi-agent cluster.

[0034] In one embodiment, the step of inputting the allocation result into the backup configuration matrix to perform data backup operations between the source node and the target node includes:

[0035] The allocation result is divided according to the backup configuration matrix to obtain multiple backup data blocks;

[0036] The backup data block is sent from the source node to the corresponding target node using a preset communication protocol;

[0037] When the target node receives the backup data block, it performs a data integrity check and obtains the check result.

[0038] Based on the verification result, perform the corresponding data backup operation between the source node and the target node.

[0039] In one embodiment, the step of performing the corresponding data backup operation between the source node and the target node based on the verification result includes:

[0040] When the verification result is successful, the backup data block is stored in the target node, and a confirmation signal is sent back to the source node.

[0041] When the verification result is a failure, a resend signal is sent to the source node to cause the source node to resend the data block.

[0042] When the source node receives a confirmation signal, it prompts the user to complete the data backup operation.

[0043] Furthermore, to achieve the above objectives, this application also proposes a data backup device for multi-agent clusters, the data backup device for multi-agent clusters comprising:

[0044] The acquisition module is used to acquire the data generation rate, remaining storage capacity, and transmission capacity of the inter-node communication links of all nodes in the multi-agent cluster.

[0045] The allocation module is used to allocate the amount of backup data according to the data generation rate, the remaining storage capacity, and the transmission capacity of the inter-node communication link, using a data integrity priority strategy or a traffic balancing strategy, and obtain the allocation result.

[0046] The execution module is used to input the allocation result into the backup configuration matrix to perform data backup operations between the source node and the target node. The rows of the backup configuration matrix represent the source node, the columns of the backup configuration matrix represent the target node, and the elements in the matrix represent the amount of backup data from the source node to the target node.

[0047] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the data backup method for multi-agent clusters as described above.

[0048] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the data backup method for multi-agent clusters as described above.

[0049] This application obtains the data generation rate, remaining storage capacity, and inter-node communication link transmission capacity of all nodes in a multi-agent cluster. Based on these factors, it allocates backup data using either a data integrity priority strategy or a traffic balancing strategy. The allocation result is then input into a backup configuration matrix to execute data backup operations between the source and target nodes. By obtaining the data generation rate, remaining storage capacity, and inter-node communication link transmission capacity of each source node in the multi-agent cluster, and dynamically allocating backup data using either a data integrity priority or traffic balancing strategy, and inputting the result into the backup configuration matrix to execute data backup, this approach improves the efficiency and reliability of data backup in complex environments, ensures data recovery in the event of node failure, and enhances system adaptability and robustness. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a flowchart illustrating the first embodiment of the data backup method for multi-agent clusters in this application;

[0052] Figure 2This is a flowchart illustrating the second embodiment of the data backup method for multi-agent clusters in this application;

[0053] Figure 3 This is a flowchart illustrating the third embodiment of the data backup method for multi-agent clusters in this application;

[0054] Figure 4 This is a flowchart illustrating the fourth embodiment of the data backup method for multi-agent clusters in this application;

[0055] Figure 5 This is a schematic diagram of the module structure of a data backup device for a multi-agent cluster, which is a first embodiment of the data backup method for a multi-agent cluster of this application.

[0056] Figure 6 This is a schematic diagram of the hardware operating environment involved in the data backup method for multi-agent clusters in this application embodiment.

[0057] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0058] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0059] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0060] In today's rapidly evolving technological landscape, multi-agent systems (such as drone swarms and robotic teams) are widely used in critical scenarios like disaster response and remote area monitoring. These applications require agents to not only execute tasks efficiently but also possess robust data collection and processing capabilities. However, in practice, especially in adversarial environments or situations lacking stable communication infrastructure, ensuring data integrity and reliability becomes a significant challenge. Traditional data backup methods rely on stable communication links and centralized control architectures, making them ill-suited to the dynamically changing communication topologies of multi-agent clusters and lacking adaptability to frequent node joining and leaving. Furthermore, the enormous consumption of storage space and channel capacity in large-scale clusters further limits data backup efficiency.

[0061] Therefore, this application proposes a more efficient and accurate data backup method to achieve data stability in multi-agent clusters under complex environments. The main solution of this application's embodiments is as follows: Obtain the data generation rate, remaining storage capacity, and transmission capacity of inter-node communication links for all nodes in the multi-agent cluster; allocate the backup data volume according to the data generation rate, remaining storage capacity, and transmission capacity of inter-node communication links using a data integrity priority strategy or a traffic balancing strategy; obtain the allocation result; and input the allocation result into the backup configuration matrix to execute data backup operations between the source node and the target node.

[0062] Based on the above, this application also provides a data backup method for multi-agent clusters, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the data backup method for multi-agent clusters in this application.

[0063] In this embodiment, the data backup method for multi-agent clusters includes steps S10 to S30:

[0064] Step S10: Obtain the data generation rate, remaining storage capacity, and transmission capacity of the inter-node communication links for all nodes in the multi-agent cluster.

[0065] It's important to note that data generation rate can be obtained through sensor monitoring and log recording. For example, when a drone swarm performs environmental monitoring tasks, its cameras and various sensors continuously generate data. This information is recorded in local logs and analyzed through integrated algorithms. The data generation rate is specifically expressed as the amount of data generated per second / minute, such as MB / s or GB / min, as well as the proportional distribution of different data types, such as the proportion of image data and sensor data. Secondly, the remaining storage capacity can be obtained through the local query interface provided by each agent. For example, the `df -h` command can be used to check disk usage on a Linux system. Alternatively, if a distributed file system (such as Ceph) is used, the storage status of each node, including used space and remaining space, can be obtained in real time through a monitoring service. The remaining storage capacity is accurate to the byte level of available storage space, and also includes an assessment of the health status of the storage media, such as the wear and tear of SSDs. Finally, for the transmission capacity of inter-node communication links, network testing tools such as iperf and ping can be used to periodically test key indicators such as bandwidth, latency, and packet loss rate between nodes to evaluate the quality and maximum transmission capacity of the communication link. Simultaneously, based on information collected during actual data transmission, adaptive algorithms can be applied to dynamically adjust the estimated transmission capacity, automatically reducing the expected maximum transmission rate under high load conditions to avoid excessive consumption of network resources. The transmission capacity of inter-node communication links includes not only the theoretical maximum transmission rate but also the average transmission rate, latency, and reliability indicators in actual operation.

[0066] Further, step S10 includes: acquiring node operational status data through distributed sensors, including data generation rate and remaining storage capacity; and obtaining the transmission capacity of the inter-node communication link through a multi-agent cooperative communication mechanism. Specifically, each agent is equipped with multiple types of sensors that can monitor and record the data generation rate and remaining storage capacity of its node in real time. For example, in environmental monitoring tasks, cameras, thermometers, and other sensors continuously generate data and transmit it to the local processing unit for preliminary analysis. This process not only helps us understand the current data production rate of each node but also allows us to promptly grasp the status of remaining storage resources, ensuring that appropriate measures are taken before storage space is exhausted. To further optimize the data backup strategy, it is also necessary to accurately assess the transmission capacity of the inter-node communication link. This is achieved through a multi-agent cooperative communication mechanism. Specifically, each agent periodically exchanges information packets, estimating the actual transmission rate and link quality by measuring parameters such as the time delay of receiving acknowledgments and packet loss rate.

[0067] Furthermore, a target neural network model is used to analyze and adjust the operational status data and the transmission capacity of inter-node communication links to obtain the final data generation rate, remaining storage capacity, and transmission capacity of inter-node communication links for all nodes in the multi-agent cluster. Before using this target neural network model, the process includes: acquiring historical operational status data of the nodes and constructing an initial target neural network model; then, inputting the historical operational status data into the initial target neural network model for training to obtain the target neural network model. Specifically, the first step is to acquire historical operational status data of the nodes, including but not limited to key parameters such as data generation rate, remaining storage capacity, and transmission capacity of inter-node communication links. This historical data can be collected through the built-in sensors and log systems of each node and centrally managed through a distributed file system or cloud storage. Once sufficient historical data is collected, the next step is to construct the initial target neural network model. This model can use a Long Short-Term Memory (LSTM) network or a Convolutional Neural Network (CNN) suitable for time series prediction to capture the trend of data changes over time. Inputting the historical operational status data into the initial model for training involves not only learning from existing data but also adjusting the model's internal parameters to minimize prediction errors. After multiple rounds of iterative training, the optimized target neural network model is finally obtained. After obtaining the target neural network model, the running status data and the transmission capacity of the inter-node communication link are input into the target neural network model to obtain the predicted data value. This preset data value includes the predicted data generation rate of the node, the predicted remaining storage capacity, and the predicted transmission capacity of the inter-node communication link within the future time period. Then, the data backup priority of the node and the target node selection are dynamically adjusted according to the predicted data value to obtain the data generation rate, remaining storage capacity, and transmission capacity of the inter-node communication link of all nodes in the multi-agent cluster.

[0068] The key parameters obtained through the above methods can provide a solid foundation for subsequent data backup strategies. For example, the optimal data transmission path can be selected based on the transmission capacity of the inter-node communication link to reduce transmission time and improve efficiency; backup tasks can be reasonably arranged based on the data generation rate and remaining storage capacity to prevent the risk of data loss due to insufficient storage space; and when a node fails, a pre-calculated backup configuration matrix can be used to quickly locate a replacement node to ensure the security and integrity of the data.

[0069] Step S20: Based on the data generation rate, remaining storage capacity, and transmission capacity of the inter-node communication link, allocate the amount of backup data using a data integrity priority strategy or a traffic balancing strategy to obtain the allocation result.

[0070] It should be noted that allocating backup data based on the node's data generation rate, remaining storage capacity, and the transmission capacity of inter-node communication links can improve the efficiency and reliability of data management in multi-agent clusters by employing either a data integrity-first strategy or a traffic balancing strategy. These two strategies each have their own focus and are suitable for different application scenarios and needs.

[0071] Specifically, the data integrity priority strategy focuses on reducing the difficulty of data recovery. When implementing this strategy, all nodes are first sorted according to their data generation rate, and the remaining amount of data to be backed up for each node is initialized. Next, the main loop iterates through all nodes in the multi-agent cluster. For each node, its data output channels (outgoing edges) are first sorted in descending order of channel capacity, and then the node with the highest data transmission rate is selected to maximize data transmission efficiency. Once a node reaches its backup capacity limit, the system switches to the next available node with the highest remaining data transmission rate. This method dynamically adjusts the strategy priority by real-time evaluation of the maximum instantaneous resource margin, improving the overall efficiency of data transmission and ensuring maximum resource utilization. It is particularly suitable for scenarios with extremely high requirements for data integrity and security, such as emergency rescue missions. On the other hand, the traffic balancing strategy aims to increase the data backup rate while maintaining load balancing. First, the data generation rate of each node, the node's storage capacity, and the channel transmission capacity between nodes need to be initialized. Then, nodes are sorted based on their data generation rate, and the remaining storage capacity of each node is initialized. Next, the main loop iterates through all nodes, determining the set of all outgoing target nodes for each node, and calculating the amount of data to be backed up for each target node. This is typically determined using a formula to ensure that the remaining storage capacity of the target node is not exceeded. By dividing the backup data of each multi-agent system into multiple coded segments and distributing them across geographically dispersed nodes, this method effectively avoids single-link overload, improves the overall system throughput, and reduces the risk of data loss in the event of node failure.

[0072] Ultimately, the allocation results from these two strategies are input into the backup configuration matrix to guide the data backup operations between the source and target nodes. This dynamic and intelligent allocation and backup mechanism not only optimizes resource utilization and enhances the system's adaptability and robustness, but also ensures efficient data recovery even in the event of partial node failure, greatly improving the task execution capabilities and information security of the multi-agent cluster.

[0073] Step S30: Input the allocation result into the backup configuration matrix to perform data backup operations between the source node and the target node.

[0074] It should be noted that the backup configuration matrix is ​​a two-dimensional array, where the rows of the backup configuration matrix represent the source nodes, the columns of the backup configuration matrix represent the target nodes, and the elements in the matrix represent the amount of backup data from the source node to the target node.

[0075] In practice, the first step is to calculate, based on either a data integrity priority strategy or a traffic balancing strategy, which target nodes each source node needs to back up and how much data to each. These calculations are then systematically populated into the backup configuration matrix. For example, under a data integrity priority strategy, when a source node needs urgent backup due to approaching its storage limit, the strategy prioritizes target nodes with high remaining storage capacity and good communication link quality, and determines the specific amount of backup data. Under a traffic balancing strategy, backup tasks are evenly distributed through optimization algorithms to avoid overloading some nodes while maximizing the overall cluster's data backup rate.

[0076] Once the backup configuration matrix is ​​built, it can guide the actual data backup operation. The source node establishes a connection with the corresponding target node based on the information in the matrix and begins transmitting the specified amount of data. During this process, the system also monitors the data transmission status in real time, such as transmission rate, latency, and packet loss, to ensure the backup process proceeds smoothly. If any anomalies are detected (such as transmission interruption or sudden target node failure), the backup path can be quickly adjusted or backup tasks can be reassigned according to pre-defined rules to ensure data security.

[0077] Furthermore, after step S30, the process also includes: monitoring the fault status of nodes through a fault detection mechanism to obtain monitoring results; triggering the reallocation and backup operation of the allocation results when a node is found to be faulty based on the monitoring results; and adjusting the backup strategy to achieve normal backup operation when the monitoring results show no fault. Specifically, a high-precision fault detection module is deployed on each node. These modules can monitor the node's operating status in real time, including key indicators such as hardware failures, software anomalies, communication link interruptions, and node response delays, and report the monitoring results to the cluster management node in real time. When the monitoring results show that a node is faulty, the reallocation and backup operation of backup data will be triggered immediately. This process involves quickly identifying the faulty node, recording the fault type and time, dynamically adjusting the backup strategy according to the current cluster status, reallocating backup tasks, selecting other available nodes as new target nodes, and updating the backup configuration matrix to ensure the continuity and integrity of data backup. At the same time, corresponding recovery measures will be taken according to the severity and type of the fault, such as automatically restarting the node, repairing the communication link, or reconfiguring node resources. When monitoring indicates that no faults exist in the cluster, the backup strategy is dynamically adjusted based on the current cluster status and resource utilization. This optimizes the backup path and backup data volume to achieve efficient and reliable normal backup operations. This process includes real-time assessment of each node's remaining storage capacity, data generation rate, and communication link transmission capacity. A data integrity-first or traffic-balancing strategy is employed to ensure the efficiency of the backup operation and the rationality of resource utilization. Simultaneously, the backup process is monitored in real-time to ensure its successful completion. Through this fault detection and dynamic adjustment mechanism, various fault conditions can be flexibly addressed in complex and ever-changing multi-agent cluster environments, ensuring the reliability and stability of data backup while optimizing the backup strategy and improving the overall performance and adaptability of the system.

[0078] Furthermore, the backup configuration matrix is ​​designed to support dynamic updates, meaning that as tasks progress and the environment changes, the element values ​​in the matrix can be adjusted in real time to reflect the latest backup requirements. This allows the entire system to flexibly respond to unexpected situations, such as the addition of new nodes or the failure of existing nodes, further enhancing the system's robustness and adaptability.

[0079] This embodiment acquires the data generation rate, remaining storage capacity, and inter-node communication link transmission capacity of all nodes in a multi-agent cluster. Based on these factors, it allocates backup data using either a data integrity priority strategy or a traffic balancing strategy. The allocation result is then input into a backup configuration matrix to execute data backup operations between the source and target nodes. By acquiring the data generation rate, remaining storage capacity, and inter-node communication link transmission capacity of each source node in the multi-agent cluster, and dynamically allocating backup data using either a data integrity priority or traffic balancing strategy, and inputting the result into the backup configuration matrix to execute data backup, this approach improves the efficiency and reliability of data backup in complex environments, ensures data recovery in the event of node failure, and enhances system adaptability and robustness.

[0080] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in Embodiment 1 above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 2 The data backup method for multi-agent clusters, step S20, further includes steps S201 to S205:

[0081] Step S201: Initialize the remaining amount of data to be backed up to the total amount of original backup data for the corresponding node.

[0082] It's important to note that the remaining data to be backed up on each node is initialized to its original total backup data volume, i.e., the total amount of data the node needs to back up at the initial moment. This initialization process is uniformly coordinated by the cluster management module, ensuring that the data volume information of all nodes is accurately recorded and synchronized to the global status table. Specifically, a heartbeat protocol is used to periodically collect metadata from each node, including data generation rate, storage capacity, and the current amount of data to be backed up. For newly joined nodes, their initial data volume is dynamically allocated by the task scheduling system, and global consistency is ensured through a distributed consensus algorithm (such as the Raft protocol). Furthermore, to cope with data changes in a dynamic environment, the system employs an incremental update mechanism, monitoring the node data generation and backup progress in real time and dynamically adjusting the value of the original total backup data volume, thereby ensuring the efficient execution of the backup strategy.

[0083] Step S202: Sort all nodes in descending order according to the data generation rate to obtain a sorted list of nodes.

[0084] It should be noted that the nodes are sorted in descending order based on their data generation rate to generate the sorted list. This process first requires real-time monitoring and collection of the data generation rate information for each node. The data generation rate reflects the amount of data generated by a node per unit of time and is an important indicator for measuring the importance of node data and backup priority.

[0085] To achieve accurate sorting, efficient sorting algorithms, such as quicksort or mergesort, are employed to ensure rapid sorting operations even in large-scale cluster environments. The resulting list of nodes after sorting is... The system clearly prioritizes each node, placing nodes with higher data generation rates at the top. This means they will be processed first during the backup process. This data generation rate-based sorting strategy ensures that critical data is backed up in a timely manner, thereby improving the overall system's data backup efficiency and reliability, and laying a solid foundation for subsequent backup operations.

[0086] Step S203: Traverse the sorted list of nodes and sort the transmission capacity of the communication links of all nodes in descending order of channel capacity to obtain the sorted set of edges.

[0087] It should be noted that when traversing the sorted list of nodes, all communication links (i.e., outgoing edges) of each node are sorted in descending order according to their transmission capacity, thus obtaining the sorted set of edges. This process first requires real-time monitoring and collection of communication link transmission capacity information for each node. The transmission capacity of the communication link directly determines the efficiency of data backup. Therefore, by sorting the links in descending order of channel capacity, links with larger capacities can be prioritized for data backup, thereby improving backup speed and reducing backup time.

[0088] During the sorting process, efficient sorting algorithms, such as quicksort or mergesort, are employed to ensure rapid sorting operations in large-scale cluster environments. After sorting, the resulting edge set clearly defines the communication link priority order for each node, with links having higher transmission capacity listed first. This channel capacity-based sorting strategy ensures that efficient communication links are prioritized during backup, thereby improving the overall data backup efficiency and reliability of the system and providing crucial support for achieving efficient data backup.

[0089] Step S204: Traverse the sorted edge set and calculate the first backup data volume of the corresponding node.

[0090] It should be noted that the edge set after traversal and sorting... At that time, for each node and its corresponding communication link, the amount of the first backup data of that node on the target node is calculated. This process requires comprehensive consideration of several key factors, including the total amount of original backup data of the current node, the remaining storage capacity of the target node, and the transmission capacity of the communication link between the current node and the target node.

[0091] First, a portion of the remaining data to be backed up on each node is selected as candidate values ​​for the backup data volume. Then, the remaining storage capacity of the target node is checked to ensure it has sufficient space to receive the backup data. Simultaneously, the transmission capacity of the communication link is considered to ensure data transmission can be completed within a reasonable timeframe. Finally, for each target node... The minimum of these three values ​​is taken as the first backup data size, i.e.:

[0092]

[0093] in, This represents the remaining storage capacity of the target node. This represents the transmission capacity of the communication link between the current node and the target node. This represents the total amount of the original backup data. For the current node, For the target node, For time.

[0094] This approach ensures that the amount of backup data on each node does not exceed its own data generation capacity, nor does it exceed the storage capacity of the target node or the transmission capacity of the communication link. This precise calculation method not only improves the efficiency of data backup but also ensures the stability and reliability of the backup process. Furthermore, by dynamically adjusting the amount of backup data, it can better adapt to the dynamic changes in nodes and communication links within a multi-agent cluster, thereby achieving efficient and reliable data backup.

[0095] Step S205: Use the first backup data volume as the first allocation result.

[0096] It should be noted that after traversing the sorted edge set and calculating the first backup data volume for each node on the target node, these backup data volumes are recorded as the first allocation result. This process not only provides a clear task allocation for subsequent backup operations but also provides basic data for the dynamic adjustment and optimization of the system. The first backup data volume is written into the backup configuration matrix, where each row represents a source node, each column represents a target node, and the elements in the matrix represent the backup data volume from the source node to the target node. In this way, the backup configuration matrix can clearly record the backup task allocation for each node, ensuring that each node can perform backup operations according to the allocation result.

[0097] Furthermore, the subsequent backup strategy is dynamically adjusted based on the initial allocation result. If anomalies such as communication link failures or insufficient storage capacity of the target node are detected during the backup process, the amount of backup data is recalculated based on real-time monitoring data, and the backup configuration matrix is ​​updated. This dynamic adjustment mechanism ensures the continuity and reliability of the backup process, enabling efficient data backup even in complex multi-agent cluster environments. In this way, the system not only improves backup efficiency but also enhances its fault tolerance and adaptability.

[0098] This embodiment initializes the remaining backup data volume of each node to the total original backup data volume, sorts the nodes in descending order by data generation rate, and sorts the communication links in descending order by channel capacity. The first backup data volume is then calculated and used as the first allocation result. A data integrity-first strategy is adopted to ensure efficient and reliable data backup, improving overall system performance and data security.

[0099] Based on the first embodiment of this application, in the third embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 3 The data backup step S20 for multi-agent clusters further includes steps S301 to S305:

[0100] Step S301: Initialize the data generation rate, remaining storage capacity, and transmission capacity of the inter-node communication links for all nodes.

[0101] It's important to note that before the backup operation begins, key parameters for each node in the cluster are initialized. These parameters include the node's data generation rate, remaining storage capacity, and the transmission capacity of the inter-node communication links. The data generation rate reflects the amount of data generated by each node per unit of time, a crucial parameter for determining the priority of backup tasks. Remaining storage capacity indicates the amount of space available for storing backup data on each node, directly affecting the storage location and quantity of backup data. The transmission capacity of the communication links relates to the efficiency of data transmission between nodes, determining the speed and stability of the backup operation. To accurately obtain these parameters, monitoring modules are deployed on each node. These modules can collect and report node operating status information in real time. Simultaneously, the transmission status of the inter-node communication links is monitored and acquired in real time through network management protocols or dedicated communication mechanisms. These measures ensure that the initialized parameters accurately reflect the actual operating status of the cluster, thus providing a reliable basis for subsequent backup strategy formulation and task allocation.

[0102] Step S302: Sort all nodes in descending order according to the node data generation rate to obtain a sorted list of nodes.

[0103] It's important to note that data generation rate is a crucial metric for measuring the amount of data a node generates per unit of time, directly impacting its backup resource requirements. Therefore, sorting nodes by data generation rate ensures that nodes with higher rates are prioritized for backup, thereby improving overall system backup efficiency and data security.

[0104] Specifically, the data generation rate information of each node is first monitored and collected in real time. This information is acquired through a distributed sensor network or monitoring modules on the nodes and synchronized to the cluster management node in real time. To ensure the accuracy and efficiency of the sorting, an efficient sorting algorithm, such as quicksort or mergesort, is used to sort all nodes in descending order of their data generation rate. After sorting, the resulting node list is... The priority order of each node is clearly indicated, with nodes having a higher data generation rate listed first.

[0105] This data generation rate-based sorting strategy not only ensures timely backup of critical data but also optimizes the allocation of backup resources. In dynamic environments, the data generation rate of nodes may change, so the data is periodically reordered to adapt to dynamic changes in the cluster. This approach allows for flexible handling of various situations, ensuring that the backup process remains efficient and reliable at all times.

[0106] Step S303: Traverse the sorted list of nodes and process all nodes in turn to obtain the set of all outgoing edge endpoint nodes.

[0107] It should be noted that after sorting the nodes in descending order of data generation rate, each node is processed one by one to determine the target node set for its backup data. For each node, all outgoing edges are first obtained; these outgoing edges represent the communication links through which the current node can transmit data. The endpoint node of each outgoing edge is a potential target node, which will receive the backup data from the current node. By traversing these outgoing edges, information about all endpoint nodes is collected, forming a set of endpoint nodes. This set provides a clear target node range for subsequent backup data allocation.

[0108] In practice, the set of endpoint nodes is further filtered and optimized based on the transmission capacity of the communication link and the remaining storage capacity of the target node. This ensures that backup data from each node can be efficiently and reliably transmitted to the appropriate target node. This method of dynamically acquiring the set of endpoint nodes not only improves the flexibility of backup but also adapts to the dynamic changes in nodes and communication links within a multi-agent cluster, thereby achieving efficient and reliable distributed data backup.

[0109] Step S304: Traverse the set of endpoint nodes to obtain the second backup data volume of the corresponding node.

[0110] It should be noted that after obtaining the set of all outgoing endpoint nodes for each node, these endpoint nodes are further processed to calculate the amount of backup data for each node on each target node. For each source node, its set of endpoint nodes is traversed sequentially, and for each target node... This involves calculating the amount of backup data from the source node to the target node. This calculation requires considering several key factors, including the remaining amount of data to be backed up at the source node, the remaining storage capacity of the target node, and the transmission capacity of the communication link. The specific formula is:

[0111]

[0112] in, For the set of endpoint nodes, This represents the remaining storage capacity of the target node. This represents the transmission capacity of the communication link between the current node and the target node. This represents the total amount of the original backup data. For the current node, For the target node, This method ensures that the amount of backup data on each node does not exceed its own data generation capacity, nor does it exceed the storage capacity of the target node or the transmission capacity of the communication link. This precise calculation method not only improves the efficiency of data backup but also ensures the stability and reliability of the backup process. Furthermore, by dynamically adjusting the amount of backup data, it can better adapt to the dynamic changes of nodes and communication links in a multi-agent cluster, thereby achieving efficient and reliable data backup.

[0113] Step S305: Use the second backup data volume as the second allocation result.

[0114] It should be noted that after traversing the set of endpoint nodes and calculating the second backup data volume for each node on the target node, these backup data volumes are recorded as the second allocation result. This process not only provides a clear task allocation for subsequent backup operations but also provides basic data for dynamic adjustment and optimization. The second backup data volume is written into the backup configuration matrix, where each row represents a source node, each column represents a target node, and the elements in the matrix represent the backup data volume from the source node to the target node. In this way, the backup configuration matrix can clearly record the backup task allocation for each node, ensuring that each node can perform backup operations according to the allocation result.

[0115] Furthermore, the subsequent backup strategy will be dynamically adjusted based on the second allocation result. If anomalies such as communication link failure or insufficient storage capacity of the target node are detected during the backup process, the backup data volume will be recalculated based on real-time monitoring data, and the backup configuration matrix will be updated. Update and back up the configuration matrix Meanwhile, the remaining storage capacity of the nodes is updated according to the formula. Updates are performed. This dynamic adjustment mechanism ensures the continuity and reliability of the backup process, enabling efficient data backup even in complex multi-agent cluster environments. This approach not only improves backup efficiency but also enhances the system's fault tolerance and adaptability.

[0116] This embodiment initializes the node data generation rate, remaining storage capacity, and communication link transmission capacity, sorts nodes by data generation rate, obtains the set of outgoing edge endpoint nodes, calculates the second backup data volume, and uses it as the second allocation result. A traffic balancing strategy is adopted to optimize the backup path, improve backup efficiency and resource utilization, and enhance system reliability.

[0117] Based on the first embodiment of this application, in the fourth embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 4 The data backup method for multi-agent clusters, step S30, further includes steps S401 to S404:

[0118] Step S401: The allocation result is divided according to the backup configuration matrix to obtain multiple backup data blocks.

[0119] It's important to note that after allocating the backup data, the backup data for each node is divided into multiple smaller data blocks based on the information in the backup configuration matrix. This partitioning method not only improves data transmission efficiency but also enhances data fault tolerance. Each backup data block contains a portion of the original backup data and includes necessary verification information to quickly verify data integrity and accuracy during data recovery. Furthermore, the size of each data block is dynamically adjusted based on the target node's storage capacity and the transmission characteristics of the communication link, ensuring efficient network transmission and avoiding transmission delays or failures caused by excessively large data blocks.

[0120] Furthermore, step S401 also includes: performing redundant encoding on multiple backup data blocks to obtain encoded data blocks; storing the encoded data blocks so that if errors occur in the data within the backup data blocks, the encoded data blocks can be decoded to restore the original backup data. Specifically, during the segmentation process, redundant encoding of the data is also considered, encoding the data blocks into a form with a certain degree of redundancy to further improve the data's fault tolerance. Even if some data blocks are lost or damaged during transmission, the original data can be recovered through redundant encoding, thereby ensuring the integrity and reliability of the backup data.

[0121] By dividing the allocation results into multiple backup data blocks, the system can better adapt to the dynamic changes in nodes and communication links within a multi-agent cluster, improving the flexibility and reliability of the backup process. This partitioning strategy not only optimizes the efficiency of data backup but also enhances the system's fault tolerance and data recovery capabilities.

[0122] Step S402: Send the backup data block from the source node to the corresponding target node through a preset communication protocol.

[0123] It should be noted that after the backup data blocks are divided and encoded, they will be sent one by one to the target node according to the allocation results in the backup configuration matrix through a preset communication protocol.

[0124] The default communication protocol is specifically designed for multi-agent cluster environments, ensuring efficient data transmission between different nodes. This protocol supports multiple communication links and can dynamically adjust the data transmission rate based on the link's capacity, thereby optimizing transmission efficiency. Furthermore, the protocol includes error detection and retransmission mechanisms, enabling it to detect erroneous or lost data blocks during transmission and automatically trigger retransmissions to ensure data integrity and accuracy.

[0125] During data transmission, the source node attaches verification information to each backup data block, and the target node performs integrity verification upon receiving the data block. If the verification passes, the target node stores the data block in a designated location and sends an acknowledgment signal to the source node; if the verification fails, the target node notifies the source node to resend the data block. In this way, the system ensures that each backup data block is accurately transmitted to the target node, thereby improving the reliability and stability of the entire backup process.

[0126] Step S403: When the target node receives the backup data block, it performs data integrity verification and obtains the verification result.

[0127] It should be noted that upon receiving the backup data block from the source node, the target node immediately initiates a data integrity verification process. This process verifies the integrity and accuracy of the data block using a checksum (such as a CRC checksum, MD5 hash, or SHA-256 hash) pre-attached to the verification information. The target node first reads the verification information from the backup data block, then recalculates the checksum of the data block according to the algorithm in the verification information. The recalculated checksum is compared with the checksum carried in the data block to obtain the verification result, and then subsequent backup operations are performed.

[0128] Step S404: Perform the corresponding data backup operation between the source node and the target node based on the verification result.

[0129] It should be noted that after the target node completes the data integrity verification, it will determine the subsequent operations based on the verification results.

[0130] Specifically, when the verification result is successful, the backup data block is stored in the target node, and an acknowledgment signal is sent back to the source node. In other words, if the verification result indicates that the data block is complete and accurate, the target node will store the data block in the designated storage location and send an acknowledgment signal to the source node. Upon receiving the acknowledgment signal, the source node will mark the backup task as complete and remove it from the backup task queue. Simultaneously, the source node will update its own backup status information, recording completed backup tasks to ensure the continuity and integrity of the backup process.

[0131] Furthermore, when the verification result is a failure, a resend signal is sent to the source node to resend the data block. If the verification result indicates that the data block is erroneous or corrupted, the target node records the identification information of the data block and sends an error notification to the source node. Upon receiving the error notification, the source node resends the corresponding data block. During the resend process, the source node appends new verification information so that the target node can perform data integrity verification again. This process is repeated until the data block is successfully received and stored. Finally, when the source node receives an acknowledgment signal, it prompts that the data backup operation is complete. This prompt can be in the form of system log records, user interface prompts, or notification messages, ensuring that system administrators or users can understand the completion status of the backup task in real time. The source node further updates the status information of the backup task, recording detailed information such as the completion time of the backup task and the storage location of the target node. This information is crucial for subsequent data management and recovery operations. The source node also dynamically adjusts subsequent backup strategies based on the completion status of the backup task. For example, if a target node frequently experiences verification failures, the system may adjust the backup path and select other more reliable nodes as backup targets.

[0132] This embodiment divides the allocation result into multiple backup data blocks according to the backup configuration matrix and sends them from the source node to the target node via a preset communication protocol. Upon receiving the data, the target node performs data integrity verification and executes the backup operation based on the verification result. This method improves backup efficiency and reliability, enhances system fault tolerance, and ensures data integrity and stability.

[0133] Based on the first embodiment of this application, this application also provides a data backup device for multi-agent clusters. Please refer to... Figure 5 The device includes:

[0134] The acquisition module 10 is used to acquire the data generation rate, remaining storage capacity, and transmission capacity of the inter-node communication links of all nodes in the multi-agent cluster.

[0135] The allocation module 20 is used to allocate the amount of backup data according to the data generation rate, remaining storage capacity and the transmission capacity of the inter-node communication link, using a data integrity priority strategy or a traffic balancing strategy, and obtain the allocation result.

[0136] The execution module 30 is used to input the allocation result into the backup configuration matrix to perform data backup operations between the source node and the target node. The rows of the backup configuration matrix represent the source node, the columns of the backup configuration matrix represent the target node, and the elements in the matrix represent the amount of backup data from the source node to the target node.

[0137] The data backup device for multi-agent clusters provided in this application, employing the data backup method for multi-agent clusters in the above embodiments, can solve the technical problem of how to improve the stability of multi-agent cluster data in complex environments. Compared with the prior art, the beneficial effects of the data backup device for multi-agent clusters provided in this application are the same as those of the data backup method for multi-agent clusters provided in the above embodiments, and other technical features in the data backup device for multi-agent clusters are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0138] In one embodiment, the allocation module 20 is further configured to initialize the remaining amount of data to be backed up as the total amount of original backup data for the corresponding node; sort all nodes in descending order according to the data generation rate to obtain a sorted node list; traverse the sorted node list and sort the transmission capacity of the communication links of all nodes in descending order according to the channel capacity to obtain a sorted edge set; traverse the sorted edge set and calculate the first backup data amount for the corresponding node; and use the first backup data amount as the first allocation result, which is obtained by processing with a data integrity priority strategy.

[0139] In one embodiment, the allocation module 20 is further configured to initialize the data generation rate, remaining storage capacity, and transmission capacity of the inter-node communication links of all nodes; sort all nodes in descending order according to their data generation rate to obtain a sorted node list; traverse the sorted node list and process all nodes sequentially to obtain the set of endpoint nodes for all outgoing edges; traverse the set of endpoint nodes to obtain the second backup data volume of the corresponding node; and use the second backup data volume as the second allocation result, which is obtained by processing using a traffic balancing strategy.

[0140] In one embodiment, the acquisition module 10 is further configured to acquire the operating status data of the nodes through distributed sensors, the operating status data including the data generation rate and the remaining storage capacity; obtain the transmission capacity of the inter-node communication link through the cooperative communication mechanism between multiple agents; and analyze and adjust the operating status data and the transmission capacity of the inter-node communication link using a target neural network model to obtain the final data generation rate, remaining storage capacity and transmission capacity of the inter-node communication link of all nodes in the multi-agent cluster.

[0141] In one embodiment, the acquisition module 10 is further configured to acquire historical operating status data of nodes and construct an initial target neural network model; input the historical operating status data into the initial target neural network model for training to obtain a target neural network model; after obtaining the target neural network model, the module further includes: inputting the operating status data and the transmission capacity of the inter-node communication link into the target neural network model to obtain predicted data values, wherein the preset data values ​​include the data generation rate, remaining storage capacity, and transmission capacity of the inter-node communication link of the node in a future time period; dynamically adjusting the data backup priority of the node and the target node selection based on the predicted data values ​​to obtain the data generation rate, remaining storage capacity, and transmission capacity of the inter-node communication link of all nodes in the multi-agent cluster.

[0142] In one embodiment, the execution module 30 is further configured to divide the allocation result according to the backup configuration matrix to obtain multiple backup data blocks; send the backup data blocks from the source node to the corresponding target node through a preset communication protocol; when the target node receives the backup data blocks, perform data integrity verification and obtain the verification result; and perform the corresponding data backup operation between the source node and the target node according to the verification result.

[0143] In one embodiment, the execution module 30 is further configured to store the backup data block to the target node and send a confirmation signal to the source node when the verification result is a successful verification; when the verification result is a failed verification, send a retransmission signal to the source node to make the source node retransmit the data block; and when the source node receives the confirmation signal, prompt that the data backup operation is complete.

[0144] This application provides a data backup device for a multi-agent cluster, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to execute the data backup method for a multi-agent cluster described in Embodiment 1 above.

[0145] The following is for reference. Figure 6 This document illustrates a structural diagram of a data backup device suitable for implementing embodiments of this application for a multi-agent cluster. The data backup device for a multi-agent cluster in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 6 The data backup device for multi-agent clusters shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0146] like Figure 6As shown, a data backup device for a multi-agent cluster may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the data backup device for the multi-agent cluster. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows data backup devices for multi-agent clusters to exchange data wirelessly or via wired communication with other devices. Although various data backup devices for multi-agent clusters are shown in the figures, it should be understood that implementation or possession of all of them is not required. More or fewer may be implemented alternatively.

[0147] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0148] The data backup device for multi-agent clusters provided in this application, employing the data backup method for multi-agent clusters in the above embodiments, can solve the technical problem of how to improve the stability of multi-agent cluster data in complex environments. Compared with the prior art, the beneficial effects of the data backup device for multi-agent clusters provided in this application are the same as those of the data backup method for multi-agent clusters provided in the above embodiments, and other technical features in this data backup device for multi-agent clusters are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0149] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0150] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0151] This application provides a computer-readable medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the data backup method for multi-agent clusters in the above embodiments.

[0152] The computer-readable medium provided in this application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor devices, or any combination thereof. More specific examples of computer-readable media may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable medium may be any tangible medium containing or storing a program that can be executed by instructions, used by a device, or used in conjunction with it. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0153] The aforementioned computer-readable medium may be included in a data backup device for a multi-agent cluster; or it may exist independently and not be assembled into a data backup device for a multi-agent cluster.

[0154] The aforementioned computer-readable medium carries one or more programs that, when executed by a data backup device for a multi-agent cluster, enable the data backup device to write computer program code for performing the operations of this application in one or more programming languages ​​or a combination thereof. These programming languages ​​include object-oriented programming languages—such as Java, Smalltalk, and C++—and conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0155] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and computer program products according to various embodiments of this application. In this regard, all blocks in the flowcharts or block diagrams may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that all blocks in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using dedicated hardware-based implementations that perform the specified functions or operations, or using a combination of dedicated hardware and computer instructions.

[0156] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0157] The readable medium provided in this application is a computer-readable medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described data backup method for multi-agent clusters, thereby solving the technical problem of how to improve the stability of multi-agent cluster data in complex environments. Compared with the prior art, the beneficial effects of the computer-readable medium provided in this application are the same as those of the data backup method for multi-agent clusters provided in the above embodiments, and will not be repeated here.

[0158] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the data backup method for multi-agent clusters as described above.

[0159] The computer program product provided in this application can solve the technical problem of how to improve the stability of multi-agent cluster data in complex environments. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the data backup method for multi-agent clusters provided in the above embodiments, and will not be repeated here.

[0160] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A data backup method for multi-agent clusters, characterized in that, The method includes: Obtain the data generation rate, remaining storage capacity, and transmission capacity of inter-node communication links for all nodes in a multi-agent cluster; Based on the data generation rate, the remaining storage capacity, and the transmission capacity of the inter-node communication link, the amount of backup data is allocated using a data integrity priority strategy or a traffic balancing strategy to obtain the allocation result; The allocation result is input into the backup configuration matrix to perform data backup operations between the source node and the target node. The rows of the backup configuration matrix represent the source node, the columns of the backup configuration matrix represent the target node, and the elements in the matrix represent the amount of backup data from the source node to the target node. The step of allocating backup data based on the data generation rate, the remaining storage capacity, and the transmission capacity of the inter-node communication link using a data integrity priority strategy or a traffic balancing strategy to obtain the allocation result includes: The remaining storage capacity is initialized to the total amount of original backup data for the corresponding node; All nodes are sorted in descending order according to the data generation rate to obtain a sorted list of nodes. Traverse the sorted list of nodes and sort the transmission capacity of the communication links of all nodes in descending order of channel capacity to obtain the sorted set of edges. Traverse the sorted set of edges and calculate the first backup data volume of the corresponding node. The first backup data volume is obtained by comparing the total original backup data volume of the current node, the remaining storage capacity of the target node, and the transmission capacity of the communication link between the current node and the target node. The specific formula is as follows: in, This represents the remaining storage capacity of the target node. The transmission capacity of the communication link between the current node and the target node. This represents the total amount of the original backup data. For the current node, For the target node, For time; The first backup data volume is used as the first allocation result, which is obtained by processing with a data integrity priority strategy. The step of allocating the backup data volume according to the data generation rate, the remaining storage capacity, and the transmission capacity of the inter-node communication link using a data integrity priority strategy or a traffic balancing strategy to obtain the allocation result further includes: Initialize the data generation rate, remaining storage capacity, and transmission capacity of the inter-node communication links for all nodes; Sort all nodes in descending order according to the data generation rate to obtain a sorted list of nodes. Traverse the sorted list of nodes and process all nodes in turn to obtain the set of all outgoing edge endpoint nodes; Traverse the set of endpoint nodes to obtain the second backup data volume of the corresponding node, using the following formula: in, For the set of endpoint nodes, This represents the remaining storage capacity of the target node. The transmission capacity of the communication link between the current node and the target node. This represents the total amount of the original backup data. For the current node, For the target node, For time; The second backup data volume is used as the second allocation result, which is obtained by processing with a traffic balancing strategy; The step of inputting the allocation result into the backup configuration matrix to perform data backup operations between the source node and the target node includes: The allocation result is divided according to the backup configuration matrix to obtain multiple backup data blocks; The backup data block is sent from the source node to the corresponding target node using a preset communication protocol; When the target node receives the backup data block, it performs a data integrity check and obtains the check result. Based on the verification result, perform the corresponding data backup operation between the source node and the target node.

2. The method as described in claim 1, characterized in that, The steps of obtaining the data generation rate, remaining storage capacity, and transmission capacity of the inter-node communication links of all nodes in the multi-agent cluster include: The node's operational status data is acquired through distributed sensors, and the operational status data includes data generation rate and remaining storage capacity; The transmission capacity of the inter-node communication link is obtained through a cooperative communication mechanism among multiple agents. The target neural network model is used to analyze and adjust the running status data and the transmission capacity of the inter-node communication links to obtain the data generation rate, remaining storage capacity and transmission capacity of the inter-node communication links of all nodes in the final multi-agent cluster.

3. The method as described in claim 2, characterized in that, Before the step of analyzing and adjusting the operational status data and the transmission capacity of the inter-node communication links using the target neural network model to obtain the final data generation rate, remaining storage capacity, and transmission capacity of the inter-node communication links for all nodes in the multi-agent cluster, the method further includes: Obtain the historical operating status data of the node and construct an initial target neural network model; The historical operating status data is input into the initial target neural network model for training to obtain the target neural network model; After obtaining the target neural network model, the process further includes: The running status data and the transmission capacity of the inter-node communication link are input into the target neural network model to obtain the predicted data value. The predicted data value includes the predicted data generation rate of the node, the predicted remaining storage capacity, and the predicted transmission capacity of the inter-node communication link in the future time period. Based on the predicted data values, the data backup priority and target node selection of the nodes are dynamically adjusted to obtain the data generation rate, remaining storage capacity, and transmission capacity of the inter-node communication links of all nodes in the multi-agent cluster.

4. The method as described in claim 1, characterized in that, The step of performing the corresponding data backup operation between the source node and the target node based on the verification result includes: When the verification result is successful, the backup data block is stored in the target node, and a confirmation signal is sent back to the source node. When the verification result is a failure, a resend signal is sent to the source node to cause the source node to resend the data block. When the source node receives a confirmation signal, it prompts the user to complete the data backup operation.

5. A data backup device for multi-agent clusters, characterized in that, The data backup device for multi-agent clusters executes the data backup method for multi-agent clusters according to any one of claims 1 to 4, and the device comprises: The acquisition module is used to acquire the data generation rate, remaining storage capacity, and transmission capacity of the inter-node communication links of all nodes in the multi-agent cluster. The allocation module is used to allocate the amount of backup data according to the data generation rate, the remaining storage capacity, and the transmission capacity of the inter-node communication link, using a data integrity priority strategy or a traffic balancing strategy, and obtain the allocation result. The execution module is used to input the allocation result into the backup configuration matrix to perform data backup operations between the source node and the target node. The rows of the backup configuration matrix represent the source node, the columns of the backup configuration matrix represent the target node, and the elements in the matrix represent the amount of backup data from the source node to the target node.

6. A data backup device for multi-agent clusters, characterized in that, The device includes: a memory, a processor, and a data backup program for a multi-agent cluster stored in the memory and running on the processor, the data backup program for a multi-agent cluster being configured to implement the steps of the data backup method for a multi-agent cluster as described in any one of claims 1-4.

7. A storage medium, characterized in that, The storage medium stores a data backup program for multi-agent clusters, which, when executed by a processor, implements the steps of the data backup method for multi-agent clusters as described in any one of claims 1-4.

Images