Blockchain network dynamic scheduling method and system based on machine learning

Abstract

The present application provides a method and system for dynamic scheduling of blockchain networks based on machine learning. First, according to the dynamic scheduling tasks generated by the blockchain cluster, multiple blockchain nodes to be scheduled are selected from the blockchain cluster. When a node whose performance needs to be optimized is identified, the node group to which it belongs is determined by its node authenticator. These node groups are composed of nodes with a collaborative working relationship, which is convenient for unified management and optimization. If the scheduling group to which the target node belongs contains multiple nodes working in collaboration, the method will further obtain these related nodes and use machine learning algorithms to perform performance prediction and dynamic scheduling optimization on them. In this way, the performance bottlenecks in the blockchain network can be intelligently identified and optimized, and the operating efficiency and stability of the entire network can be improved. At the same time, it can also adapt to the complex and changeable blockchain network environment and realize the rational allocation and efficient use of resources.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and more specifically, to a blockchain network dynamic scheduling method and system based on machine learning. Background Art

[0002] In the current field of blockchain technology, with the continuous expansion of blockchain network scale and the increasing diversification of application scenarios, how to efficiently manage and schedule blockchain nodes to ensure network stability and performance optimization has become an urgent problem to be solved. Traditional blockchain node scheduling methods are often based on simple load balancing or polling algorithms. Although these methods can achieve uniform distribution of nodes to a certain extent, they lack in-depth analysis and optimization of node performance.

[0003] Especially when faced with complex and ever-changing dynamic scheduling tasks, traditional methods often cannot accurately identify the performance differences and collaborative working relationships between nodes, leading to problems such as uneven resource allocation, network congestion and performance bottlenecks. In addition, with the continuous development of blockchain technology, the collaborative working relationship between nodes has become more and more complex. How to achieve efficient scheduling and optimization of nodes in this environment has become a challenging technical problem. Summary of the invention

[0004] In view of this, the purpose of this application is to provide a blockchain network dynamic scheduling method and system based on machine learning, which can intelligently identify the performance characteristics and collaborative working relationships of blockchain nodes, and perform performance prediction and dynamic scheduling optimization of nodes through machine learning algorithms, thereby improving the operating efficiency and response speed of the blockchain network, and effectively avoiding resource waste and network congestion. That is, through intelligent dynamic scheduling strategies, it aims to solve the problem that traditional methods cannot accurately identify and optimize blockchain node performance, thereby improving the performance and stability of the entire blockchain network.

[0005] Based on the first aspect of the present application, a blockchain network dynamic scheduling method based on machine learning is provided, the method comprising:

[0006] Based on the dynamic scheduling task generated for the blockchain cluster, a plurality of blockchain nodes to be scheduled are obtained from the blockchain cluster;

[0007] When a first blockchain node whose performance is to be optimized is obtained from among the multiple blockchain nodes to be scheduled, a node authenticator corresponding to the first blockchain node is obtained, and a node group to which the first blockchain node belongs is determined based on the node authenticator, wherein the node group includes one or more blockchain nodes belonging to a scheduling group, and the scheduling group is composed of one blockchain node or multiple blockchain nodes in a collaborative working relationship;

[0008] When the target scheduling group to which the first blockchain node belongs is composed of the multiple blockchain nodes that have a collaborative working relationship, one or more second blockchain nodes recorded in the node group to which the first blockchain node belongs are obtained, and a machine learning algorithm is used to perform performance prediction and dynamic scheduling optimization on the first blockchain node and the one or more second blockchain nodes, and the second blockchain node is located in the target scheduling group.

[0009] In a possible implementation of the first aspect, obtaining a node authenticator corresponding to the first blockchain node, and determining, based on the node authenticator, a node group to which the first blockchain node belongs, includes:

[0010] Before obtaining the node authenticator corresponding to the first blockchain node and determining the node group to which the first blockchain node belongs based on the node authenticator, the method further includes:

[0011] Obtain a task scheduling center associated with the blockchain cluster, wherein the task scheduling center is used to record each preconfigured node group matched by the blockchain cluster;

[0012] The obtaining a node authenticator corresponding to the first blockchain node, and determining the node group to which the first blockchain node belongs based on the node authenticator, includes:

[0013] Obtain the node authenticator, and obtain from the task scheduling center the preconfigured node group to which the first blockchain node belongs and which is associated with the node authenticator.

[0014] In a possible implementation of the first aspect, before the dynamic scheduling task generated for the blockchain cluster obtains a plurality of blockchain nodes to be scheduled from the blockchain cluster, the method further includes:

[0015] Obtain multiple blockchain nodes to be evaluated in the blockchain cluster;

[0016] Determine a collaborative working relationship corresponding to each blockchain node among the multiple blockchain nodes to be evaluated;

[0017] Based on the corresponding collaborative working relationship, the node authenticators based on the various blockchain nodes are divided into the pre-configured node groups, and the node authenticators corresponding to the blockchain nodes located in the same scheduling group among the multiple blockchain nodes to be evaluated are divided into the same pre-configured node group.

[0018] In a possible implementation of the first aspect, the step of determining a collaborative working relationship corresponding to each blockchain node in the multiple blockchain nodes to be evaluated includes:

[0019] Continuously collect network communication data between various blockchain nodes through network monitoring tools, and the network communication data includes the amount of transmitted data, transmission frequency, and transmission delay;

[0020] Analyzing the network communication data to determine the communication density and communication stability between blockchain nodes, and generating an initial blockchain node sequence that may have a collaborative working relationship based on the communication density and communication stability between the determined blockchain nodes;

[0021] Obtain all node activity data of each initial blockchain node in the initial blockchain node sequence, wherein the node activity data includes block generation activity data, verification activity data, and propagation activity data;

[0022] Analyzing the node activity data to determine the characteristics of the mutual cooperation pattern of each initial blockchain node when processing tasks;

[0023] Each initial blockchain node is regarded as a vertex in the graph, the characteristics of the mutual cooperation pattern between nodes are regarded as edges, and weights are assigned to the edges in the graph, and the weights represent the strength of the cooperation between nodes;

[0024] The graph is divided by using a community detection algorithm to identify node communities that satisfy preset collaborative working relationships, and the collaborative working relationships corresponding to each blockchain node in the multiple blockchain nodes to be evaluated are output based on the node communities that satisfy the preset collaborative working relationships. The nodes in each node community are regarded as a node group with a reinforced collaborative working relationship.

[0025] In a possible implementation of the first aspect, the using a machine learning algorithm to perform performance prediction and dynamic scheduling optimization on the first blockchain node and the one or more second blockchain nodes includes at least one of the following:

[0026] When a first operation performance indicator corresponding to the first blockchain node and a second operation performance indicator corresponding to each of the one or more second blockchain nodes are extracted, performance prediction and dynamic scheduling optimization are performed on the first blockchain node and the one or more second blockchain nodes based on the first operation performance indicator and the second operation performance indicator;

[0027] When a first response interval indicator of the first blockchain node processing a task and a second response interval indicator of each of the one or more second blockchain nodes processing the task are extracted, the first response interval indicator and the second response interval indicator are used to perform performance prediction and dynamic scheduling optimization on the first blockchain node and the one or more second blockchain nodes;

[0028] When the first operating performance indicator, the second operating performance indicator, the first response interval indicator, and the second response interval indicator are extracted, a machine learning algorithm is used to perform performance prediction and dynamic scheduling optimization on the first blockchain node and the one or more second blockchain nodes based on the first operating performance indicator, the second operating performance indicator, the first response interval indicator, and the second response interval indicator.

[0029] In a possible implementation of the first aspect, the using a machine learning algorithm to perform performance prediction and dynamic scheduling optimization on the first blockchain node and the one or more second blockchain nodes based on the first operation performance indicator, the second operation performance indicator, the first response interval indicator, and the second response interval indicator includes:

[0030] Pre-collecting sample operating indicators of the first blockchain node and the second blockchain node, and obtaining a priori performance status label data related to the performance of the first blockchain node and the second blockchain node, wherein the sample operating indicators include operating performance indicators and response interval indicators;

[0031] Using the sample operation indicators of the first blockchain node and the second blockchain node as input model input data, and using the prior performance status label data as model output data, training the machine learning network to generate a trained performance status prediction model;

[0032] Analyze the first operation performance indicator, the second operation performance indicator, the first response interval indicator, and the second response interval indicator based on the trained performance status prediction model to generate performance prediction results of the first blockchain node and the one or more second blockchain nodes;

[0033] According to the performance prediction result, the task allocation of the first blockchain node and the second blockchain node is dynamically adjusted to achieve load balancing.

[0034] In a possible implementation of the first aspect, the step of acquiring a plurality of blockchain nodes to be scheduled from the blockchain cluster based on the dynamic scheduling task generated for the blockchain cluster includes:

[0035] A monitoring mechanism for the blockchain cluster is pre-set, and the dynamic scheduling tasks generated in the blockchain cluster are continuously monitored through the monitoring mechanism, and the dynamic scheduling tasks are triggered by the cluster manager and the task scheduler;

[0036] When a new dynamically scheduled task is monitored, the task information of the dynamically scheduled task is recorded, wherein the task information includes the task type, priority, and required resources;

[0037] Performing demand analysis on the monitored dynamic scheduling tasks to determine the task conditions of the dynamic scheduling tasks for the blockchain nodes, wherein the task conditions include computing power conditions, storage capacity conditions, and network bandwidth conditions;

[0038] According to the task conditions, evaluating the node resource situation in the blockchain cluster to determine candidate blockchain nodes that meet the task conditions;

[0039] The screened candidate blockchain nodes are sorted, and different priorities are assigned to the candidate blockchain nodes according to the sorting results, and the information of the screened and sorted candidate blockchain nodes is updated into the node status database.

[0040] In a possible implementation of the first aspect, the method further includes:

[0041] The dynamic scheduling optimization result is sent to the terminal corresponding to the dynamic scheduling task.

[0042] Based on the second aspect of the present application, a blockchain service system is provided, which includes a machine-readable storage medium and a processor, wherein the machine-readable storage medium stores machine-executable instructions, and when the processor executes the machine-executable instructions, the blockchain service system implements the aforementioned machine learning-based blockchain network dynamic scheduling method.

[0043] Based on the third aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are executed, the aforementioned machine learning-based blockchain network dynamic scheduling method is implemented.

[0044] Based on any one of the above aspects, the technical effect of the present application is:

[0045] The embodiment of the present application can realize the efficient processing of dynamic scheduling tasks in the blockchain cluster. By intelligently obtaining multiple blockchain nodes to be scheduled from the blockchain cluster, the system can quickly respond to and process various dynamic scheduling tasks, thereby improving the operating efficiency and response speed of the entire blockchain network. Secondly, by obtaining the node authenticator and determining the node group, the first blockchain node with performance to be optimized can be accurately identified, and the scheduling group to which it belongs can be further determined, which helps to implement the dynamic scheduling optimization strategy more accurately. Furthermore, when the target scheduling group to which the first blockchain node belongs is composed of multiple blockchain nodes with a collaborative working relationship, the second blockchain node associated with the first blockchain node can be intelligently obtained and recorded, providing data support for subsequent performance prediction and dynamic scheduling optimization. Finally, by using machine learning algorithms to perform performance prediction and dynamic scheduling optimization on the first blockchain node and one or more second blockchain nodes, it is possible to achieve reasonable allocation and efficient utilization of blockchain network resources, which not only helps to improve the overall performance of the blockchain network, but also effectively avoids network congestion and resource waste, thereby providing users with more stable and reliable blockchain services. Thus, through the intelligent dynamic scheduling method, the processing capacity, response speed and resource utilization of the blockchain network are significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other corresponding drawings can be obtained based on these drawings without paying creative work.

[0047] Figure 1 A schematic diagram of the process of a blockchain network dynamic scheduling method based on machine learning provided in an embodiment of the present application is shown;

[0048] Figure 2 A schematic diagram of the component structure of a blockchain service system for implementing the above-mentioned machine learning-based blockchain network dynamic scheduling method provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0049] The embodiments of the present application are described below in conjunction with the drawings in the present application. It should be understood that the implementation methods described below in conjunction with the drawings are exemplary descriptions for explaining the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions of the embodiments of the present application.

[0050] It will be understood by those skilled in the art that, unless specifically stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the terms "including" and "comprising" used in the embodiments of the present application refer to that the corresponding features can be implemented as the presented features, information, data, steps, operations, elements and / or components, but do not exclude the implementation as other features, information, data, steps, operations, elements, components and / or combinations thereof supported by the technical field. It should be understood that when an element is "connected" or "coupled" to another element, the element may be directly connected or coupled to the other element, or may refer to the element and the other element establishing a connection relationship through an intermediate element. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling, and the term "and / or" used herein indicates at least one of the items defined by the term, such as "A and / or B" may be implemented as "A", or as "B", or as "A and B".

[0051] In order to make the purpose, technical solution and advantages of the present application clearer, the implementation mode of the present application will be further described in detail with reference to the accompanying drawings. The technical solution of the embodiment of the present application and the technical effect produced by the technical solution of the present application are explained below by describing several exemplary implementation modes. It should be pointed out that the following implementation modes can refer to, draw lessons from or combine with each other, and the same terms, similar features and similar implementation steps in different implementation modes are not described repeatedly.

[0052] Figure 1 The flowchart of the blockchain network dynamic scheduling method and system based on machine learning provided by the embodiment of the present application is shown. It should be understood that in other embodiments, the order of some steps of the blockchain network dynamic scheduling method based on machine learning in this embodiment can be shared with each other according to actual needs, or some steps can be omitted or maintained. The detailed steps of the blockchain network dynamic scheduling method based on machine learning include:

[0053] Step S110, based on the dynamic scheduling task generated for the blockchain cluster, obtain multiple blockchain nodes to be scheduled from the blockchain cluster.

[0054] The blockchain service system receives a dynamic scheduling task from the blockchain cluster, which requires adjusting the node resource allocation in the cluster to optimize the overall performance. In order to perform this dynamic scheduling task, the blockchain service system first needs to obtain multiple blockchain nodes to be scheduled from the blockchain cluster.

[0055] For example, in an exemplary scenario, a blockchain platform of a large financial institution needs to optimize the resources of its blockchain cluster due to a surge in transaction volume. As part of cluster management, the blockchain service system is responsible for reallocating node resources based on dynamic scheduling tasks. The blockchain service system captures this dynamic scheduling task triggered by the cluster manager through a built-in monitoring mechanism. After analyzing the requirements of the dynamic scheduling task, the blockchain service system determines that it needs to find blockchain nodes that meet certain standards in computing power, storage capacity, and network bandwidth. The blockchain service system queries the node status database in the cluster and selects a batch of qualified candidate blockchain nodes for subsequent scheduling optimization.

[0056] Step S120, when a first blockchain node whose performance is to be optimized is obtained among the multiple blockchain nodes to be scheduled, a node authenticator corresponding to the first blockchain node is obtained, and a node group to which the first blockchain node belongs is determined based on the node authenticator, wherein the node group includes one or more blockchain nodes belonging to a scheduling group, and the scheduling group is composed of one blockchain node or multiple blockchain nodes having a collaborative working relationship.

[0057] In this embodiment, after obtaining multiple blockchain nodes to be scheduled, the blockchain service system begins to analyze the performance data of these blockchain nodes. In this process, the blockchain service system finds a node whose performance needs to be optimized, which is called the first blockchain node.

[0058] For example, among the candidate blockchain nodes screened, the blockchain service system discovered through performance monitoring tools that one blockchain node named "NodeA" had CPU and memory usage rates far exceeding its upper limit in the last few hours, indicating that its performance needs to be optimized. In order to more effectively allocate the resources of this blockchain node, the blockchain service system needs to determine the node group to which it belongs. The blockchain service system obtains the node authenticator of "NodeA" from the configuration information of the blockchain node, and by querying the task scheduling center, finds the preconfigured node group associated with this node authenticator. The preconfigured node group contains multiple blockchain nodes that have a collaborative working relationship with "NodeA" and jointly undertake part of the blockchain network tasks.

[0059] Step S130, when the target scheduling group to which the first blockchain node belongs is composed of the multiple blockchain nodes that have a collaborative working relationship, obtain one or more second blockchain nodes recorded in the node group to which the first blockchain node belongs, and use a machine learning algorithm to perform performance prediction and dynamic scheduling optimization on the first blockchain node and the one or more second blockchain nodes, and the second blockchain node is located in the target scheduling group.

[0060] In this embodiment, after determining the node group to which the first blockchain node "NodeA" belongs, the blockchain service system further analyzes other nodes in this group, which are called second blockchain nodes. In order to optimize the performance of the entire group, the blockchain service system decides to use machine learning algorithms to perform performance prediction and dynamic scheduling optimization on these nodes.

[0061] For example, the blockchain service system obtains information about all other nodes in the node group to which "NodeA" belongs, including their operating performance indicators, response interval indicators, etc. Then, the blockchain service system calls a pre-trained machine learning model that can predict the future performance of the node based on its historical operating data. By inputting the performance indicator data of "NodeA" and other nodes, the model outputs the performance prediction results of each node. Based on these results, the blockchain service system develops a dynamic scheduling plan to reallocate some of the tasks originally assigned to "NodeA" to other nodes with better performance in the group, thereby achieving load balancing and performance improvement for the entire node group. At the same time, the blockchain service system also sends this optimization plan to the cluster manager for adjustment in actual operation.

[0062] Based on the above steps, the embodiment of the present application can realize efficient processing of dynamic scheduling tasks in the blockchain cluster. By intelligently obtaining multiple blockchain nodes to be scheduled from the blockchain cluster, the system can quickly respond to and process various dynamic scheduling tasks, thereby improving the operating efficiency and response speed of the entire blockchain network. Secondly, by obtaining the node authenticator and determining the node group, the first blockchain node with performance to be optimized can be accurately identified, and the scheduling group to which it belongs can be further determined, which helps to implement the dynamic scheduling optimization strategy more accurately. Furthermore, when the target scheduling group to which the first blockchain node belongs is composed of multiple blockchain nodes with a collaborative working relationship, the second blockchain node associated with the first blockchain node can be intelligently obtained and recorded, providing data support for subsequent performance prediction and dynamic scheduling optimization. Finally, by using machine learning algorithms to perform performance prediction and dynamic scheduling optimization on the first blockchain node and one or more second blockchain nodes, it is possible to achieve reasonable allocation and efficient utilization of blockchain network resources, which not only helps to improve the overall performance of the blockchain network, but also effectively avoids network congestion and resource waste, thereby providing users with more stable and reliable blockchain services. Thus, through the intelligent dynamic scheduling method, the processing capacity, response speed and resource utilization of the blockchain network are significantly improved.

[0063] In a possible implementation, before step S120, a task scheduling center associated with the blockchain cluster may be obtained, and the task scheduling center is used to record each preconfigured node group matched by the blockchain cluster.

[0064] Step S120 includes: obtaining the node authenticator, and obtaining from the task scheduling center the preconfigured node group to which the first blockchain node belongs and which is associated with the node authenticator.

[0065] In this embodiment, before the blockchain service system starts processing dynamic scheduling tasks and identifies the first blockchain node whose performance needs to be optimized, it is first necessary to access the task scheduling center associated with the blockchain cluster. The dynamic scheduling task scheduling center is a key component that records and manages the information of each preconfigured node group in the blockchain cluster.

[0066] In detail, the blockchain service system is connected to the task scheduling center of the blockchain cluster through a secure network communication protocol. The task scheduling center stores the configuration information of all nodes in the blockchain cluster, including node authenticators, performance parameters, and the node groups they belong to. This information is crucial for the blockchain service system because it will be used for subsequent node performance analysis and dynamic scheduling decisions.

[0067] Once the blockchain service system is connected to the task scheduling center, it will start processing the first blockchain node that has been identified and whose performance needs to be optimized. In order to determine the node group to which this first blockchain node belongs, the blockchain service system first needs to obtain the node authenticator of the first blockchain node.

[0068] In detail, the blockchain service system extracts a unique identifier of the first blockchain node, namely, a node authenticator, from the previously collected performance monitoring data. The authenticator is usually assigned by the cluster manager when the node joins the blockchain cluster and is used to uniquely identify each node within the cluster.

[0069] After obtaining the node authenticator of the first blockchain node, the blockchain service system will next query the preconfigured node group associated with this authenticator from the task scheduling center.

[0070] In detail, the blockchain service system sends a query request to the task scheduling center, which includes the node authenticator of the first blockchain node. After receiving the request, the task scheduling center searches for a record matching the authenticator in its database. Once a matching record is found, the task scheduling center returns information about a preconfigured node group associated with the node authenticator to the blockchain service system. The node group includes other nodes that have a collaborative working relationship with the first blockchain node. These nodes may jointly participate in certain specific blockchain tasks or share certain resources.

[0071] After the blockchain service system obtains the pre-configured node group information associated with the node authenticator of the first blockchain node from the task scheduling center, it formally determines the group to which the node belongs. In detail, after receiving the response from the task scheduling center, the blockchain service system parses the node group information to which the first blockchain node belongs, which includes the identifiers, performance parameters, and collaborative working relationships of other nodes in the group. With this information, the blockchain service system can more accurately evaluate the performance status of the entire group, and accordingly formulate a more effective dynamic scheduling strategy to optimize the overall performance of the cluster.

[0072] In a possible implementation, before step S110, the method further includes:

[0073] Step S101, obtaining multiple blockchain nodes to be evaluated in the blockchain cluster.

[0074] Step S102: determining the collaborative working relationship corresponding to each blockchain node among the multiple blockchain nodes to be evaluated.

[0075] Step S103, based on the corresponding collaborative working relationship, divide the node authenticators based on the various blockchain nodes into the various preconfigured node groups, and the node authenticators corresponding to the blockchain nodes in the same scheduling group among the multiple blockchain nodes to be evaluated are divided into the same preconfigured node group.

[0076] In this embodiment, before the blockchain service system starts to execute the dynamic scheduling task, it is first necessary to obtain multiple blockchain nodes to be evaluated from the blockchain cluster. These blockchain nodes will be the objects of subsequent scheduling and performance optimization.

[0077] In detail, the blockchain service system connects to the communication interface of the blockchain cluster and sends a request to obtain a list of all available blockchain nodes in the cluster. These nodes may be in different states, such as idle, busy, or under maintenance. The blockchain service system selects those nodes that are in an idle state and can participate in scheduling, and uses them as blockchain nodes to be evaluated.

[0078] After obtaining the blockchain nodes to be evaluated, the blockchain service system needs to further analyze the collaborative working relationship between these blockchain nodes. This is to understand how blockchain nodes depend on and collaborate with each other, so that node groups can be divided more reasonably in the future.

[0079] In detail, the blockchain service system accesses the configuration database or related management interface of the blockchain cluster to obtain the collaborative work information between each node to be evaluated. This information may include communication records between nodes, cooperative relationships for transaction verification, frequency of data synchronization, etc. By analyzing this information, the blockchain service system can identify which nodes often work together and which nodes have strong dependencies.

[0080] After understanding the collaborative working relationship between the nodes to be evaluated, the blockchain service system begins to divide these nodes into different pre-configured node groups according to their collaborative relationship. The purpose of this is to optimize the scheduling strategy and ensure that the collaborative nodes can be scheduled at the same time, thereby improving the overall performance of the blockchain cluster.

[0081] Specifically, the blockchain service system divides nodes with close cooperation relationships into the same preconfigured node group based on the collaborative working relationship obtained from the previous analysis. For example, if nodes A, B, and C often participate in transaction verification and data synchronization together, they will be divided into the same group. During the division process, the blockchain service system uses the node's authenticator as a unique identifier to ensure that each node can be accurately assigned to the corresponding group.

[0082] After completing the preliminary grouping of blockchain nodes, the blockchain service system needs to further consider the concept of scheduling groups. Scheduling groups refer to a group of nodes that need to be activated at the same time during the actual scheduling process to ensure the atomicity and consistency of blockchain operations.

[0083] In detail, the blockchain service system identifies the node combinations that need to be scheduled at the same time according to the scheduling requirements. These nodes are usually located in the same scheduling group. In order to ensure that these nodes can be activated at the same time, the blockchain service system divides their corresponding node authenticators into the same preconfigured node group. The advantage of this is that in the subsequent scheduling process, the blockchain service system can activate the entire group at one time, thereby improving the efficiency and accuracy of scheduling. For example, if nodes D, E, and F need to be scheduled as a whole to complete a task, then their authenticators will be divided into the same preconfigured node group.

[0084] In a possible implementation, step S102 may include:

[0085] Step S1021, continuously collect network communication data between each blockchain node through a network monitoring tool, and the network communication data includes the amount of transmitted data, transmission frequency, and transmission delay.

[0086] In this embodiment, the blockchain service system first starts a network monitoring tool, which can capture and record the network communication data between each node in the blockchain cluster in real time. These network communication data are crucial for the subsequent analysis of the collaborative working relationship between nodes.

[0087] Specifically, the blockchain service system configures a network monitoring tool to monitor the network traffic of the blockchain cluster. Over a period of time, the tool continuously collects and records key indicators such as the amount of data transmitted, the transmission frequency, and the transmission delay between each blockchain node. These network communication data are stored in the database of the blockchain service system for subsequent analysis.

[0088] Step S1022: Analyze the network communication data to determine the communication density and communication stability between blockchain nodes, and generate an initial blockchain node sequence that may have a collaborative working relationship based on the communication density and communication stability between the determined blockchain nodes.

[0089] After collecting the network communication data, the blockchain service system begins to analyze the network communication data. The purpose of the analysis is to determine which nodes communicate frequently and stably, so as to preliminarily identify nodes that may have a collaborative working relationship.

[0090] In detail, the blockchain service system runs a data analysis algorithm to process network communication data. The algorithm calculates the communication density between each pair of nodes, such as by counting the amount and frequency of data transmission within a specific time window. At the same time, the algorithm also evaluates the stability of communication, such as by checking the fluctuation of transmission delay. Based on these analysis results, the blockchain service system generates an initial blockchain node sequence, and the initial blockchain node pairs in this initial blockchain node sequence may have a collaborative working relationship.

[0091] Step S1023, obtaining all node activity data of each initial blockchain node in the initial blockchain node sequence, wherein the node activity data includes block generation activity data, verification activity data, and propagation activity data.

[0092] In order to gain a deeper understanding of how these nodes that may have collaborative relationships collaborate with each other in actual work, the blockchain service system needs to further obtain their node activity data.

[0093] In detail, the blockchain service system obtains the activity data of each initial blockchain node in the initial blockchain node sequence through the blockchain cluster API or related monitoring tools. These activity data include block generation activity data (such as the generation time and content of new blocks), verification activity data (such as transaction verification records and timestamps), and propagation activity data (such as the broadcast of block or transaction information). These data provide the blockchain service system with the specific behavior patterns of nodes in the blockchain network.

[0094] Step S1024, analyzing the node activity data to determine the mutual cooperation pattern characteristics of each initial blockchain node when processing tasks.

[0095] After obtaining the node activity data, the blockchain service system begins to analyze these node activity data to reveal the collaboration patterns of the nodes when processing tasks.

[0096] In detail, the blockchain service system uses data analysis techniques, such as pattern recognition and time series analysis, to explore hidden patterns in node activity data. By analyzing the time synchronization, activity frequency, and similarity of activity types during the block generation, verification, and propagation process, the blockchain service system is able to identify the characteristics of the collaboration patterns between nodes. For example, some nodes may often participate in the verification and propagation of blocks together in the same time period, indicating a close collaboration between them.

[0097] Step S1025, taking each initial blockchain node as a vertex in the graph, taking the mutual cooperation pattern characteristics between the nodes as edges, and assigning weights to the edges in the graph, wherein the weights represent the intensity of the collaborative work between the nodes.

[0098] In order to better visualize and quantify the collaborative working relationships between nodes, the blockchain service system represents these relationships as a weighted graph.

[0099] In detail, the blockchain service system builds a graph model in which each initial blockchain node is represented as a vertex. Based on the characteristics of the mutual cooperation pattern between nodes obtained from the previous analysis, the blockchain service system adds edges to the graph to represent these cooperation relationships and assigns a weight to each edge, which reflects the intensity or closeness of the cooperation between nodes. For example, if two nodes often participate in the verification and propagation of blocks together, the weight of the edge connecting these two vertices will be relatively high.

[0100] Step S1026, using a community detection algorithm to divide the graph, identify node communities that satisfy preset collaborative working relationships, and output the collaborative working relationships corresponding to each blockchain node in the multiple blockchain nodes to be evaluated based on the node communities that satisfy the preset collaborative working relationships. The nodes in each node community are regarded as a node group with a strengthened collaborative working relationship.

[0101] Finally, the blockchain service system uses a community detection algorithm to partition the weighted graph to identify groups of nodes with enhanced collaborative working relationships.

[0102] In detail, the blockchain service system uses advanced community detection algorithms (such as Louvain algorithm or GN algorithm) to partition the graph. These algorithms can identify closely connected node communities based on the weight of the edges and the connection between vertices. Each identified node community represents a group of nodes with enhanced collaborative working relationships. The blockchain service system outputs the collaborative working relationships corresponding to each node in multiple blockchain nodes to be evaluated based on these node communities. In this way, the blockchain service system can provide valuable reference information for subsequent dynamic scheduling tasks.

[0103] In a possible implementation manner, step S130 includes at least one of the following:

[0104] Step S131: When a first operating performance indicator corresponding to the first blockchain node and a second operating performance indicator corresponding to each of the one or more second blockchain nodes are extracted, performance prediction and dynamic scheduling optimization are performed on the first blockchain node and the one or more second blockchain nodes based on the first operating performance indicator and the second operating performance indicator.

[0105] In this embodiment, when the blockchain service system extracts the first operation performance indicator of the first blockchain node and the second operation performance indicator of each of the one or more second blockchain nodes, the blockchain service system first collects and organizes these operation performance indicators, which may include the CPU usage rate, memory occupancy rate, disk I / O performance, network bandwidth utilization rate, etc. of the node. These indicators can reflect the operation status and load of the node at the current time point.

[0106] Then, the blockchain service system can use these operating performance indicators as input to perform performance predictions through trained machine learning models (such as random forests, support vector machines, etc.). The model will predict the performance of the node in the future based on historical data and current data, such as whether the node will encounter performance bottlenecks, resource exhaustion, etc.

[0107] Based on the prediction results, the blockchain service system can perform dynamic scheduling optimization. If it is predicted that a node is about to have performance problems, the blockchain service system can migrate some tasks on the node to other nodes with better performance, or adjust the task allocation strategy to balance the load of each node and ensure the stable operation of the entire blockchain cluster.

[0108] Step S132, when the first response interval indicator of the first blockchain node processing the task and the second response interval indicator of each of the one or more second blockchain nodes processing the task are extracted, the first response interval indicator and the second response interval indicator are used to perform performance prediction and dynamic scheduling optimization on the first blockchain node and the one or more second blockchain nodes.

[0109] When the blockchain service system extracts the first response interval indicator of the first blockchain node processing tasks and the second response interval indicator of each node processing tasks in one or more second blockchain nodes, the blockchain service system first collects and organizes these response interval indicators, which reflect the response speed and efficiency of the nodes when processing tasks. A shorter response interval means that the node can complete the task more quickly, while a longer response interval may indicate that the node has a performance bottleneck or resource competition problem.

[0110] The blockchain service system can then use these response interval indicators as input to make performance predictions through machine learning models. The model predicts the response speed and efficiency of the node in processing tasks in the future based on historical data and current data.

[0111] Based on the prediction results, the blockchain service system can perform corresponding dynamic scheduling optimization. If it is predicted that the response interval of a certain node will become longer, that is, the speed of processing tasks will become slower, the blockchain service system can migrate some tasks on this node to a node with faster response to improve the processing efficiency and response speed of the entire blockchain cluster.

[0112] Step S133, when the first operation performance indicator, the second operation performance indicator, the first response interval indicator, and the second response interval indicator are extracted, a machine learning algorithm is used to perform performance prediction and dynamic scheduling optimization on the first blockchain node and the one or more second blockchain nodes based on the first operation performance indicator, the second operation performance indicator, the first response interval indicator, and the second response interval indicator.

[0113] When the blockchain service system extracts both operating performance indicators and response interval indicators, it will use these indicators in combination to perform more comprehensive performance prediction and dynamic scheduling optimization.

[0114] The blockchain service system can first analyze and predict the operation performance index and response interval index according to the methods of step one and step two respectively. Then, the blockchain service system can combine the prediction results of these two aspects to obtain a more comprehensive node performance prediction result. Based on the comprehensive prediction results, the blockchain service system can perform more refined dynamic scheduling optimization. For example, the blockchain service system can formulate a comprehensive load balancing strategy based on the operation performance index and response interval index of the node, taking into account both the operation status and load of the node, and the response speed and efficiency of the node. Through such an optimization strategy, the blockchain service system can ensure that the entire blockchain cluster operates in an efficient and stable state.

[0115] In a possible implementation, step S133 includes:

[0116] Step S1331, pre-collecting sample operating indicators of the first blockchain node and the second blockchain node, and obtaining prior performance status label data related to the performance of the first blockchain node and the second blockchain node, the sample operating indicators including operating performance indicators and response interval indicators.

[0117] In this embodiment, the blockchain service system will start to collect sample operation indicators of the first blockchain node and one or more second blockchain nodes in the early preparation stage. These sample operation indicators include but are not limited to CPU usage, memory usage, disk I / O, network latency and other operation performance indicators, as well as task processing time, response speed and other response interval indicators.

[0118] At the same time, the blockchain service system also obtains prior performance status label data related to the performance of these nodes. This data can be obtained through historical records, expert evaluations, or other reliable methods to identify the performance status of the node under specific operating indicators, such as "high performance status", "medium performance status", or "low performance status".

[0119] In detail, the blockchain service system regularly collects operation data from each blockchain node, such as recording the node's CPU and memory usage and the average response time of task processing every 10 minutes. At the same time, based on historical data and expert evaluation, these data are labeled with performance status, such as "excellent", "good", "medium", and "poor".

[0120] Step S1332: Use the sample operation indicators of the first blockchain node and the second blockchain node as input model input data, and use the prior performance status label data as model output data to train the machine learning network and generate a trained performance status prediction model.

[0121] The blockchain service system uses the collected sample operation indicators as input data and the prior performance status label data as output data, and inputs them into the machine learning network for training. During this process, the blockchain service system can use a deep learning framework (such as TensorFlow or PyTorch) to build and train neural network models.

[0122] Through multiple iterations and optimizations, the machine learning network will gradually learn how to accurately predict the performance status of the node based on the input operating indicators.

[0123] In detail, the blockchain service system uses the TensorFlow framework to build a deep neural network model. The previously collected sample data with performance status labels is fed into the model for training. By adjusting network parameters and multiple iterations, the accuracy of the model on the validation set gradually improves, and finally reaches an acceptable performance status prediction level.

[0124] Step S1333: Analyze the first operating performance indicator, the second operating performance indicator, the first response interval indicator, and the second response interval indicator based on the trained performance status prediction model to generate performance prediction results for the first blockchain node and the one or more second blockchain nodes.

[0125] Once the machine learning network training is completed and the performance status prediction model is generated, the blockchain service system can use this model to analyze and predict new operating indicator data. The blockchain service system can input the real-time operating indicators (including the first operating performance indicator, the second operating performance indicator, the first response interval indicator, and the second response interval indicator) of the first blockchain node and one or more second blockchain nodes into the model, and the model will output the current performance status prediction results of these nodes.

[0126] Specifically, the blockchain service system collects the operating indicator data of each blockchain node in real time and inputs the data into the previously trained performance status prediction model. The model quickly gives the performance status prediction results of each node, such as "node A is currently in a high performance state", "node B is currently in a medium performance state", etc.

[0127] Step S1334: dynamically adjust the task allocation of the first blockchain node and the second blockchain node according to the performance prediction result to achieve load balancing.

[0128] According to the performance prediction results output by the model, the blockchain service system can dynamically adjust the task allocation strategy of the first blockchain node and one or more second blockchain nodes to achieve load balancing and maximize performance. If a node is predicted to be about to enter a low-performance state, the blockchain service system can migrate the tasks originally assigned to the node to other nodes with better performance; conversely, if a node is predicted to be about to enter a high-performance state, the blockchain service system may assign more tasks to the node to utilize its powerful processing capabilities.

[0129] In detail, based on the output of the performance status prediction model, the blockchain service system finds that node A is about to enter a low-performance state while node B will maintain a high-performance state. Therefore, the blockchain service system decides to migrate some of the computationally intensive tasks originally assigned to node A to node B for execution to ensure that these tasks can be completed efficiently and reduce the overall processing time. At the same time, the blockchain service system will continuously adjust the task allocation strategy based on real-time performance data to maintain the system's optimal performance state.

[0130] In a possible implementation, step S110 includes:

[0131] Step S111, pre-setting a monitoring mechanism for the blockchain cluster, and continuously monitoring the dynamically scheduled tasks generated in the blockchain cluster through the monitoring mechanism, wherein the dynamically scheduled tasks are triggered by the cluster manager and the task scheduler.

[0132] In this embodiment, the blockchain service system will first set up a special monitoring mechanism, which can continuously monitor the dynamics in the blockchain cluster. The monitoring mechanism is implemented by establishing communication connections with key components such as the cluster manager and task scheduler to ensure that once a new dynamic scheduling task is generated, the blockchain service system can be immediately informed.

[0133] In detail, the blockchain service system configures a listening service that keeps communicating with the cluster manager and task scheduler of the blockchain cluster in real time. Whenever the cluster manager or task scheduler triggers a new dynamically scheduled task, the listening service will immediately capture this information.

[0134] Step S112, when a new dynamically scheduled task is monitored, the task information of the dynamically scheduled task is recorded, and the task information includes task type, priority, and required resources.

[0135] When a new dynamically scheduled task is detected, the blockchain service system can immediately record the relevant information of the dynamically scheduled task. This information includes but is not limited to the task type (such as data verification, transaction processing, etc.), the priority of the task (such as high, medium, low), and the resources required for task execution (such as the number of CPU cores, memory size, storage space, etc.).

[0136] Specifically, after the monitoring service captures a new dynamically scheduled task, it will automatically record the task information in the task log of the blockchain service system. For example, the record shows that this is a data verification task with a high priority, which requires 4 CPU cores and 8GB of memory resources to execute.

[0137] Step S113, performing demand analysis on the monitored dynamic scheduling tasks to determine the task conditions of the dynamic scheduling tasks for the blockchain nodes, wherein the task conditions include computing power conditions, storage capacity conditions, and network bandwidth conditions.

[0138] The blockchain service system can analyze the needs of the recorded dynamic scheduling tasks and clarify the specific requirements of the dynamic scheduling tasks for the blockchain nodes. These requirements usually include the computing power of the nodes (such as the number of transactions that can be processed per second), storage capacity (such as at least 1TB of storage space) and network bandwidth (such as at least 100Mbps upload and download speed).

[0139] In detail, the blockchain service system analyzes the information in the task log and determines that this data verification task requires the node to have strong computing power to quickly process data, and also requires sufficient storage space to save the intermediate data in the verification process. Therefore, the blockchain service system sets computing power conditions and storage capacity conditions.

[0140] Step S114: According to the task conditions, the node resource situation in the blockchain cluster is evaluated to determine the candidate blockchain nodes that meet the task conditions.

[0141] According to the task conditions, the blockchain service system can evaluate the resource status of each node in the blockchain cluster. The evaluation process mainly compares whether the actual resources of the node (such as CPU performance, hard disk capacity, network speed, etc.) match the task conditions, so as to find candidate blockchain nodes that have the conditions required to execute the task.

[0142] In detail, the blockchain service system traverses all nodes in the blockchain cluster and checks the computing performance, storage capacity and network bandwidth of each node. By comparison, the blockchain service system finds that the resource conditions of several nodes meet or exceed the previously set task conditions, and then marks these nodes as candidate blockchain nodes.

[0143] Step S115, sort the screened candidate blockchain nodes, assign different priorities to the candidate blockchain nodes according to the sorting results, and update the screened and sorted candidate blockchain node information into the node status database.

[0144] The blockchain service system can further analyze and sort the selected candidate blockchain nodes. The sorting can be based on the node's comprehensive performance, historical task completion, current load, etc. After the sorting is completed, the blockchain service system can assign a priority to each candidate blockchain node, which will determine the order of the node in subsequent task allocation.

[0145] In detail, the blockchain service system ranks candidate blockchain nodes according to their comprehensive performance and current load. For example, if a node has strong performance but is currently heavily loaded, its priority may be lowered. After the ranking is completed, the blockchain service system updates the information and priority of each candidate blockchain node to the node status database for subsequent task allocation operations.

[0146] In a possible implementation manner, this embodiment may send the dynamic scheduling optimization result to a terminal corresponding to the dynamic scheduling task.

[0147] After completing the performance prediction and dynamic scheduling optimization of blockchain nodes, the blockchain service system needs to promptly notify the relevant task initiators or managers of the optimization results so that they can understand the scheduling of the task and perform subsequent operations or decisions.

[0148] In detail, after completing a series of dynamic scheduling optimization processes, the blockchain service system obtains an optimization result, which shows which blockchain nodes are selected to perform tasks, as well as the distribution of tasks and the expected execution efficiency. In order to inform the initiator or manager of the task of this result, the blockchain service system will perform the following operations:

[0149] The blockchain service system will first organize and optimize the data related to the results, including information about the selected blockchain nodes, specific details of task assignments, estimated execution time and efficiency, etc. These data will be formatted into reports or charts that are easy to read and understand.

[0150] According to the pre-set communication protocol or user preference, the blockchain service system will select an appropriate method to send the optimization results to the corresponding terminal. This can include email, instant messaging, API interface call, web interface update, etc.

[0151] Once the sending method is determined, the blockchain service system will start sending the prepared optimization result data to the terminal corresponding to the dynamic scheduling task. For example, if email is selected as the sending method, the blockchain service system will generate an email containing the optimization results and send it to the email box of the task initiator or manager.

[0152] After sending the optimization results, the blockchain service system may also wait for confirmation from the recipient to ensure that the optimization results have been successfully received. If the recipient has any questions or needs further operational guidance, they can communicate with the blockchain service system through the preset feedback channel.

[0153] Through such a process, the blockchain service system not only completes the optimization of dynamic scheduling tasks, but also ensures that the relevant task initiators or managers are aware of the optimization results in a timely manner, so as to make further decisions or operations.

[0154] Figure 2 A blockchain service system 100 provided in an embodiment of the present application is shown, including a processor 1001, a memory 1003 and a program code stored on the memory 1003, and the processor 1001 executes the above program code to implement the steps of the blockchain network dynamic scheduling method based on machine learning.

[0155] Figure 2 The blockchain service system 100 shown includes: a processor 1001 and a memory 1003. The processor 1001 and the memory 1003 are connected, such as through a bus 1002. Optionally, the blockchain service system 100 may also include a transceiver 1004, which may be used for data interaction between the blockchain service system and other blockchain service systems, such as data transmission and / or data reception. It should be noted that the transceiver 1004 is not limited to one in actual scheduling, and the structure of the blockchain service system 100 does not constitute a limitation on the embodiments of the present application.

[0156] Processor 1001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware data structures or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the transaction records disclosed in this application. Processor 1001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0157] The bus 1002 may include a path to transmit information between the above components. The bus 1002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 1002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 2 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0158] The memory 1003 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), magnetic disk storage medium, other magnetic storage devices, or any other medium that can be used to have or store program code and can be read by a computer, without limitation herein.

[0159] The memory 1003 is used to store program codes for executing the embodiments of the present application, and the execution is controlled by the processor 1001. The processor 1001 is used to execute the program codes stored in the memory 1003 to implement the steps shown in the above method embodiments.

[0160] An embodiment of the present application provides a computer-readable storage medium having program code stored thereon. When the program code is executed by a processor, the steps of the aforementioned method embodiment and the corresponding transaction records can be implemented.

[0161] It should be understood that, although each operation step is indicated by arrows in the flowchart of the embodiment of the present application, the implementation order of these steps is not limited to the order indicated by the arrows. Unless there is clear explanation in this article, in some implementation scenarios of the embodiment of the present application, the implementation steps in each flowchart can be performed in other orders based on demand. In addition, some or all of the steps in each flowchart may include multiple sub-steps or multiple stages according to the actual implementation scenario, and some or all of these sub-steps or stages may be executed at the same time, and each sub-step or stage in these sub-steps or stages may also be executed at different times respectively. In different scenarios at the execution time, the execution order of these sub-steps or stages can be flexibly configured based on demand, and the embodiment of the present application does not limit this.

[0162] The above is only an optional implementation method for some implementation scenarios of the present application. It should be pointed out that for ordinary technicians in this technical field, without departing from the technical concept of the scheme of the present application, other similar implementation methods based on the technical ideas of the present application are also within the protection scope of the embodiments of the present application.

Claims

1. A blockchain network dynamic scheduling method based on machine learning, characterized in that: The method comprises: Based on the dynamic scheduling task generated for the blockchain cluster, a plurality of blockchain nodes to be scheduled are obtained from the blockchain cluster; When a first blockchain node whose performance is to be optimized is obtained from among the multiple blockchain nodes to be scheduled, a node authenticator corresponding to the first blockchain node is obtained, and a node group to which the first blockchain node belongs is determined based on the node authenticator, wherein the node group includes one or more blockchain nodes belonging to a scheduling group, and the scheduling group is composed of one blockchain node or multiple blockchain nodes in a collaborative working relationship; When the target scheduling group to which the first blockchain node belongs is composed of the plurality of blockchain nodes in a collaborative working relationship, one or more second blockchain nodes recorded in the node group to which the first blockchain node belongs are obtained, and a machine learning algorithm is used to perform performance prediction and dynamic scheduling optimization on the first blockchain node and the one or more second blockchain nodes, where the second blockchain node is located in the target scheduling group; Before obtaining the node authenticator corresponding to the first blockchain node and determining the node group to which the first blockchain node belongs based on the node authenticator, the method further includes: Obtain a task scheduling center associated with the blockchain cluster, wherein the task scheduling center is used to record each preconfigured node group matched by the blockchain cluster; The obtaining a node authenticator corresponding to the first blockchain node, and determining the node group to which the first blockchain node belongs based on the node authenticator, includes: Obtaining the node authenticator, and obtaining from the task scheduling center the preconfigured node group to which the first blockchain node belongs and which is associated with the node authenticator; Before the dynamic scheduling task generated based on the blockchain cluster obtains a plurality of blockchain nodes to be scheduled from the blockchain cluster, the method further includes: Obtain multiple blockchain nodes to be evaluated in the blockchain cluster; Determine a collaborative working relationship corresponding to each blockchain node among the multiple blockchain nodes to be evaluated; Based on the corresponding collaborative working relationship, each node authenticator based on each blockchain node is divided into each preconfigured node group, and the node authenticators corresponding to the blockchain nodes in the same scheduling group among the multiple blockchain nodes to be evaluated are divided into the same preconfigured node group; The step of determining the collaborative working relationship corresponding to each blockchain node among the multiple blockchain nodes to be evaluated includes: Continuously collect network communication data between various blockchain nodes through network monitoring tools, and the network communication data includes the amount of transmitted data, transmission frequency, and transmission delay; Analyzing the network communication data to determine the communication density and communication stability between blockchain nodes, and generating an initial blockchain node sequence that may have a collaborative working relationship based on the communication density and communication stability between the blockchain nodes; Obtain all node activity data of each initial blockchain node in the initial blockchain node sequence, wherein the node activity data includes block generation activity data, verification activity data, and propagation activity data; Analyzing the node activity data to determine the characteristics of the mutual cooperation pattern of each initial blockchain node when processing tasks; Each initial blockchain node is regarded as a vertex in the graph, the characteristics of the mutual cooperation pattern between nodes are regarded as edges, and weights are assigned to the edges in the graph, and the weights represent the strength of the cooperation between nodes; The graph is divided by using a community detection algorithm to identify node communities that satisfy a preset collaborative working relationship, and the collaborative working relationship corresponding to each blockchain node in the plurality of blockchain nodes to be evaluated is output based on the node communities that satisfy the preset collaborative working relationship, and the nodes in each node community are regarded as a node group with a strengthened collaborative working relationship; The using of a machine learning algorithm to perform performance prediction and dynamic scheduling optimization on the first blockchain node and the one or more second blockchain nodes includes at least one of the following: When a first operation performance indicator corresponding to the first blockchain node and a second operation performance indicator corresponding to each of the one or more second blockchain nodes are extracted, performance prediction and dynamic scheduling optimization are performed on the first blockchain node and the one or more second blockchain nodes based on the first operation performance indicator and the second operation performance indicator; When a first response interval indicator of the first blockchain node processing a task and a second response interval indicator of each of the one or more second blockchain nodes processing the task are extracted, performance prediction and dynamic scheduling optimization are performed on the first blockchain node and the one or more second blockchain nodes based on the first response interval indicator and the second response interval indicator; When the first operation performance indicator, the second operation performance indicator, the first response interval indicator, and the second response interval indicator are extracted, a machine learning algorithm is used to perform performance prediction and dynamic scheduling optimization on the first blockchain node and the one or more second blockchain nodes based on the first operation performance indicator, the second operation performance indicator, the first response interval indicator, and the second response interval indicator; The using of a machine learning algorithm to perform performance prediction and dynamic scheduling optimization on the first blockchain node and the one or more second blockchain nodes based on the first operation performance indicator, the second operation performance indicator, the first response interval indicator, and the second response interval indicator includes: Pre-collecting sample operating indicators of the first blockchain node and the second blockchain node, and obtaining a priori performance status label data related to the performance of the first blockchain node and the second blockchain node, wherein the sample operating indicators include operating performance indicators and response interval indicators; Using the sample operation indicators of the first blockchain node and the second blockchain node as input model input data, and using the prior performance status label data as model output data, training the machine learning network to generate a trained performance status prediction model; Analyze the first operation performance indicator, the second operation performance indicator, the first response interval indicator, and the second response interval indicator based on the trained performance status prediction model to generate performance prediction results of the first blockchain node and the one or more second blockchain nodes; According to the performance prediction result, the task allocation of the first blockchain node and the second blockchain node is dynamically adjusted to achieve load balancing.

2. The blockchain network dynamic scheduling method based on machine learning according to claim 1 is characterized in that: The step of obtaining a plurality of blockchain nodes to be scheduled from the blockchain cluster based on the dynamic scheduling task generated for the blockchain cluster includes: A monitoring mechanism for the blockchain cluster is pre-set, and the dynamic scheduling tasks generated in the blockchain cluster are continuously monitored through the monitoring mechanism, and the dynamic scheduling tasks are triggered by the cluster manager and the task scheduler; When a new dynamically scheduled task is monitored, the task information of the dynamically scheduled task is recorded, wherein the task information includes the task type, priority, and required resources; Performing demand analysis on the monitored dynamic scheduling tasks to determine the task conditions of the dynamic scheduling tasks for the blockchain nodes, wherein the task conditions include computing power conditions, storage capacity conditions, and network bandwidth conditions; According to the task conditions, evaluating the node resource situation in the blockchain cluster to determine candidate blockchain nodes that meet the task conditions; The screened candidate blockchain nodes are sorted, and different priorities are assigned to the candidate blockchain nodes according to the sorting results, and the information of the screened and sorted candidate blockchain nodes is updated into the node status database.

3. The blockchain network dynamic scheduling method based on machine learning according to claim 1 is characterized in that: The method further comprises: The dynamic scheduling optimization result is sent to the terminal corresponding to the dynamic scheduling task.

4. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores machine-executable instructions, which, when executed by a processor, implement the machine learning-based blockchain network dynamic scheduling method described in any one of claims 1 to 3.

5. A blockchain service system, characterized in that: It includes a processor and a computer-readable storage medium, wherein the computer-readable storage medium stores machine-executable instructions, and when the machine-executable instructions are executed by the processor, the machine learning-based blockchain network dynamic scheduling method described in any one of claims 1 to 3 is implemented.

Images