Network resource optimization method, device, system, computer equipment and computer storage medium

Abstract

The application discloses a network resource optimization method, device, system, computer equipment and computer storage medium, and the method comprises the steps of acquiring physical network information, generating a virtual network topology graph based on the physical network information; the virtual network topology graph comprises nodes, links between the nodes and weights corresponding to the links; determining the shortest path information from a source node to each node in the virtual network topology graph; determining a first time period and a second time period contained in a basic time period according to monitoring requirements and / or network traffic characteristics; monitoring the performance indicators of the first link within the basic time period, performing fault detection on the first link based on the monitored performance indicators, the first time period and the second time period, and obtaining a fault detection result; determining a virtual routing flow table according to the fault detection result, and the virtual routing flow table is used for optimizing and configuring the network equipment and / or the connection between the network equipment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular to a network resource optimization method, device, system, computer equipment and computer storage medium. BACKGROUND

[0002] As a new type of network architecture, cloud network provides efficient and flexible network services for users. At present, with the continuous expansion of cloud network scale and the increasing complexity of application scenarios, the optimal configuration of network resources has become a problem to be solved. Traditional network resource optimization methods are usually based on experience and rules, such as load balancing, resource scheduling, performance optimization, etc., which are difficult to adapt to the dynamic changes and complex demands of cloud network.

[0003] Digital twin technology, as a new technology for simulating physical entities through digital models, provides a new idea and method for cloud network resource optimization. By building a digital twin model of cloud network, real-time monitoring, simulation optimization and intelligent decision-making of network resources can be realized, thereby improving the utilization efficiency and service quality of network resources, constructing accurate virtual models to reflect and control physical networks, achieving efficient resource configuration and performance improvement, and promoting the development of cloud network.

[0004] However, the current network resource optimization method based on digital twin technology lacks response strategies for network sudden events such as dynamic changes of network resources, network traffic surge, and network link performance mutation, which leads to difficulty in predicting potential fault risks in time, and even misjudgment and omission, resulting in unsatisfactory network resource optimization results, and seriously affecting the overall network performance and resource utilization. SUMMARY

[0005] To solve the existing technical problems, the embodiments of the present application provide a network resource optimization method, device, system, computer equipment and computer storage medium.

[0006] To achieve the above purpose, the technical scheme of the embodiments of the present application is as follows:

[0007] In a first aspect, the embodiments of the present application provide a network resource optimization method, which comprises: acquiring physical network information, and generating a virtual network topology graph based on the physical network information; wherein the virtual network topology graph comprises nodes, links between nodes, and weights corresponding to the links; the nodes correspond to network devices in the physical network information; the links correspond to the connection relationship between network devices; and the weights represent the index parameters of the links;

[0008] determining the shortest path information from a source node to each node in the virtual network topology graph; the source node is any node in the virtual network topology graph;

[0009] determine a first time period and a second time period contained in the basic time period according to monitoring requirements and / or network traffic characteristics;

[0010] monitor a performance indicator of a first link in the basic time period, perform fault detection on the first link based on the monitored performance indicator, the first time period and the second time period, and obtain a fault detection result; the first link is a link formed by two nodes contained in the shortest path information;

[0011] determine a virtual routing flow table according to the fault detection result, the virtual routing flow table being used for optimizing configuration of network devices and / or connections between network devices.

[0012] In the above scheme, the physical network information includes network devices in a physical entity network, connection relationships between network devices, and performance indicators corresponding to connections between network devices; the virtual network topology graph is generated based on the physical network information, including: mapping the network devices in the physical network information as nodes, and mapping the connection relationships between the network devices in the physical network information as links between nodes to generate an initial virtual network topology graph; determining weights based on the performance indicators corresponding to the connections between the network devices, and mapping the weights to the links corresponding to the connections between the network devices to generate the virtual network topology graph.

[0013] In the above scheme, the determination of the shortest path information from a source node to each node in the virtual network topology graph includes: obtaining a first array and a second array corresponding to the source node and each node based on the virtual network topology graph, and determining the shortest path information from the source node to each node based on the first array and the second array; wherein the first array includes shortest distances corresponding to shortest paths from the source node to each node, and the second array includes predecessor nodes of each node on the shortest path; the shortest path satisfies the lowest delay and / or the least bandwidth consumption of the link, and the shortest distance represents the corresponding delay and / or bandwidth consumption of the shortest path.

[0014] In the scheme, the first array and the second array corresponding to each node are obtained by: initializing the first array, the second array and a first node set, the first node set including the visited nodes; selecting a first node from a second node set and adding the first node to the first node set, the second node set including the unvisited nodes, and the first node being the node closest to the source node; determining a second node from the adjacent nodes of the first node, and taking the first node as the predecessor node of the second node, the second node satisfying: the sum of the shortest distance between the first node and the source node and the weight between the first node and the second node being less than the shortest distance between the second node and the source node, and the second node not belonging to the first node set; and updating the first array and the second array corresponding to the shortest path from the source node to the second node based on the second node.

[0015] In the scheme, the first mark of the link and / or the second mark of the node in the fault detection result are further included in the virtual network topology graph; the weight of the link associated with the first mark is the maximum value; and the first mark and / or the second mark are used to update the first array and the second array.

[0016] In the scheme, the first mark and / or the second mark are used to update the first array and the second array, including: determining the shortest path between a third node and a fourth node according to the nodes not associated with the second mark, the nodes corresponding to the links not associated with the first mark and the links not associated with the first mark in the virtual network topology graph; the third node being the predecessor node of the node associated with the second mark, and the fourth node being the last node in the link associated with the first mark; and updating the corresponding first array and second array based on the shortest path between the third node and the fourth node, and deleting the first mark and / or the second mark.

[0017] In the scheme, the performance indicators of the first link are monitored in the basic time period, the first link is detected for fault based on the monitored performance indicators, and the fault detection result is obtained, including: obtaining a first performance indicator value of the first link at a first time in the basic time period; obtaining a second performance indicator value of the first time period and a third performance indicator value of the second time period based on the first performance indicator value; and judging whether the first link has a fault in the basic time period based on the second performance indicator value and the third performance indicator value, and obtaining the fault detection result.

[0018] In the solution, the determining whether the first link fails in the basic time period based on the second performance index value and the third performance index value comprises: in a case that the third performance index value is greater than or equal to a first threshold value and the second performance index value is less than the first threshold value, determining that the first link fails and / or determining that a predecessor node in the first link fails.

[0019] In the solution, the performance index comprises one or more of the following: delay amount, packet loss rate, bandwidth utilization.

[0020] In the solution, the determining the virtual routing flow table according to the fault detection result comprises: in a case that the virtual network topology graph is not marked, determining the virtual routing flow table according to the first array and the second array corresponding to the updated shortest path from the source node to each node.

[0021] In a second aspect, an embodiment of the present application further provides a network resource optimization device, the device comprising: a basic environment module, a multi-path redundancy module and a virtual verification module; wherein,

[0022] The basic environment module is configured to acquire physical network information and generate a virtual network topology graph based on the physical network information; wherein the virtual network topology graph comprises nodes, links between the nodes and weights corresponding to the links; the nodes correspond to network devices in the physical network information; the links correspond to connection relationships between the network devices; and the weights represent index parameters of the links.

[0023] The multi-path redundancy module is configured to determine shortest path information from a source node to each node in the virtual network topology graph; the source node is any node in the virtual network topology graph; and the multi-path redundancy module is further configured to determine a virtual routing flow table according to a fault detection result, the virtual routing flow table being used for optimizing and configuring network devices and / or connections between the network devices.

[0024] The virtual verification module is configured to determine a first time period and a second time period included in a basic time period according to monitoring requirements and / or network traffic characteristics; monitor performance indexes of a first link in the basic time period; perform fault detection on the first link based on the monitored performance indexes, the first time period and the second time period, and obtain a fault detection result; and the first link is a link formed by two nodes included in the shortest path information.

[0025] In a third aspect, an embodiment of the present application further provides a network resource optimization system, the system comprising: a physical layer, a connection layer, a twin data layer and an application service layer; wherein,

[0026] The physical layer is configured to provide physical network information, the physical network information including network devices in a physical entity network, connection relationships between the network devices, and performance indexes corresponding to the connections between the network devices.

[0027] The connection layer is configured to transmit the physical network information provided by the physical layer to the twin data layer.

[0028] The twin data layer is configured to generate a virtual network topology graph based on the physical network information, and determine shortest path information from a source node to each node in the virtual network topology graph; the virtual network topology graph includes nodes, links between the nodes, and weights corresponding to the links; the nodes correspond to the network devices in the physical network information; the links correspond to the connection relationships between the network devices; the weights represent index parameters of the links; the source node is any node in the virtual network topology graph; the twin data layer is further configured to determine a first time period and a second time period included in a basic time period according to a monitoring requirement and / or network traffic characteristics; monitor a performance index of a first link in the basic time period, perform fault detection on the first link based on the monitored performance index, the first time period, and the second time period, and obtain a fault detection result; the first link is a link formed by two nodes included in the shortest path information; determine a virtual routing flow table according to the fault detection result, and map the virtual routing flow table to physical network resource optimization information.

[0029] The application service layer is configured to send a monitoring requirement to the twin data layer, and further configured to obtain the physical network resource optimization information from the twin data layer, and perform an optimized configuration on the network devices and / or the connections between the network devices according to the physical network resource optimization information, so as to implement business deployment.

[0030] In a fourth aspect, an embodiment of the present application further provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements steps of the method of the embodiment of the present application when executing the program.

[0031] In a fifth aspect, an embodiment of the present application further provides a computer readable storage medium, which stores a computer program, and the program implements steps of the method of the embodiment of the present application when executed by a processor.

[0032] In a sixth aspect, an embodiment of the present application further provides a computer program product, including a computer program, and the computer program implements the method of the embodiment of the present application when executed by a processor.

[0033] The network resource optimization method, apparatus, system, computer equipment, and computer storage medium provided in this invention generate a virtual network topology map based on acquired physical network information. The virtual network topology map includes nodes corresponding to network devices in the physical network information, and links corresponding to the connection relationships of these network devices. It dynamically adjusts a first time period and a second time period within a basic time period based on detection requirements and / or network traffic characteristics. Based on the first time period, the second time period, and performance indicators obtained by monitoring the first link within the basic time period, it performs fault detection on the shortest path information from the source node to each node in the determined virtual network topology map. A virtual routing flow table is determined based on the fault detection results. By introducing a scalable time period fault detection mechanism based on digital twin technology to achieve a virtual mapping of the physical network, it flexibly adjusts the time range of different stages within the basic time period based on network traffic characteristics or requirements. This allows for accurate capture of sudden changes in link performance based on fault results, enabling rapid and precise responses when network links encounter faults. This achieves optimized configuration and efficient utilization of network resources, reduces false positives and false negatives, improves network reliability and stability, ensures network service continuity, and provides new ideas and methods for network resource optimization based on digital twin technology. Attached Figure Description

[0034] Figure 1 This is a schematic diagram of a network resource optimization system architecture according to an embodiment of the present invention;

[0035] Figure 2 This is a schematic diagram of a platform architecture for a digital twin network side according to an embodiment of the present invention;

[0036] Figure 3 This is a flowchart illustrating the network resource optimization method according to an embodiment of the present invention. Figure 1 ;

[0037] Figure 4 This is a flowchart illustrating the network resource optimization method according to an embodiment of the present invention. Figure 2 ;

[0038] Figure 5 This is a flowchart illustrating an example of a network resource optimization method according to an embodiment of the present invention.

[0039] Figure 6 This is a schematic diagram of the composition structure of the network resource optimization device according to an embodiment of the present invention;

[0040] Figure 7 This is a schematic diagram of the composition structure of the network resource optimization system according to an embodiment of the present invention;

[0041] Figure 8 This is a schematic diagram of the structural operation process of the network resource optimization system according to an embodiment of the present invention;

[0042] Figure 9 This is a schematic diagram of the hardware composition structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0043] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0044] It should be understood that devices with communication functions in the network / system of this application embodiment can be referred to as communication devices. Communication devices may include network devices and terminals with communication functions. Network devices and terminal devices can be the specific devices described above, which will not be repeated here. Communication devices may also include other devices in the communication system, such as network controllers, mobility management entities, and other network entities. This embodiment of the present invention does not limit these.

[0045] It should be understood that the terms "system" and "network" are often used interchangeably in this document. The term "and / or" in this document merely describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Furthermore, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0046] The terms “first,” “second,” etc., used in this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0047] Before providing a detailed description of the access method in the embodiments of the present invention, a brief description of the related technologies will be given first.

[0048] Traditional network resource optimization methods rely on experience and rules, such as load balancing, resource scheduling, and performance optimization. However, due to the dynamic and complex nature of cloud networks—including their elastic scalability, multi-user nature, and complex interrelationships of services—these traditional methods struggle to quickly and accurately respond to resource changes, meet multi-user needs, and achieve global collaborative optimization. Consequently, new, smarter, and more flexible optimization strategies and technologies are urgently needed to adapt to the evolving cloud network environment.

[0049] Digital twin technology is a technique that uses digital models to reflect or simulate real-world entities, systems, or processes. This digital model is a virtual copy capable of comprehensively simulating the state, behavior, and performance of the actual object or system in real time. Digital twin technology integrates multiple advanced technologies such as the Internet of Things, big data analytics, cloud computing, and artificial intelligence. It features virtual-real mapping, real-time monitoring, simulation and prediction, optimized design, and continuous learning. As digital twin technology matures, it is now widely used in intelligent manufacturing, smart cities, transportation, and healthcare, and is one of the key technologies for future intelligent upgrades. It also provides new ideas and methods for optimizing cloud network resources.

[0050] Currently, existing methods for optimizing network resources based on digital twin technology include the following:

[0051] The first method constructs a data mirror model by integrating a general digital twin architecture, hardware device groups, and network data types. It adopts a virtual load data construction method, parsing network data to obtain key elements, packaging them into event packets, and populating them to provide diverse samples and accurate basis for network resource optimization. Based on network latency and preset values, it calculates latency coefficients and adjusts the virtual routing table according to the actual traffic demand value of the latency coefficients and event packet traffic values. This efficiently optimizes network communication, strongly supports the optimization process, ensures stable system operation, and enhances network performance.

[0052] However, in the process of using virtual load data construction, the keyword preset and data filling methods of this method are not flexible enough, making it difficult to accurately simulate complex and ever-changing network scenarios. This affects the universality and accuracy of the optimization scheme and makes it less adaptable to extreme or new network conditions.

[0053] The second approach integrates digital twin technology with reinforcement learning algorithms to achieve network self-optimization in data centers. Specifically, it constructs a system comprising physical and digital twin network layers, collects physical network data to build a foundational and functional model, and uses the Q-learning algorithm—with the traffic matrix as the state, link weights as actions, and the negative maximum link utilization rate as the reward function—to iteratively train and optimize strategies through multi-module interaction. During deployment, it intelligently decides the path algorithm based on traffic volume, and constructs a real-time mirror of the physical network through digital twin network simulation to verify the strategy. This enables precise control and optimization of the physical network, effectively overcoming the training sample difficulties and adverse interference to the physical network inherent in traditional reinforcement learning algorithms, endowing the physical network with simulation capabilities, and improving network management and control.

[0054] However, this method has a complex system architecture, is difficult to develop and maintain, and is costly. Furthermore, when dealing with large-scale network dynamic changes, the model updates are slow and the algorithm adjustments are lagging, making it difficult to guarantee the continuity and quality of network services. Moreover, the system architecture is tightly coupled, making it difficult to combine with new network technologies, which hinders technological evolution and seriously affects the long-term optimization, upgrading and sustainable development of cloud networks.

[0055] The third method applies digital twin technology to communication network fault analysis. It employs unique fully closed test signal parameter processing to obtain the network structure distribution to construct the topology, builds a mirrored digital model, and integrates multi-cycle operational data to predict node data. Based on historical node data, it optimizes the values ​​of constant coefficients to improve prediction accuracy. In the fault determination stage, based on the positive and negative characteristics of operational data, it constructs refined rules with the deviation between measured and predicted data as the core to accurately assess the fault risk level of nodes, achieving intelligent and precise fault detection, improving the reliability and stability of communication networks, reducing the scope and duration of fault impact, and providing strong support for communication network maintenance.

[0056] However, this method involves many process steps, a high degree of technical integration, and a large demand for professional knowledge, from network topology construction to prediction model parameter determination and fault judgment rule design. This makes the implementation and maintenance by technical personnel too difficult and costly.

[0057] The fourth method involves constructing a digital twin network, combining virtual communication device data processing and business intent inference mechanisms. During the business inference process, the network is optimized by changing device parameters and routing algorithms based on the model, and then fed back to the physical network. The digital twin network achieves dynamic network optimization by deploying communication elements, setting up middleware and probe software in collaboration, thereby improving the intelligence of operation and maintenance and the ability to predict and control risks, thus driving the optimization and upgrading of the communication network in all aspects.

[0058] However, this method does not take into account network emergencies such as surges in network traffic, and does not have a pre-set emergency mechanism for traffic surges. This can easily cause delays in data processing and transmission, exacerbate network congestion, drastically increase response time, make it difficult to predict potential faults in a timely manner, greatly affect user experience, waste network resources, and fail to effectively prevent such incidents, thus failing to guarantee network availability.

[0059] It is evident that current network resource optimization methods based on digital twin technology lack strategies to address network emergencies such as dynamic changes in network resources, surges in network traffic, and sudden changes in network link performance. This makes it difficult to predict potential faults in a timely manner, and may even lead to misjudgments and omissions, resulting in unsatisfactory network resource optimization results and seriously affecting overall network performance and resource utilization.

[0060] Based on this, the present invention proposes a network resource optimization system architecture for user platforms. This system architecture can be deployed on a cloud service network, specifically on a mobile cloud virtual network console. For example, based on user business needs, the specific process of optimizing the resource configuration of the physical network using this network resource optimization system architecture is as follows:

[0061] When the system receives user data related to routing table selection, virtual private cloud routing table creation, and subnet product selection on the mobile cloud virtual control service platform, the network resource optimization system will automatically begin executing the network resource optimization process deployed on the digital twin network side. Through an interface-introduced mechanism, the system transmits the intelligent decision-making network resource information generated on the digital twin network side to the cloud network server. The server then issues instructions to transmit the optimized decision information to the corresponding physical interface of the cloud network console. Based on the cloud management and control platform, the system intelligently optimizes and selects network resources from the routing table on the user's web interface and deploys the corresponding network resources in the user's own cloud product network environment.

[0062] Figure 1 This is a schematic diagram of a network resource optimization system architecture according to an embodiment of the present invention; as shown below. Figure 1 As shown, the process of optimizing the resource allocation of physical entity networks based on the network resource optimization system architecture is as follows:

[0063] First, in the early stage of network resource optimization, relevant information of each network device in the physical network is collected, such as hardware information, traffic data information, and connection relationships between each network element. The physical network information is obtained through data filtering to realize the physical physical space for network connection and data traffic processing of each physical network element.

[0064] Secondly, the physical network information is uploaded through the IoT gateway and transmitted to the router connection point on the digital twin network side. The router then forwards the information to the working platform on the virtual digital twin network side of the cloud server. After receiving the physical network information, the platform performs network function modeling and algorithm model invocation. Based on the predicted traffic of the current scalable period, it performs fault detection and management on the network devices in the current period, optimizes network resources, and obtains the updated optimized network resource configuration, i.e., the digital twin network side information.

[0065] Finally, in the later stages of network resource optimization, the working platform on the digital twin network side maps the information on the digital twin network side into post-mapping network resource information, and transmits this information to the router connection point of the cloud management and control platform. Through bidirectional driving judgment of the driving model, the optimized network resource configuration (i.e. post-mapping network resource information) is distributed to the corresponding physical entity routing network information and sent to the physical entity network through flow tables to complete the predetermined traffic task forwarding and cloud service product facility configuration for multiple users.

[0066] based on Figure 1 The network resource optimization system architecture shown is as follows: Figure 2 This is a schematic diagram of a platform architecture for a digital twin network side according to an embodiment of the present invention; as shown. Figure 2 As shown, the platform architecture of the digital twin network mainly consists of four layers, from bottom to top: the physical layer, the connectivity layer, the twin data layer, and the application service layer. The descriptions of each layer are as follows:

[0067] The physical layer is deployed in the physical network element system environment of the real world. In the physical network element system environment, various hardware devices are used to realize control, latest data collection, and storage of historical data. The physical layer is used to obtain the network connection status of each basic network element in the physical entity network and the identifiable traffic environment information of each network element link in the physical world.

[0068] The connection layer, acting as a "bridge" for information exchange between the physical layer and the digital twin layer, can be deployed on routers connecting the physical entity network and the digital twin network. It transmits physical network information collected by the physical layer to the digital twin layer through routing requests, and then transmits the optimized network resource configuration generated by the digital twin layer to the physical layer through flow table distribution, ensuring the synchronization, consistency and integrity of information between the physical and virtual layers.

[0069] In the digital twin network platform architecture of this invention, the twin data layer is at the core level, deploying multiple sub-models based on digital twin technology, such as data mapping, data warehouse, and service mapping models. Data mapping is used to map physical network information collected from the physical layer into virtual data. The data warehouse has data collection, storage, management, and service functions, primarily responsible for storing the mapped virtual network data. The service mapping model includes a basic sub-model and functional sub-models. The basic sub-model is mainly responsible for collecting and storing various network data, providing the functional sub-model with various data services (such as network resource optimization, predictive fault diagnosis, predictive maintenance, etc.) and a unified interface, modeling based on data and providing data model examples for various network applications. The functional sub-model has functions such as data maintenance, prediction, optimization, management, and operation. Based on the data service interface provided by the basic sub-model, it achieves network resource optimization configuration of virtual network data through iterative algorithm optimization, and realizes simulation verification of network resource optimization configuration through steps such as scalable time-period predictive operation verification and predictive fault diagnosis. This layer of functionality maximizes the agility and programmability of network services, enabling higher-level maintenance, optimization, and intelligent optimization.

[0070] The application service layer is deployed on the cloud service network. It is mainly responsible for transmitting the user's business requirements to the twin data layer. After fully verifying the network optimization strategy, it deploys the business in a model-based manner according to the optimized network resource configuration provided by the twin data layer.

[0071] In this embodiment of the invention, a four-layer architecture is designed using twin digital virtual routing table dynamic adjustment technology. The functions of each layer are reasonably designed and the layers cooperate closely, fully supporting cloud network resource optimization. From basic resources to business mapping applications, it realizes efficient interaction and intelligent control as well as real-time dynamic adjustment of routing policies.

[0072] Based on the aforementioned network resource optimization architecture and the platform architecture of the digital twin network side, this embodiment of the invention proposes a network resource optimization method. Figure 3 This is a flowchart illustrating the network resource optimization method according to an embodiment of the present invention. Figure 1 ;like Figure 3 As shown, the method includes:

[0073] Step 101: Obtain physical network information and generate a virtual network topology map based on the physical network information; wherein, the virtual network topology map includes nodes, links between nodes, and weights corresponding to the links; the nodes correspond to the network devices in the physical network information; the links correspond to the connection relationships between the network devices; the weights represent the index parameters of the links;

[0074] Step 102: Determine the shortest path information from the source node to each node in the virtual network topology graph; the source node is any node in the virtual network topology graph;

[0075] Step 103: Determine the first and second time periods included in the basic time period based on monitoring needs and / or network traffic characteristics;

[0076] Step 104: Monitor the performance indicators of the first link within the basic time period, and perform fault detection on the first link based on the monitored performance indicators, the first time period, and the second time period to obtain the fault detection results; the first link is the link formed by the two nodes included in the shortest path information.

[0077] Step 105: Determine the virtual routing flow table based on the fault detection results. The virtual routing flow table is used to optimize the configuration of network devices and / or the connections between network devices.

[0078] In this embodiment, physical network information related to the actual physical network is obtained, and twinning processing is performed on the physical network information to generate a virtual network topology map corresponding to the physical network.

[0079] Among them, physical network information is information collected based on the physical network element system environment in the real world, including various types of information such as real-time collected network information and / or stored historical data, and may also include network information filtered based on user business needs.

[0080] The virtual network topology map corresponds to the actual physical network, specifically including multiple nodes, links between nodes, and weights corresponding to each link; each node in the virtual network topology map corresponds to a network device in the physical network information; the links between nodes correspond to the connection relationships between network devices in the actual physical network; the weights are used to represent the index parameters of the links, specifically determined based on the link's latency and / or bandwidth consumption metrics.

[0081] In this embodiment, after obtaining the virtual network topology map corresponding to the physical network information, the shortest path information from the source node to each node in the virtual network topology map is determined, providing data information for the subsequent construction of the virtual routing flow table, thereby effectively guiding the efficient transmission of data in the network.

[0082] The source node can be any node in the virtual network topology graph, or it can be a specific node in the virtual network topology graph determined based on the obtained user's business requirements.

[0083] In this embodiment, after obtaining the shortest path information from the source node to each node in the virtual network topology graph, the first time period and the second time period included in the basic time period are dynamically adjusted according to monitoring requirements and / or network traffic characteristics, so as to enable subsequent fault detection of the shortest path information.

[0084] Among them, monitoring demand refers to the real-time monitored traffic demand or the predicted traffic demand value. The embodiments of the present invention do not impose specific restrictions on the method of monitoring traffic demand or predicting traffic demand value; network traffic characteristics refer to the network traffic fluctuation, such as large network traffic fluctuation caused by a surge or drop in network traffic, or small network traffic fluctuation caused by low network traffic demand.

[0085] The first time period refers to the first time period in the basic time period, and the second time period refers to the second time period in the basic time period. The first time period and the second time period dynamically form a complete basic time period, that is, the basic time period is divided into two parts: the first time period and the second time period.

[0086] For example, Based on the base time period length, within the base time period Based on this, define the first time period (i.e., the preceding time period). The second time period (i.e., the later time period) This constitutes a scalable time period, which can be flexibly adjusted according to monitoring needs and / or network traffic characteristics. and ,satisfy .

[0087] It should be noted that when network traffic fluctuations are large, it is necessary to detect the latest link in the base time period. Therefore, the adjustment to the first and second time periods is to shorten the first time period and extend the second time period to more accurately capture the link performance change trend. When network traffic fluctuations are small, the detection results of the previous and subsequent time periods are relatively similar, and it is not necessary to detect the latest link. Therefore, the adjustment to the first and second time periods is to extend the first time period and shorten the second time period to obtain the stability of link performance.

[0088] In this embodiment, after determining the first time period and the second time period in the basic time period, the performance indicators of the first link formed by the two nodes contained in the shortest path information are monitored within the basic time period. Based on the monitored performance indicators, the first time period and the second time period, the first link is fault detected to determine the faulty link and / or faulty node in the shortest path and obtain the corresponding fault detection results.

[0089] It should be noted that since a failure at a node in the shortest path will cause a failure in the link formed by that node and subsequent nodes, when a faulty node in the shortest path is identified, it can be further determined that the link starting from that faulty node is also faulty, i.e., the faulty link in the shortest path is identified. In other words, the fault detection result corresponding to the faulty link in this embodiment of the invention is similar to the fault detection result corresponding to the faulty node.

[0090] In this embodiment, after obtaining the fault detection result, a virtual routing flow table is determined based on the fault detection result. The network devices and / or the connection relationships between network devices are optimized based on the virtual routing flow table, thereby realizing the optimized configuration of physical network resources.

[0091] In this embodiment of the invention, a virtual network topology map is generated based on the acquired physical network information. The first and second time periods within the basic time period are dynamically adjusted according to detection requirements and / or network traffic characteristics. Based on the first and second time periods, and performance indicators obtained from monitoring the first link within the basic time period, fault detection is performed on the shortest path information from the source node to each node in the determined virtual network topology map. A virtual routing flow table is determined based on the fault detection results. Building upon the use of digital twin technology to achieve a virtual mapping of the physical network, a scalable fault detection mechanism is introduced. The time range of different stages within the basic time period is flexibly adjusted based on network traffic characteristics or requirements. This allows for accurate capture of sudden changes in link performance based on fault results, enabling rapid and precise responses when network links encounter faults. This achieves optimized allocation and efficient utilization of network resources, reduces false positives and false negatives, improves network reliability and stability, and ensures network service continuity.

[0092] In some optional implementations, the physical network information includes network devices in the physical entity network, the connection relationships between network devices, and the performance indicators corresponding to the connections between network devices; generating a virtual network topology based on the physical network information includes: mapping the network devices in the physical network information to nodes, and mapping the connection relationships between the network devices in the physical network information to links between nodes, to generate an initial virtual network topology; determining weights based on the performance indicators corresponding to the connections between network devices, and mapping the weights to the links corresponding to the connections between the network devices, to generate the virtual network topology.

[0093] In this embodiment, physical network information includes network devices (e.g., routers) in a physical entity network (e.g., a cloud server network), the connection relationships between network devices, and the performance indicators corresponding to the connections between network devices. The performance indicators include indicators such as latency and bandwidth consumption between network devices.

[0094] For example, the process of generating a network topology map corresponding to a physical network is as follows: mapping network devices in the physical network information to nodes to obtain a total set of nodes. V The total number of nodes Mapping the connections between network devices in the physical network information to links between nodes yields a set of links. E Based on the performance metrics corresponding to the connections between network devices, each link formed between two nodes with a connection relationship is defined. Assign a corresponding weight The weights are mapped to the links corresponding to the connections between network devices, thus obtaining a virtual network topology that includes all nodes and all links. G ( V, E It should be noted that the embodiments of the present invention do not impose specific limitations on the method for determining the weight of a link based on performance indicators.

[0095] In some optional implementations, determining the shortest path information from the source node to each node in the virtual network topology graph includes: obtaining a first array and a second array corresponding to the source node and each node based on the virtual network topology graph; and determining the shortest path information from the source node to each node based on the first array and the second array; wherein the first array includes the shortest distance corresponding to the shortest path from the source node to each node, and the second array includes the predecessor node of each node on the shortest path; the shortest path satisfies the minimum link latency and / or minimum bandwidth consumption, and the shortest distance represents the corresponding latency and / or bandwidth consumption of the shortest path.

[0096] In this embodiment, the process of determining the shortest path information from the source node to each node in the virtual network topology graph is as follows: obtain the first array and the second array corresponding to the links between the source node and each node in the virtual network topology graph, and determine the shortest path information between the source node and each node based on the first array and the second array.

[0097] The first array is a distance array, specifically including the shortest distances corresponding to the shortest paths from the source node to each node in the virtual network topology graph. Each shortest path corresponds to a first array. In this embodiment, the distance represents the path's latency and / or bandwidth consumption. That is, the shortest path refers to the link between the source node and each node with the lowest latency and / or the least bandwidth consumption.

[0098] The second array is a predecessor node array, specifically including the predecessor node of each node on the shortest path, that is, the node before each node on the shortest path. In this embodiment of the invention, there is no specific restriction on the storage method of the second array; each shortest path can correspond to a sequence of predecessor nodes.

[0099] Figure 4This is a flowchart illustrating the network resource optimization method according to an embodiment of the present invention. Figure 2 ;like Figure 4 As shown, obtaining the source node and the first and second arrays corresponding to each node includes:

[0100] Step 201: Initialize the first array, the second array, and the first node set, where the first node set includes the visited nodes.

[0101] In this embodiment, during the process of obtaining the source node and the first and second arrays corresponding to each node, it is first necessary to initialize the data structure, specifically including the initialization of the first array, the second array, and the first node set including the visited nodes.

[0102] For example, initialize the source node to each node. ( V The first array corresponding to the total set of nodes) D Set source node The corresponding first array Configure other nodes The corresponding first array .

[0103] Initialize the source node to the second array corresponding to each node. P Except for the source node s In addition, all other nodes v The corresponding second array is empty ( Null ).

[0104] Initialize the first node set S An empty set indicates that there are currently no visited nodes.

[0105] Step 202: Select a first node from the second node set and add the first node to the first node set; wherein the second node set includes unvisited nodes and the first node is the node closest to the source node.

[0106] In this embodiment, after initializing the first array, the second array, and the first node set, if there are nodes in the second node set including unvisited nodes, the first node closest to the source node is selected from the second node set and added to the first node set.

[0107] For example, when the second node set I= ( V-S There are unvisited nodes in the total node set (i.e., the total node set). V First node set S The number of nodes in the data is not equal to the total number of nodes. When ), select a node from the second set.u , making That is, select a node u This makes the node u Compared to other currently unvisited nodes and the source node s The node with the shortest distance will be selected. u Add to the first node set S In, that is .

[0108] Step 203: Determine a second node among the adjacent nodes of the first node, and take the first node as the predecessor node of the second node; wherein, the second node satisfies the following conditions: the sum of the shortest distance between the first node and the source node and the weights between the first node and the second node is less than the shortest distance between the second node and the source node, and the second node does not belong to the set of the first nodes.

[0109] In this embodiment, after selecting the first node, the second node is selected as the next node of the first node from all the adjacent nodes of the first node in the second set, that is, the first node is selected as the predecessor node of the second node.

[0110] Specifically, the method for determining the second node is to calculate the sum of the shortest distance between the first node and the source node and the weights between the first node and each of its adjacent nodes. If the sum is less than the shortest distance between the adjacent node and the source node, and the adjacent node is not in the set of the first nodes, then the adjacent node is determined to be the second node.

[0111] For example, for the first node u Each adjacent node ,like ,and That is, the first node u and source node s shortest distance between D [ u ] and the first node s and each adjacent node v Weights between Less than the shortest distance between the adjacent node and the source node D [ v ], then the adjacent node v As the second node, the first node u As the second node v The predecessor node.

[0112] Step 204: Update the first array and the second array corresponding to the shortest path from the source node to the second node based on the second node.

[0113] In this embodiment, after determining the second node, the shortest path between the source node and the second node is determined, and the first array and the second array corresponding to the shortest path from the source node to the second node are updated.

[0114] For example, based on the source node s To the first node u shortest distance D[u] and from the first node to the second node v weight Update the first array corresponding to the shortest path from the source node to the second node. The first node is about to be completed. u and source node s shortest distance between D [ u ] and the first node s and each adjacent node v Weights between The sum is updated to the shortest distance between the adjacent node and the source node. D [ v Based on the second array corresponding to the shortest path from the source node to the first node, the first node is used as the predecessor node of the second node. This is added to the second array corresponding to the shortest path from the source node to the first node, thereby updating the second array corresponding to the shortest path from the source node to the second node.

[0115] It should be noted that, in this embodiment of the invention, the Dijkstra algorithm is used to repeatedly perform the algorithm iteration processing from the source node to each node in steps 201 to 204, thereby obtaining the first array and the second array corresponding to the shortest path between the source node and each node.

[0116] In this embodiment, by running the algorithm in steps 201 to 204 above, the algorithm parameters are optimized based on the virtual network topology graph after virtual mapping, the shortest path is calculated by adjusting the link weight according to the link evaluation index, and the algorithm model is optimized based on the execution result of the algorithm to achieve dynamic bidirectional optimization, thereby improving network resources and enabling the network state to be dynamically adjusted to maintain the optimal operating state.

[0117] In some optional implementations, the virtual network topology graph further includes a first marker of the faulty link in the fault detection result, and / or a second marker of the faulty node; wherein the weight of the link associated with the first marker is the maximum value; the first marker and / or the second marker are used to update the first array and the second array.

[0118] In this embodiment, after determining the first array and the second array corresponding to the shortest path from the source node to each node, fault detection is performed on the shortest path. After obtaining the corresponding fault detection results, the faulty links and / or faulty nodes determined in the fault detection results are marked, so that the virtual network topology graph includes the first mark of the faulty link in the fault detection results and / or the second mark of the faulty node.

[0119] Here, the first and / or second markers indicate that the current link and / or node is in an update-pending state. That is, the first and / or second markers are used to update the first and second arrays corresponding to the shortest paths associated with the marked links and / or nodes. Since there are faulty links and / or faulty nodes in the virtual network topology graph, it is necessary to adjust the weight of the faulty links to the maximum value. .

[0120] It should be noted that, for multiple shortest paths, when a faulty node and / or faulty link is identified in one of the shortest paths, the information related to that faulty node and / or faulty link in the first and second arrays corresponding to the other shortest paths will be marked.

[0121] In some optional implementations, the first tag and / or the second tag are used to update the first array and the second array, including: determining the shortest path between a third node and a fourth node based on nodes in the virtual network topology graph that are not associated with the second tag, nodes corresponding to links that are not associated with the first tag, and links that are not associated with the first tag; wherein the third node is a predecessor node of a node associated with the second tag, and the fourth node is the end node in a link associated with the first tag; updating the corresponding first array and second array based on the shortest path between the third node and the fourth node, and deleting the first tag and / or the second tag.

[0122] In this embodiment, the process of updating the first and second arrays corresponding to the shortest paths associated with the marked links and / or nodes is as follows: determining the nodes in the virtual network topology graph that are not associated with the second mark, the nodes corresponding to the links that are not associated with the first mark, and the links that are not associated with the first mark, that is, determining all nodes in the virtual network topology graph except for the faulty nodes and / or the nodes corresponding to the faulty links, and all links in each shortest path except for the faulty links, and determining the shortest path between the third node and the fourth node based on the above nodes and links.

[0123] The third node represents the predecessor node of the node corresponding to the second mark of the faulty node, and the fourth node represents the end node in the first mark of the faulty link.

[0124] For example, with the source nodes To the end node k The shortest path between e =( s , k ), that is, the complete link is ( s,..., h, i, j,... k The first marked faulty link in ) and the second marked fault node i For example, the third node is the faulty node. i predecessor node h The fourth node is a faulty link. End node j Remove nodes from the virtual network topology graph i and nodes j The nodes outside the link are treated as nodes not associated with the second tag and nodes corresponding to links not associated with the first tag, resulting in the corresponding node set. The current shortest path is then defined. Troubleshooting Link External links (i.e.) and For links not associated with the first tag, obtain the corresponding link set. Based on the above node set and link set, determine the links. The shortest path.

[0125] It should be noted that after determining the third and fourth nodes, the shortest path between the third and fourth nodes can be determined by repeatedly iterating through steps 201 to 204.

[0126] In this embodiment, after determining the shortest path between the third node and the fourth node, the corresponding first array and second array are updated based on the shortest path, including information related to the current faulty node and / or faulty link in the first array and second array corresponding to other shortest paths, and the corresponding first tag and / or second tag are deleted after the update is completed.

[0127] In some optional implementations, the step of monitoring the performance indicators of the first link within the base time period, and performing fault detection on the first link based on the monitored performance indicators to obtain fault detection results includes: obtaining a first performance indicator value of the first link at a first moment within the base time period; obtaining a second performance indicator value for the first time period and a third performance indicator value for the second time period based on the first performance indicator value; determining whether the first link has failed within the base time period based on the second performance indicator value and the third performance indicator value, and obtaining fault monitoring results.

[0128] In this embodiment, the process of obtaining the fault detection result by detecting the first link within the basic time period is as follows: by monitoring the performance index of the first link at the first moment within the basic time period, the first performance index value at the first moment is obtained; based on the first performance index value, the performance index values ​​of the first time period and the second time period are predicted respectively to obtain the corresponding second performance index value and third performance index value; and based on the second performance index value and the third performance index value, it is determined whether the first link has failed within the basic time period, thereby obtaining the fault detection result.

[0129] For example, by analyzing the link e= ( i, j During the base period The first moment inside t Performance metrics were monitored to obtain the corresponding primary performance metric. The first time period is predicted using a differential method. The average performance index value is used to obtain the second performance index value. The prediction method for the second performance index value is designed as follows:

[0130] (1)

[0131] in, i and j These represent the preceding and following nodes in the current fault detection link, respectively. T The maximum moment of the base time period; This indicates the first time period, which is the preceding time period set within the basic time period.

[0132] As shown in formula (1), the first performance index is described by differentiation. from Time to t Time (i.e., the first time period) The linear change within (within) is then divided by the first time period. This yields the average performance index value of the first time period within the base time period, which is the second performance index value.

[0133] Meanwhile, the second time period is predicted using a differential method. The average performance index value is used to obtain the third performance index value. The prediction method for the third performance index value is designed as follows:

[0134] (2)

[0135] in, i and j These represent the preceding and following nodes in the current fault detection link, respectively. T Basic time period The maximum moment; t The first moment in the basic time period; This refers to the first time period, also known as the basic time period. The preceding time period is set in the middle; This indicates the second time period, i.e., the base time period. The time period set in the middle.

[0136] As shown in formula (2), the first performance index is described by differentiation. from Time to Time (i.e., the second time period) The linear change within (within) is then divided by the second time period. This yields the average performance index value of the later time period within the base time period, i.e., the third performance index value.

[0137] In some optional implementations, determining whether the first link has failed within the basic time period based on the second performance index value and the third performance index value includes: determining that the first link has failed when the third performance index value is greater than or equal to a first threshold and the second performance index value is less than the first threshold, and / or determining that the predecessor node in the first link has failed.

[0138] In this embodiment, after predicting the performance index values ​​of the first link in the first time period and the second time period within the basic time period, the second performance index value and the third performance index value are compared with the preset first threshold. If the third performance index value is greater than or equal to the first threshold and the second performance index value is less than the first threshold, it is determined that the first link has failed, and / or, it is determined that the predecessor node in the first link has failed. The specific cause of the failure is related to the monitored performance index.

[0139] For example, for a link e= ( i, j ), with a preset first threshold For example, based on the above formulas (1) and (2), the conditions for determining whether a link and / or node has failed during the basic time period can be designed as follows:

[0140] ,and (3)

[0141] in, i and j These represent the preceding and following nodes in the current fault detection link, respectively. T Basic time period The maximum moment; t The first moment in the basic time period; This refers to the first time period, also known as the basic time period. The preceding time period is set in the middle; This indicates the second time period, i.e., the base time period. The time period set in the middle; Indicates the primary performance indicator; This is the preset first threshold.

[0142] As shown in formula (3), when the value of the third performance index is greater than or equal to the first threshold and the value of the second performance index is less than the first threshold, the link is determined. e= ( i, j A fault occurs, and / or, determine the node. i A malfunction has occurred.

[0143] In some alternative implementations, the performance metrics include one or more of the following: latency, packet loss rate, and bandwidth utilization.

[0144] In this embodiment, the performance metrics monitored for the first link include one or more of the following metrics: latency, packet loss rate, and bandwidth utilization. In other words, the first performance metric value of the first link monitored at the first moment within the scalable base time period includes one or more of the following metrics: latency, packet loss rate, and bandwidth utilization.

[0145] Correspondingly, the second performance index value corresponding to the first time period predicted based on the first performance index value, and the third performance index value corresponding to the second time period, also include one or more of the following: latency, packet loss rate, and bandwidth utilization.

[0146] It should be noted that this invention does not impose specific restrictions on the method of obtaining the latency, packet loss rate, and bandwidth utilization at the first moment within the basic time period.

[0147] For example, the first performance index value Specifically, this can include performance metrics that represent latency (i.e., the amount of latency). The performance index value representing the packet loss rate , representing the performance index value of bandwidth utilization One or more of them.

[0148] Accordingly, in determining whether a link failure and / or node failure has occurred on the shortest path based on the second and third performance index values, the judgment conditions used include one or more rules related to one or more of latency, packet loss rate, and bandwidth utilization. The corresponding first threshold is related to one or more of latency, packet loss rate, and bandwidth utilization. In other words, if one or more of the judgment conditions are met, it can be determined that a link failure and / or node failure has occurred on the shortest path.

[0149] For example, taking the above formula (3) as an example, the first threshold set may include a delay threshold. Packet loss rate threshold and bandwidth utilization threshold The judgment conditions may specifically include one or more of the following rules:

[0150] ,and (4)

[0151] ,and (5)

[0152] ,and (6)

[0153] Among them, among them, i and j These represent the preceding and following nodes in the current fault detection link, respectively. T Basic time period The maximum moment; t Basic time period The first moment in the middle; This refers to the first time period, also known as the basic time period. The preceding time period is set in the middle; This indicates the second time period, i.e., the base time period. The time period set in the middle; The primary performance metric related to latency; This represents the primary performance metric related to packet loss rate; This represents the primary performance metric related to bandwidth utilization.

[0154] Based on formula (4), when the value of the second performance index representing the average delay of the previous time period is less than the delay threshold, And the third performance index value, representing the average latency in the subsequent time period, is greater than or equal to the latency threshold. In this case, determine the link e= ( i, j A delay failure occurred, and / or, the node was identified. i A delay fault occurred.

[0155] Based on formula (5), the second performance index value, which represents the average packet loss rate of the previous time period, is less than the packet loss rate threshold. Furthermore, the third performance metric value, representing the average packet loss rate over the subsequent time period, is greater than or equal to the packet loss rate threshold. In this case, determine the link e= ( i, jA packet loss rate failure occurred, and / or, the node was identified. i A packet loss rate failure occurred.

[0156] Based on formula (6), the second performance index value, which represents the average bandwidth utilization rate of the previous time period, is less than the bandwidth utilization rate threshold. Furthermore, the third performance metric value, representing the average bandwidth utilization over the subsequent time period, is greater than or equal to the bandwidth utilization threshold. In this case, determine the link e= ( i, j A bandwidth utilization failure occurred, and / or, the node was identified. i A bandwidth utilization failure has occurred.

[0157] In this embodiment, a fault diagnosis mechanism with scalable time periods is introduced. Based on monitoring performance indicators within a time range before and after dynamic network adjustments, preset thresholds are used to determine fault types, accurately capturing sudden changes in link performance and reducing false positives and false negatives. Compared to fixed thresholds or other simple monitoring methods, this method can accurately locate faults under complex network fluctuations, such as accurately identifying faulty links during network traffic peaks, buying time for rapid repair, improving network reliability and stability, ensuring network service continuity, effectively reducing false positives and false negatives, and improving the accuracy and timeliness of fault detection.

[0158] In some optional implementations, determining the virtual routing flow table based on the fault detection result includes: when there are no markers in the virtual network topology graph, obtaining the virtual routing flow table based on the first array and the second array corresponding to the shortest paths from the source node to each node after the update.

[0159] In this embodiment, after completing the fault detection of the shortest links between the source node and each node in the virtual network topology graph, the information of the marked fault links and / or fault nodes in the fault detection results is updated. By checking whether there are first and / or second marks in the first and second arrays corresponding to the shortest links between the source node and each node in the virtual network topology graph, if it is determined that there are still marks in the virtual network topology graph, the corresponding first and second arrays are updated for the marked links and / or nodes until it is determined that there are no marks in the virtual network topology graph.

[0160] If no marker is found in the virtual network topology diagram, the virtual link verification is confirmed to be successful. Based on the first and second arrays corresponding to the shortest paths from the source node to each node after the update, the updated virtual routing flow table is output. Subsequently, the updated routing flow table data will be distributed to the actual cloud service network side through virtual mapping to realize the status update of the connection relationship between physical network elements and network elements. The service platform side will update the intelligently optimized information in real time and provide it to the user to generate the task node.

[0161] In this embodiment, a fault detection mechanism is used to adjust routes, marking faulty links and / or faulty nodes to update node information and obtain an updated virtual routing flow table. This mechanism ensures network connectivity and data transmission reliability, rapidly adjusts routes to avoid faulty links and / or faulty nodes in complex multi-link failure scenarios, balances network load, maintains network functional stability, reduces the impact of faults on network services, reduces operation and maintenance costs and service interruption losses, and improves overall network resilience and service quality.

[0162] As an example, Figure 5 This is a flowchart illustrating an example of a network resource optimization method according to an embodiment of the present invention; such as Figure 5 As shown, the specific process for optimizing network resources is as follows:

[0163] Step 301: Obtain physical network information.

[0164] Specifically, by collecting data from the physical network element system environment in the real world, physical network information related to the actual physical network is obtained. Physical network information includes network devices in the physical network, the connection relationships between network devices, and the performance indicators corresponding to the connections between network devices. Performance indicators include indicators such as latency and bandwidth consumption between network devices.

[0165] Step 302: Generate a virtual network topology map based on physical network information.

[0166] Specifically, the physical network information is virtually mapped, and the network devices in the physical network information are mapped as nodes to obtain a set of nodes. The connection relationships between the network devices in the physical network information are mapped as links between nodes to obtain a set of links. Based on the performance indicators corresponding to the connection between the network devices, each link between two nodes with a connection relationship is assigned a corresponding weight. The weights are then mapped to the links corresponding to the connection relationships between the network devices, thereby obtaining a virtual network topology map that includes all nodes and all links.

[0167] Step 303: Determine the shortest path information from the source node to each node in the virtual network topology graph.

[0168] Specifically, initialize the first array, the second array, and the first node set including visited nodes. Select the first node closest to the source node from the second node set including unvisited nodes. Add the first node to the first node set including visited nodes. Determine the second node from the nodes adjacent to the first node, such that the second node satisfies the following conditions: the sum of the shortest distance between the first node and the source node and the weights between the first node and the second node is less than the shortest distance between the second node and the source node, and the second node does not belong to the first node set. Use the first node as the predecessor node of the second node. Update the first array and the second array corresponding to the shortest path from the source node to the second node based on the second node.

[0169] Step 304: Determine the first time period and the second time period in the basic time period, monitor the performance index of the first link at the first moment in the basic time period, and predict the second performance index value corresponding to the first time period and the third performance index value corresponding to the second time period based on the obtained first performance index value.

[0170] Specifically, the first and second time periods included in the basic time period are dynamically adjusted according to monitoring needs and / or network traffic characteristics. The performance indicators of the first link at the first moment within the basic time period are monitored to obtain the first performance indicator value at the first moment. Based on the first performance indicator value, the performance indicator values ​​of the first and second time periods are predicted to obtain the corresponding second and third performance indicator values.

[0171] The first performance metric value of the first link monitored at the first moment within the scalable base time period includes one or more of latency, packet loss rate, and bandwidth utilization. Correspondingly, the second performance metric value predicted based on the first performance metric value for the first time period, and the third performance metric value predicted based on the second time period, also include one or more of latency, packet loss rate, and bandwidth utilization.

[0172] Step 305: Based on the second and third performance index values, determine whether the shortest path has failed.

[0173] Specifically, if the second performance index value is less than the preset first threshold and the third performance index value is greater than or equal to the preset first threshold, it is determined whether the first link has a link fault and / or node fault within the basic time period, thereby obtaining the fault detection result.

[0174] The judgment conditions used in the fault detection process include one or more rules related to one or more of the latency, packet loss rate and bandwidth utilization. The corresponding first threshold is related to one or more of the latency, packet loss rate and bandwidth utilization. If one or more of the judgment conditions are met, it can be determined that the shortest path has a link failure and / or node failure.

[0175] Step 306: Mark the fault based on the fault detection results.

[0176] Specifically, when a fault detection result is obtained that there is a faulty link and / or a faulty node in the current shortest path, the faulty link and / or faulty node identified in the fault detection result is marked, so that the virtual network topology graph includes a first mark of the faulty link in the fault detection result and / or a second mark of the faulty node. By repeating steps 303 to 305, the shortest path of the faulty link is re-determined. Based on the re-determined shortest path, the first array and the second array corresponding to the shortest path associated with the marked link and / or node are updated, and the corresponding first mark and / or second mark are deleted.

[0177] Step 307: Determine if there is a fault in the current shortest path.

[0178] Specifically, by querying whether there are any markers to be updated in the current shortest path, it can be determined whether there are faulty links and / or faulty nodes in the current shortest path.

[0179] If there are still markers to be updated on the current path, the shortest path of the faulty link is re-determined by repeating steps 303 to 306. The first and second arrays corresponding to the shortest paths associated with the marked links and / or nodes are updated according to the re-determined shortest paths, and the corresponding first and / or second markers are deleted.

[0180] Step 308: Determine the first and second arrays corresponding to the current path information.

[0181] Specifically, if it is determined that there are no markers to be updated for the current shortest path, store the first array and the second array corresponding to the updated current shortest path.

[0182] Step 309: Check for fault markers.

[0183] Specifically, query whether there are any markers to be updated for the shortest paths between the source node and each node in the virtual network topology graph. If there are still markers to be updated in the virtual network topology graph, repeat steps 303 to 308 to redetermine the shortest path of the faulty link. Update the first and second arrays corresponding to the shortest paths related to the marked links and / or nodes according to the redetermined shortest paths, and delete the corresponding first and / or second markers.

[0184] Step 310: Determine the virtual routing flow table.

[0185] Specifically, if there are no markers in the virtual network topology graph, the virtual link verification is successful. Based on the first and second arrays corresponding to the updated shortest paths from the source node to each node, the updated virtual routing flow table is output.

[0186] This invention integrates the Dijkstra algorithm with a fault detection feedback mechanism and introduces scalable time-segmentation logic, providing a new approach and method for network link fault diagnosis and network resource optimization. Through refined digital twin modeling of the network topology, efficient execution of the Dijkstra algorithm, accurate fault detection based on scalable time periods, and intelligent route adjustment after fault feedback, this method can promptly update the virtual routing flow table when a network link encounters a fault and respond quickly and accurately. This allows the network to cleverly bypass faulty links or nodes for data transmission, ensuring network connectivity and the reliability and efficiency of data transmission. It also optimizes network resource allocation and efficient utilization, significantly improving overall network performance and fault tolerance.

[0187] Furthermore, since the interaction between the virtual digital twin space and the physical space occurs continuously during the network resource optimization process, it is necessary to adopt technologies such as cloud computing, which have efficient data interaction capabilities, real-time transmission capabilities, and decision-making modules with excellent computing performance and the ability to process large amounts of data in the virtual digital twin space, so that the simulation effect can better match the actual application scenario and achieve real-time response and dynamic data feedback between the virtual and physical spaces.

[0188] Therefore, the network resource optimization method provided by the embodiments of the present invention has extremely broad application prospects in diverse network environments such as cloud computing networks and enterprise intranets, and is of great significance for ensuring the stability and continuity of network services and meeting the ever-increasing network communication needs.

[0189] Based on the above embodiments, this invention also provides a network resource optimization device. Figure 6 This is a schematic diagram of the composition structure of the network resource optimization device according to an embodiment of the present invention; as shown below. Figure 6As shown, the device includes: a basic environment module 41, a multi-path redundancy module 42, and a virtual verification module 43; wherein,

[0190] The basic environment module 41 is used to acquire physical network information and generate a virtual network topology map based on the physical network information; wherein, the virtual network topology map includes nodes, links between nodes, and weights corresponding to the links; the nodes correspond to network devices in the physical network information; the links correspond to the connection relationships between network devices; and the weights represent the index parameters of the links;

[0191] The multi-path redundancy module 42 is used to determine the shortest path information from the source node to each node in the virtual network topology graph; the source node is any node in the virtual network topology graph; it is also used to determine a virtual routing flow table based on the fault detection result, the virtual routing flow table being used to optimize the configuration of network devices and / or the connections between network devices;

[0192] The virtual verification module 43 is used to determine the first time period and the second time period included in the basic time period according to monitoring requirements and / or network traffic characteristics; to monitor the performance indicators of the first link within the basic time period; to perform fault detection on the first link based on the monitored performance indicators, the first time period and the second time period, and to obtain fault detection results; the first link is a link formed by two nodes included in the shortest path information.

[0193] In an optional embodiment of the present invention, the physical network information includes network devices in the physical entity network, the connection relationships between network devices, and the performance indicators corresponding to the connections between network devices; the basic environment module 41 is used to map the network devices in the physical network information as nodes, and to map the connection relationships between the network devices in the physical network information as links between nodes, thereby generating an initial virtual network topology; and to determine weights based on the performance indicators corresponding to the connections between network devices, and to map the weights to the links corresponding to the connections between the network devices, thereby generating the virtual network topology.

[0194] In an optional embodiment of the present invention, the multi-path redundancy module 42 is used to obtain a first array and a second array corresponding to the source node and each node based on the virtual network topology map, and to determine the shortest path information between the source node and each node based on the first array and the second array; wherein, the first array includes the shortest distance corresponding to the shortest path from the source node to each node, and the second array includes the predecessor node of each node on the shortest path; the shortest path satisfies the minimum link latency and / or minimum bandwidth consumption, and the shortest distance represents the corresponding latency and / or bandwidth consumption of the shortest path.

[0195] In an optional embodiment of the present invention, the multi-path redundancy module 42 is used to initialize a first array, a second array, and a first node set, wherein the first node set includes visited nodes; select a first node from the second node set and add the first node to the first node set; wherein the second node set includes unvisited nodes, and the first node is the node closest to the source node; determine a second node among the adjacent nodes of the first node, and use the first node as the predecessor node of the second node; wherein the second node satisfies: the sum of the shortest distance between the first node and the source node and the weights between the first node and the second node is less than the shortest distance between the second node and the source node, and the second node does not belong to the first node set; update the first array and the second array corresponding to the shortest path from the source node to the second node based on the second node.

[0196] In an optional embodiment of the present invention, the virtual network topology graph further includes a first marker of the link that has failed in the fault detection result, and / or a second marker of the node that has failed; wherein the weight of the link associated with the first marker is the maximum value; the first marker and / or the second marker are used to update the first array and the second array.

[0197] In an optional embodiment of the present invention, the multi-path redundancy module 42 is configured to determine the shortest path between a third node and a fourth node based on nodes not associated with the second marker, nodes corresponding to links not associated with the first marker, and links not associated with the first marker in the virtual network topology graph; wherein the third node is a predecessor node of a node associated with the second marker, and the fourth node is the last node in a link associated with the first marker; based on the shortest path between the third node and the fourth node, the corresponding first array and second array are updated, and the first marker and / or the second marker are deleted.

[0198] In an optional embodiment of the present invention, the virtual verification module 43 is configured to obtain a first performance index value of the first link at a first moment within the basic time period; obtain a second performance index value of the first time period and a third performance index value of the second time period based on the first performance index value; determine whether the first link has failed within the basic time period based on the second performance index value and the third performance index value, and obtain a fault monitoring result.

[0199] In an optional embodiment of the present invention, the virtual verification module 43 is configured to determine that the first link has failed, and / or determine that the predecessor node in the first link has failed, when the third performance index value is greater than or equal to the first threshold and the second performance index value is less than the first threshold.

[0200] In one optional embodiment of the present invention, the performance indicators include one or more of the following: latency, packet loss rate, and bandwidth utilization.

[0201] In an optional embodiment of the present invention, the multi-path redundancy module 42 is used to determine, when there is no marker in the virtual network topology graph, to obtain a virtual routing flow table based on the first array and the second array corresponding to the shortest paths from the source node to each node after the update.

[0202] In an optional embodiment of the present invention, the device further includes a communication processing module 44, which is used to connect the basic environment module 41, the multipath redundancy module 42 and the virtual verification module 43; and is also used to initialize communication configuration and build a communication protocol service, transmit the virtual network topology map generated by the basic environment module 41 to the multipath redundancy module 42 through the protocol, and transmit the virtual routing flow table to the virtual verification module 43.

[0203] For example, the communication processing module 44 can encapsulate the algorithm into a service interface, such as using a RESTful (Representational State Transfer) application programming interface (API), a message queue, or a full-duplex communication interface (such as WebSockets).

[0204] In this invention, the basic environment module 41, multipath redundancy module 42, and virtual verification module 43 in the device can be implemented by a central processing unit (CPU), digital signal processor (DSP), microcontroller unit (MCU), or field-programmable gate array (FPGA) in practical applications; the communication processing module 44 in the device can be implemented by a communication module (including: basic communication kit, operating system, communication module, standardized interface and protocol, etc.) and transceiver antenna in practical applications.

[0205] It should be noted that the network resource optimization device described above is only illustrated by the division of the above-described program modules. In practical applications, the above processing can be assigned to different program modules as needed, that is, the internal structure of the device can be divided into different program modules to complete all or part of the processing described above. Furthermore, the network resource optimization device and the network resource optimization method provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiments, which will not be repeated here.

[0206] This invention also provides a network resource optimization system, which is deployed on the digital twin network side of the network resource optimization system architecture, and... Figure 2 The platform architecture shown corresponds to the digital twin network side.

[0207] Figure 7 This is a schematic diagram of the composition structure of the network resource optimization system according to an embodiment of the present invention; as shown below. Figure 7 As shown, the system includes: a physical layer 51, a connection layer 52, a twin data layer 53, and an application service layer 54; wherein,

[0208] The physical layer 51 is used to provide physical network information, which includes network devices in the physical entity network, the connection relationships between network devices, and the performance indicators corresponding to the connections between network devices.

[0209] The connection layer 52 is used to transmit the physical network information provided by the physical layer 51 to the twin data layer 53;

[0210] The twin data layer 53 is used to generate a virtual network topology map based on the physical network information, and to determine the shortest path information from the source node to each node in the virtual network topology map. The virtual network topology map includes nodes, links between nodes, and weights corresponding to the links. Each node corresponds to a network device in the physical network information. Each link corresponds to a connection relationship between network devices. The weights represent the link's performance parameters. The source node is any node in the virtual network topology map. It is also used to determine a first time period and a second time period included in the basic time period based on monitoring requirements and / or network traffic characteristics. Within the basic time period, the performance indicators of the first link are monitored. Based on the monitored performance indicators, the first time period, and the second time period, fault detection is performed on the first link to obtain fault detection results. The first link is a link formed by two nodes included in the shortest path information. A virtual routing flow table is determined based on the fault detection results, and the virtual routing flow table is mapped to physical network resource optimization information.

[0211] The application service layer 54 is used to send monitoring requests to the twin data layer 53; it is also used to obtain physical network resource optimization information from the twin data layer 53, and optimize the configuration of network devices and / or connections between network devices according to the physical network resource optimization information to achieve service deployment.

[0212] In this embodiment of the invention, a four-layer architecture is designed. The physical layer provides the physical network element network connection and traffic data foundation; the connection layer acts as a bridge for the transmission of physical and virtual information; the twin data layer contains key sub-models such as data mapping, data warehouse, and service mapping model for data processing and modeling; and the application service layer realizes business requirement input, model-based deployment, and control updates of the physical layer. Compared with existing technologies, this four-layer architecture is more refined and has closer collaboration. The twin data layer is further subdivided into data mapping, data warehouse, and service mapping model subsystems to achieve precise data processing, improve the depth and breadth of network resource optimization, extend optimization from local to global, improve network performance stability and resource utilization, adapt to the complex and ever-changing needs of cloud networks, and allow various tests and optimization decisions to be made without affecting the actual network operation, thereby improving the reliability and security of cloud networks.

[0213] As an example, the twin data layer 53 is based on Figure 6 The implementation of each module in the network resource optimization device shown is specifically achieved by calling the encapsulated modules through API interfaces to realize the various functions in the twin data layer 53. Figure 8 This is a schematic diagram of the structural operation flow of the network resource optimization system according to an embodiment of the present invention; as follows: Figure 8 As shown, the structural operation flow of network resource optimization based on the modules in the network resource optimization device is as follows:

[0214] The basic environment module 41 corresponds to the data mapping and data warehouse in the twin data layer 53. It is used to receive physical network information sent by the physical layer 51 through the connection layer 52, and to model the network element link scenarios that require twinning. Specifically, it includes mapping twin network element information of network devices in the physical network information, mapping twin link information of the connection relationship between network elements in the physical network information, and storing the traffic parameter information related to the physical network.

[0215] The multi-path redundancy module 42 implements some functions of the service mapping model in the twin data layer 53. Specifically, it loads and runs the algorithm program. The input of this module is determined by the communication processing module 44. The basic sub-module in the service mapping model receives the content input through the communication processing module 44. The input content can be parameter information related to business requirements entered by the user in the custom interface, and / or information such as routing initial state parameters, destination address, and traffic information stored in the basic environment module 41. The functional sub-module in the service mapping model performs positive data processing on the input content through algorithm iteration. After processing, it outputs the initial virtual routing flow table to the virtual verification module 43. In addition, it updates the initial virtual routing flow table based on the fault detection results fed back by the virtual verification module 43 and transmits the updated virtual routing flow table to the virtual verification module 43.

[0216] The virtual verification module 43 is used to implement some functions of the service mapping model in the twin data layer 53. Specifically, it receives the initial virtual routing flow table from the multi-path redundancy module 42 through the communication protocol built by the communication processing module 44, and verifies the initial virtual routing flow table in combination with the fault detection feedback mechanism of scalable time period. After detecting the fault path and / or fault node, it feeds back the fault detection result to the multi-path redundancy module 42 to obtain the updated virtual routing flow table. After obtaining the verification success information, it selects the relevant data parameters, configures the data parameters to map to physical network elements and data information, forwards them to the cloud server platform side through routing, and obtains the corresponding decision feedback to realize early optimization measures in scenarios such as network resource optimization, time period traffic prediction, and fault detection and early warning.

[0217] The communication processing module 44 is used to initialize the communication configuration, build the communication protocol service, and connect the data warehouse, data mapping and service mapping model in the twin data layer 53 to realize the information interaction between the above functional modules. It obtains the virtual routing flow table output by the multi-path redundancy module 42 by receiving message notifications and transmits the virtual routing flow table to the virtual verification module 43 by sending message notifications.

[0218] In this embodiment of the invention, by designing the aforementioned multiple modules, a basic environment module constructs a network element link scenario model; a multi-path redundancy module runs an algorithm and, combined with a fault detection feedback mechanism, determines the virtual routing flow table after network resource optimization; a communication processing module handles communication configuration and information transmission; and a virtual verification module performs network element link path planning and feedback in the virtual mapping environment. Based on the verification results, network parameters are intelligently adjusted to improve network adaptability, enabling the network to operate stably and efficiently under complex conditions such as different loads and faults, thus enhancing the network's adaptability to complex and ever-changing cloud network environments. Through the collaborative work between these modules, the network resource optimization process is automated and intelligent, systematically ensuring the interaction of module functions, and allowing for flexible expansion of module functions to adapt to various network scenarios.

[0219] This invention also provides a computer device. Figure 9 This is a schematic diagram of the hardware composition structure of a computer device according to an embodiment of the present invention; as shown below. Figure 9 As shown, the computer device includes a memory 62, a processor 61, and a computer program stored on the memory 62 and executable on the processor 61.

[0220] Optionally, when the processor 61 executes the program, it implements the steps of the network resource optimization method of the present invention.

[0221] Optionally, the computer device also includes at least one communication component 64. The various components in the computer device can be coupled together via a bus system 63. It is understood that the bus system 63 is used to implement communication between these components. In addition to a data bus, the bus system 63 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in… Figure 9 The general labeled all buses as Bus System 63.

[0222] It is understood that memory 62 can be volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), ferromagnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM); magnetic surface memory can be disk storage or magnetic tape storage. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), SyncLink Dynamic Random Access Memory (SLDRAM), and Direct Rambus Random Access Memory (DRRAM).The memory 62 described in this embodiment of the invention is intended to include, but is not limited to, these and any other suitable types of memory.

[0223] The methods disclosed in the above embodiments of the present invention can be applied to processor 61, or implemented by processor 61. Processor 61 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware in processor 61 or by instructions in the form of software. The processor 61 may be a general-purpose processor, DSP, or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Processor 61 can implement or execute the methods, steps and logic block diagrams disclosed in the embodiments of the present invention. The general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the methods disclosed in the embodiments of the present invention can be directly manifested as being executed by a hardware decoding processor, or being executed by a combination of hardware and software modules in the decoding processor. The software modules may be located in a storage medium, which is located in memory 62. Processor 61 reads the information in memory 62 and completes the steps of the aforementioned method in combination with its hardware.

[0224] In an exemplary embodiment, the computer device may be implemented by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), FPGAs, general-purpose processors, controllers, MCUs, microprocessors, or other electronic components to perform the aforementioned methods.

[0225] This invention also provides a computer-readable storage medium having a computer program stored thereon.

[0226] Optionally, the computer-readable storage medium can be applied to the network resource optimization apparatus of the present invention; then, when the program is executed by the processor, it implements the steps of the network resource optimization method of the present invention.

[0227] This invention also provides a computer program product, including a computer program that can be executed by a processor 61 of a computer device to perform the steps of the network resource optimization method described in this invention.

[0228] The methods disclosed in the several method embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments.

[0229] The features disclosed in the several product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.

[0230] The features disclosed in the several method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method or device embodiments.

[0231] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0232] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0233] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0234] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0235] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

[0236] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for optimizing network resources, characterized in that, The method includes: Obtain physical network information and generate a virtual network topology map based on the physical network information; wherein, the virtual network topology map includes nodes, links between nodes, and weights corresponding to the links; the nodes correspond to network devices in the physical network information; the links correspond to the connection relationships between network devices; and the weights represent the index parameters of the links; Determine the shortest path information from the source node to each node in the virtual network topology graph; the source node is any node in the virtual network topology graph; The first and second time periods included in the basic time period are determined based on monitoring needs and / or network traffic characteristics. The performance indicators of the first link are monitored during the basic time period. Based on the monitored performance indicators, the first time period, and the second time period, fault detection is performed on the first link to obtain fault detection results. The first link is a link formed by two nodes included in the shortest path information. The virtual routing flow table is determined based on the fault detection results. The virtual routing flow table is used to optimize the configuration of network devices and / or the connections between network devices. The virtual network topology graph also includes a first marker of the faulty link in the fault detection result, and / or a second marker of the faulty node; wherein the weight of the link associated with the first marker is the maximum value; the first marker and / or the second marker are used to update the first array and the second array corresponding to each node; the first array includes the shortest distance corresponding to the shortest path from the source node to each node, and the second array includes the predecessor node of each node on the shortest path; the shortest path satisfies the minimum link latency and / or minimum bandwidth consumption, and the shortest distance represents the corresponding latency and / or bandwidth consumption of the shortest path.

2. The method according to claim 1, characterized in that, The physical network information includes network devices in the physical network, the connection relationships between network devices, and the performance indicators corresponding to the connections between network devices; generating a virtual network topology map based on the physical network information includes: The network devices in the physical network information are mapped as nodes, and the connection relationships between the network devices in the physical network information are mapped as links between nodes to generate an initial virtual network topology. Weights are determined based on the performance metrics corresponding to the connections between network devices, and the weights are mapped to the links corresponding to the connections between the network devices to generate the virtual network topology.

3. The method according to claim 1, characterized in that, Determining the shortest path information from the source node to each node in the virtual network topology graph includes: Based on the virtual network topology, the first array and the second array corresponding to the source node and each node are obtained, and the shortest path information between the source node and each node is determined based on the first array and the second array.

4. The method according to claim 3, characterized in that, The process of obtaining the source node and the first and second arrays corresponding to each node includes: Initialize the first array, the second array, and the first node set, where the first node set includes the visited nodes; Select a first node from the second node set and add the first node to the first node set; wherein the second node set includes unvisited nodes and the first node is the node closest to the source node; Determine a second node among the neighboring nodes of the first node, and take the first node as the predecessor node of the second node; wherein the second node satisfies the following conditions: the sum of the shortest distance between the first node and the source node and the weights between the first node and the second node is less than the shortest distance between the second node and the source node, and the second node does not belong to the set of the first nodes; Update the first and second arrays corresponding to the shortest path from the source node to the second node based on the second node.

5. The method according to claim 1, characterized in that, The first and / or the second marker are used to update the first and second arrays, including: Based on the nodes not associated with the second marker, the nodes corresponding to the links not associated with the first marker, and the links not associated with the first marker in the virtual network topology graph, determine the shortest path between the third node and the fourth node; wherein, the third node is the predecessor node of the node associated with the second marker, and the fourth node is the last node in the link associated with the first marker. Based on the shortest path between the third node and the fourth node, update the corresponding first array and second array, and delete the first tag and / or the second tag.

6. The method according to any one of claims 1 to 5, characterized in that, The monitoring of performance indicators of the first link during the base time period, and the fault detection of the first link based on the monitored performance indicators, to obtain fault detection results, includes: The first performance index value of the first link is obtained at the first moment within the basic time period; Based on the first performance index value, the second performance index value for the first time period and the third performance index value for the second time period are obtained respectively. Based on the second performance index value and the third performance index value, it is determined whether the first link has failed within the basic time period, and the fault monitoring result is obtained.

7. The method according to claim 6, characterized in that, The step of determining whether the first link has failed within the base time period based on the second performance index value and the third performance index value includes: If the third performance index value is greater than or equal to the first threshold and the second performance index value is less than the first threshold, it is determined that the first link has failed, and / or, it is determined that the predecessor node in the first link has failed.

8. The method according to claim 1, characterized in that, The performance metrics include one or more of the following: latency, packet loss rate, and bandwidth utilization.

9. The method according to claim 5, characterized in that, The step of determining the virtual routing flow table based on the fault detection result includes: If no marker is found in the virtual network topology graph, the virtual routing flow table is obtained based on the first and second arrays corresponding to the shortest paths from the source node to each node after the update.

10. A network resource optimization device, characterized in that, The device includes: a basic environment module, a multi-path redundancy module, and a virtual verification module; wherein... The basic environment module is used to acquire physical network information and generate a virtual network topology map based on the physical network information; wherein, the virtual network topology map includes nodes, links between nodes, and weights corresponding to the links; the nodes correspond to network devices in the physical network information; the links correspond to the connection relationships between network devices; and the weights represent the index parameters of the links; The multi-path redundancy module is used to determine the shortest path information from the source node to each node in the virtual network topology graph; the source node is any node in the virtual network topology graph; it is also used to determine a virtual routing flow table based on the fault detection results, the virtual routing flow table being used to optimize the configuration of network devices and / or the connections between network devices; The virtual verification module is used to determine the first time period and the second time period included in the basic time period according to monitoring requirements and / or network traffic characteristics; to monitor the performance indicators of the first link within the basic time period; to perform fault detection on the first link based on the monitored performance indicators, the first time period and the second time period, and to obtain fault detection results; the first link is the link formed by the two nodes included in the shortest path information. The virtual network topology graph also includes a first marker of the faulty link in the fault detection result, and / or a second marker of the faulty node; wherein the weight of the link associated with the first marker is the maximum value; the first marker and / or the second marker are used to update the first array and the second array corresponding to each node; the first array includes the shortest distance corresponding to the shortest path from the source node to each node, and the second array includes the predecessor node of each node on the shortest path; the shortest path satisfies the minimum link latency and / or minimum bandwidth consumption, and the shortest distance represents the corresponding latency and / or bandwidth consumption of the shortest path.

11. A network resource optimization system, characterized in that, The system comprises: a physical layer, a connectivity layer, a twin data layer, and an application service layer; wherein... The physical layer is used to provide physical network information, which includes network devices in the physical entity network, the connection relationships between network devices, and the performance indicators corresponding to the connections between network devices. The connection layer is used to transmit the physical network information provided by the physical layer to the twin data layer; The twin data layer is used to generate a virtual network topology map based on the physical network information, and to determine the shortest path information from the source node to each node in the virtual network topology map; wherein, the virtual network topology map includes nodes, links between nodes, and weights corresponding to the links; the nodes correspond to network devices in the physical network information; the links correspond to the connection relationships between network devices; the weights represent the index parameters of the links; the source node is any node in the virtual network topology map; the virtual network topology map also includes a first marker of the link that has failed in the fault detection result, and / or a second marker of the node that has failed; wherein, the weight of the link associated with the first marker is the maximum value; the first marker and / or the second marker are used to update the first array and the second array corresponding to the source node and each node; the first The first array includes the shortest distances corresponding to the shortest paths from the source node to each node, and the second array includes the predecessor nodes of each node on the shortest path; the shortest path satisfies the minimum latency and / or minimum bandwidth consumption of the link, and the shortest distance represents the corresponding latency and / or bandwidth consumption of the shortest path; it is also used to determine a first time period and a second time period included in the basic time period according to monitoring requirements and / or network traffic characteristics; to monitor the performance indicators of the first link within the basic time period, and to perform fault detection on the first link based on the monitored performance indicators, the first time period, and the second time period to obtain fault detection results; the first link is a link formed by two nodes included in the shortest path information; a virtual routing flow table is determined according to the fault detection results, and the virtual routing flow table is mapped to physical network resource optimization information; The application service layer is used to send monitoring requests to the twin data layer; it is also used to obtain physical network resource optimization information from the twin data layer, and optimize the configuration of network devices and / or connections between network devices according to the physical network resource optimization information to achieve service deployment.

12. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the method according to any one of claims 1 to 9.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1 to 9.

14. A computer program product, comprising a computer program, characterized in that, The computer program, when executed by a processor, implements the method according to any one of claims 1 to 9.

Images