Path selection method, device, system and storage medium

Abstract

The present disclosure proposes a path selection method, device, system and storage medium, which relate to the field of network technology. A path selection method disclosed in the present disclosure includes: determining a candidate path based on a digital twin model of a target network, wherein the digital twin model is generated according to the parameters of the target network, and the digital twin model dynamically interacts with the target network; determining the target path according to at least one of the latency, packet loss rate or bandwidth load of the candidate path; and sending the target path to the target network so that the target network uses the target path as an alternative path. Through such a method, the trial and error cost of selecting an alternative path can be reduced, and the reliability of the determined alternative path can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of network technology, and in particular to a path selection method, device, system and storage medium. Background Art

[0002] In a three-layer network, service flows usually select round-trip paths during operation. If the circuit quality is poor or interrupted, the line will be reselected. If the end-to-end service quality is poor, the service load needs to be transferred to other optional paths. For multiple optional paths, it is necessary to determine the optimal path in advance or load-share multiple paths to load the service.

[0003] In the related art, the path is generally selected based on the minimum routing hops or the shortest path. In addition, there is a method of selecting the optimal alternative path by superimposing the costs of alternative links, and a method of selecting a path for an SDN (Software Defined Network) network based on load balancing and QoE (Quality of Experience) metrics. Summary of the invention

[0004] One object of the present disclosure is to improve the accuracy and efficiency of path selection.

[0005] According to one aspect of some embodiments of the present disclosure, a path selection method is proposed, including: determining a candidate path based on a digital twin model of a target network, wherein the digital twin model is generated according to parameters of the target network, and the digital twin model dynamically interacts with the target network; determining a target path according to at least one of the latency, packet loss rate, or bandwidth load of the candidate path; and sending the target path to the target network so that the target network uses the target path as an alternative path.

[0006] In some embodiments, the method also includes: generating a digital twin model of the target network based on the parameters of the target network, wherein the parameters of the target network include target end-to-end business performance data, network topology data, network configuration data, and device resource data; synchronizing the business data of the target network into the digital twin model.

[0007] In some embodiments, based on the digital twin model of the target network, determining candidate paths includes: determining poor quality paths based on at least one of the business performance of the digital twin model or end-to-end SLA (Service Level Agreement) performance indicators; excluding poor quality paths to determine candidate paths.

[0008] In some embodiments, based on the digital twin model of the target network, determining the candidate paths also includes: collecting traffic data and device data of the digital twin model; performing predetermined cleaning processing on the traffic data and device data of the digital twin model to obtain cleaned data; determining the business performance of the digital twin model based on the cleaned data, and the business performance includes at least one of packet loss rate, one-way delay, two-way delay, or bandwidth load.

[0009] In some embodiments, determining a poor quality path based on the business performance of the digital twin model or at least one of the end-to-end SLA performance indicators includes: determining a congested path based on the business performance of the digital twin model and an anomaly detection model, and treating the congested path as a poor quality path.

[0010] In some embodiments, determining a poor quality path based on the business performance of the digital twin model or at least one of the end-to-end SLA performance indicators includes: determining an SLA excess path based on the business performance and SLA performance indicators of the digital twin model, and using the SLA excess path as a poor quality path, wherein the SLA performance indicators include one or more of maximum packet loss rate, maximum single-way delay, maximum two-way delay, or maximum bandwidth load.

[0011] In some embodiments, determining the poor quality path based on at least one of the business performance of the digital twin model or the end-to-end SLA performance indicators includes: determining the congested path based on the business performance and anomaly detection model of the digital twin model; determining the SLA over-limit path based on the business performance and SLA performance indicators of the digital twin model; and obtaining the intersection of the congested path and the SLA over-limit path to determine the poor quality path.

[0012] In some embodiments, determining the SLA over-limit path based on the business performance and service level agreement SLA performance indicators of the digital twin model includes: when any business performance does not meet the corresponding SLA performance indicator, determining that the path corresponding to the business performance belongs to the SLA over-limit path.

[0013] In some embodiments, determining the target path based on at least one of the latency, packet loss rate or bandwidth load of the candidate paths includes: when the number of candidate paths is greater than 1, transmitting the service flow on each path in the candidate paths respectively, and obtaining at least two of the latency, packet loss rate or bandwidth load corresponding to each path in the candidate paths; obtaining a weighted evaluation value of the corresponding path based on at least two of the latency, packet loss rate or bandwidth load; and determining the target path based on the weighted evaluation value of each path in the candidate paths.

[0014] In some embodiments, obtaining a weighted evaluation value of a corresponding path according to at least two of delay, packet loss rate or bandwidth load includes: normalizing the delay, packet loss rate and bandwidth load to obtain normalized delay parameters, normalized packet loss rate parameters and normalized bandwidth load parameters; obtaining a weighted evaluation value according to the normalized delay parameters, normalized packet loss rate parameters, normalized bandwidth load parameters, and corresponding weights, wherein at least one of the weights corresponding to the normalized delay parameters, normalized packet loss rate parameters and normalized bandwidth load parameters is not 0; determining a target path according to the weighted evaluation values ​​of each path in the candidate paths includes: determining a path with a minimum weighted evaluation value as the target path.

[0015] In some embodiments, the delay is a two-way delay and the target path is a round-trip path; or the delay is a one-way delay and the target path is a one-way path.

[0016] In some embodiments, determining the target path according to the delay and packet loss rate of the candidate paths further includes: when the number of candidate paths is 1, using the candidate path as the target path.

[0017] According to one aspect of some embodiments of the present disclosure, a path selection device is proposed, including: a candidate path determination unit, configured to determine a candidate path based on a digital twin model of a target network, wherein the digital twin model is generated according to parameters of the target network, and the digital twin model dynamically interacts with the target network; a target path determination unit, configured to determine a target path based on at least one of the latency, packet loss rate or bandwidth load of the candidate path; and a sending unit, configured to send the target path to the target network so that the target network uses the target path as an alternative path.

[0018] In some embodiments, the device also includes a model building unit configured to: generate a digital twin model of the target network based on the parameters of the target network, wherein the parameters of the target network include target end-to-end business performance data, network topology data, network configuration data, and equipment resource data; and synchronize the business data of the target network into the digital twin model.

[0019] According to one aspect of some embodiments of the present disclosure, a path selection device is provided, comprising: a memory; and a processor coupled to the memory, wherein the processor is configured to execute any one of the above path selection methods based on instructions stored in the memory.

[0020] According to one aspect of some embodiments of the present disclosure, a computer-readable storage medium is provided, on which computer program instructions are stored, and when the instructions are executed by a processor, the steps of any one of the above path selection methods are implemented.

[0021] According to one aspect of some embodiments of the present disclosure, a path selection system is proposed, comprising: any one of the path selection devices described above; and a target network, configured to use a target path determined by the path selection device as an alternative path, and to switch end-to-end services to the alternative path when the path is switched. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The drawings described herein are used to provide a further understanding of the present disclosure and constitute a part of the present disclosure. The illustrative embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation on the present disclosure. In the drawings:

[0023] Figure 1 The following is a flow chart of some embodiments of the path selection method of the present disclosure.

[0024] Figure 2 Flow charts of other embodiments of the path selection method disclosed in the present invention.

[0025] Figure 3 Schematic diagrams of some embodiments of the path selection device disclosed in the present invention.

[0026] Figure 4 Schematic diagrams of other embodiments of the path selection device disclosed in the present invention.

[0027] Figure 5 Schematic diagrams of some further embodiments of the path selection device disclosed in the present invention.

[0028] Figure 6 Schematic diagrams of some embodiments of the path selection system disclosed herein.

[0029] Figure 7 Schematic diagrams of other embodiments of the path selection system disclosed in the present invention. DETAILED DESCRIPTION

[0030] The technical solution of the present disclosure is further described in detail below through the accompanying drawings and embodiments.

[0031] The inventors found that in the related art, due to the lack of comprehensive performance characteristic data for optimal analysis and selection, the shortest path may be overloaded, which is prone to network failure. When trying different solutions in the existing network, it may affect the normal operation of the business, fail to guarantee user perception, and the trial and error cost is high.

[0032] In response to the above problems, the present disclosure proposes a path selection method, device and storage medium, which can perform path strategy analysis in advance and recommend the optimal path without affecting the existing network services, thereby reducing the trial and error cost of the existing network.

[0033] The flowcharts of some embodiments of the path selection method disclosed in the present invention are as follows: Figure 1 As shown, steps 130 to 150 are included.

[0034] In step 130, candidate paths are determined based on the digital twin model of the target network. The digital twin model is generated based on the parameters of the target network, and the digital twin model dynamically interacts with the target network. When the state of the target network, the services carried, etc. change, the changes will be synchronized to the digital twin model in real time. The target network is the actual network that carries the services.

[0035] In some embodiments, a digital twin model of the target network may be generated in advance and updated in real time to ensure synchronization between the digital twin model and the target network.

[0036] In some embodiments, the operation of determining candidate paths in the digital twin model may include: first determining poor quality paths based on the business performance of the digital twin model, or at least one of the end-to-end SLA performance indicators, and further excluding poor quality paths from feasible end-to-end paths, and using the remaining paths as candidate paths. In some embodiments, the SLA performance indicators include one or more of the maximum packet loss rate, the maximum single-item delay, the maximum bilateral delay, or the maximum bandwidth load. Poor quality refers to the low quality of business services, data transmission, or network environment and the threshold. Poor quality paths refer to paths with poor transmission quality that can no longer meet the needs of transmission or providing services to users. Through such a method, it is possible to comprehensively consider various parameters that affect network performance, and use the elimination method to exclude paths that do not meet performance standards, thereby ensuring that the selected candidate paths can support business carrying and improving the reliability of the determined candidate paths.

[0037] In some embodiments, before determining the poor quality path based on at least one of the business performance or end-to-end SLA performance indicators of the digital twin model, the data of the digital twin model may be extracted and processed first, for example, the flow data and device data of the digital twin model may be extracted first; further, a predetermined cleaning process may be performed on the flow data and device data of the digital twin model to obtain the cleaning data, and then the business performance of the digital twin model may be determined based on the cleaning data. In some embodiments, the business performance may include at least one of the packet loss rate, one-way delay, two-way delay, or bandwidth load. By using such a method, the robustness and accuracy of subsequent data processing can be improved.

[0038] In step 140, the target path is determined based on at least one of the delay, packet loss rate, or bandwidth load of the candidate path. In some embodiments, factors affecting the end-to-end transmission quality, such as the delay, packet loss rate, and bandwidth load of the candidate path, can be used as reference items to obtain evaluation values ​​of each item, and then combined with the weight of each reference item to obtain a comprehensive evaluation value of each candidate path. The candidate path with the best performance is selected as the target path according to the comprehensive evaluation value.

[0039] In step 150, the target path is sent to the target network so that the target network uses the target path as an alternative path. In some embodiments, the target path can be sent to a control device of the target network, or to a related forwarding node on the path, and when the current corresponding end-to-end service of the target network has poor quality or interruption, it is switched to the alternative path in time. In some embodiments, the path number can be pre-set, and the number of the target path can be sent to the target network so that the target network can quickly identify the target path.

[0040] Digital twin technology is driven by virtual networks and models, and realizes actual functional services and application requirements such as monitoring, simulation, prediction, and optimization through virtual-real interaction. Based on the method in the embodiment shown above, digital twin technology can be used to evaluate the end-to-end service quality and recommend paths in a virtual network environment without affecting the normal business operation in the actual network, reducing the cost of trial and error; comprehensive end-to-end service performance data and SLA indicator data are used to evaluate and verify the end-to-end service quality of multiple factors, comprehensively evaluate the quality of each path, improve the reliability of the determined alternative path, and improve the end-to-end service performance after switching to the alternative path.

[0041] In some embodiments, Figure 1 As shown, the path selection method of the present disclosure may also include a process of constructing a digital twin model, specifically, including step 110 and step 120.

[0042] In step 110, a digital twin model of the target network is generated according to the parameters of the target network. In some embodiments, the parameters of the target network include target end-to-end service performance data, network topology data, network configuration data, and device resource data.

[0043] In step 120, the service data of the target network is synchronized to the digital twin model. In some embodiments, the service data may include the source node, target node, current transmission path, service flow parameters, etc. In some embodiments, when the service data of the target network changes, it is synchronized to the digital twin model in real time.

[0044] Based on the method in this embodiment, a digital twin model synchronized with the target network can be generated, and the digital twin model is ensured to be synchronized with the target network in terms of model structure, performance, and current business and network status parameters, so that it can be run in a timely manner when an alternative path needs to be determined later, thereby improving the accuracy and timeliness of the target path determination.

[0045] In some embodiments, in the above step 130, a congested path can be determined based on the business performance and anomaly detection model of the digital twin model, and the congested path can be used as a poor quality path. In some embodiments, a congested link can be determined based on any congestion detection method in the relevant technology, such as a link anomaly detection model. For example, the path traffic is input into the anomaly detection model to obtain a congestion output result, wherein the output for the congested path is 1 and the output for the uncongested path is 0. This value is only an example and does not constitute an improper limitation on the present disclosure. Through such a method, it is possible to avoid the existence of congested paths in the candidate paths, improve the business reliability when switching to the alternative paths, and reduce the frequency of path switching.

[0046] In some embodiments, in the above step 130, the SLA over-limit path can be determined based on the business performance and SLA performance indicators of the digital twin model, and the SLA over-limit path can be used as a poor quality path. In some embodiments, the business performance parameters can be compared with the corresponding preset threshold values ​​in the SLA performance indicators. When any business performance does not meet the corresponding SLA performance indicators, it is determined that the path corresponding to the business performance belongs to the SLA over-limit path. For example, the path packet loss rate, one-way delay, two-way delay, and bandwidth load are compared with the corresponding performance indicators in the SLA performance indicators. If one performance indicator exceeds the corresponding SLA value, it is determined to be of poor quality, such as: the path bandwidth load is 60%, the maximum bandwidth load in the SLA is 50%, 60%>50%, it is determined to be over-limit, and the end-to-end service quality is poor.

[0047] This method can avoid the existence of SLA-exceeding paths in the candidate paths, such as paths whose parameters such as packet loss rate, single-way delay, double-way delay, and bandwidth load exceed the preset thresholds, thereby improving the service reliability when switching to the alternative paths and reducing the frequency of path switching.

[0048] In some embodiments, in the above step 130, the congested path can be determined based on the business performance and anomaly detection model of the digital twin model; the SLA overrun path can be determined based on the business performance and SLA performance indicators of the digital twin model. Furthermore, the intersection of the congested path and the SLA overrun path is obtained, and the path in the intersection is used as a poor quality path. Through such a method, it is possible to avoid the existence of congested paths in the candidate paths while avoiding the existence of SLA overrun paths in the candidate paths, further improving the business reliability when switching to the alternative path, and further reducing the frequency of path switching.

[0049] In some embodiments, flow charts of other embodiments of the path selection method disclosed herein are as follows: Figure 2 shown.

[0050] In step 231, a candidate path is determined based on the digital twin model of the target network. In some embodiments, the candidate path can be determined based on the method in any one of the embodiments in step 130 above. In some embodiments, if there is no candidate path, the current process ends and the path selection fails; if there is a candidate path, step 241 is executed.

[0051] In step 241 , it is determined whether the number of candidate paths is greater than 1. If the number of candidate paths is greater than 1, step 242 is executed; if the number of candidate paths is 1, step 245 is executed.

[0052] In step 242, the service flow is transmitted on each path in the candidate path respectively, and at least two items of the delay, packet loss rate or bandwidth load corresponding to each path in the candidate path are obtained. In some embodiments, if the path to be determined is a unidirectional path, the delay may be a unidirectional delay, and the direction corresponding to the delay is the same as the direction of the path; if the path to be determined is a bidirectional path, the delay may be a bidirectional delay.

[0053] In step 243, a weighted evaluation value of the corresponding path is obtained based on at least two of the delay, packet loss rate, or bandwidth load. In some embodiments, the delay, packet loss rate, and bandwidth load parameters may be normalized to eliminate the units of the parameters. In some embodiments, a normalization algorithm Z-score may be used; further, the parameters are multiplied by preset weights to obtain weighted evaluation values ​​of the corresponding dimensions; the weighted evaluation values ​​of each dimension are added to obtain the weighted evaluation value of the path. For example, the weighted evaluation value Y is determined by the following formula.

[0054] Y=W 1 ×Delay normalization value+W 2 ×Normalized value of packet loss rate

[0055] +W 3 ×Bandwidth load normalized value

[0056] Among them, W 1 , W 2 , W 3 They are the weights of latency, packet loss rate, and bandwidth load respectively.

[0057] In some embodiments, if the parameters of a certain dimension or multiple dimensions are not considered, the weight of the corresponding dimension can be set to 0, such as setting W 1 , W 2 , W 3 One or both of them are set to 0, thereby meeting the user's flexible needs for alternative path selection.

[0058] In step 244, the target path is determined according to the weighted evaluation value of each path in the candidate paths. In some embodiments, if the path quality is better and the weighted evaluation value is larger, the path with the largest weighted evaluation value is selected as the target path; in some embodiments, if the path quality is better and the weighted evaluation value is smaller, the path with the smallest weighted evaluation value is selected as the target path. In some embodiments, taking the weighted evaluation value Y calculated in step 243 as an example, the path with the smallest Y value is selected as the target path.

[0059] In step 245 , when the number of candidate paths is 1, the candidate path is taken as the target path.

[0060] Through this method, when there are multiple candidate paths, multiple factors such as latency, packet loss rate, and bandwidth load can be comprehensively considered, and the path that can provide the best quality can be selected as the target path, thereby further improving the quality of the alternative paths and reducing the path switching frequency of the target network end-to-end service.

[0061] Schematic diagrams of some embodiments of the path selection device 30 of the present disclosure are as follows Figure 3 shown.

[0062] The candidate path determination unit 302 can determine the candidate path based on the digital twin model of the target network. The digital twin model is generated according to the parameters of the target network, and the digital twin model dynamically interacts with the target network. When the state of the target network, the business carried, etc. change, the change will be synchronized to the digital twin model in real time. The target network is the actual network that carries the business. In some embodiments, the candidate path determination unit 302 can perform the operations in any of the embodiments in step 130 above.

[0063] The target path determination unit 303 can determine the target path according to at least one of the delay, packet loss rate or bandwidth load of the candidate path. In some embodiments, the factors affecting the end-to-end transmission quality such as the delay, packet loss rate, bandwidth load of the candidate path can be used as reference items to obtain the evaluation value of each item, and then combined with the weight of each reference item to obtain the comprehensive evaluation value of each candidate path. The candidate path with the best performance is selected as the target path according to the comprehensive evaluation value. In some embodiments, the target path determination unit 303 can perform the operations in any of the embodiments in step 140 and steps 241 to 245 above.

[0064] The sending unit 304 can send the target path to the target network so that the target network uses the target path as an alternative path. In some embodiments, the target path can be sent to a control device of the target network, or to a related forwarding node on the path, and when the current corresponding end-to-end service of the target network has poor quality or interruption, it is switched to the alternative path in time. In some embodiments, the sending unit 304 can perform the operations in any embodiment of step 150 above.

[0065] Such a path selection device can use digital twin technology to evaluate end-to-end service quality and recommend paths in a virtual network environment without affecting normal business operations in the actual network and reducing trial and error costs; it can conduct multi-factor end-to-end service quality evaluation and verification based on comprehensive end-to-end service performance data and SLA indicator data, comprehensively evaluate the quality of each path, improve the reliability of the determined alternative paths, and improve the end-to-end service performance after switching to the alternative paths.

[0066] In some embodiments, Figure 3 As shown, the path selection device 30 also includes a model construction unit 301. The model construction unit 301 can generate a digital twin model of the target network based on the parameters of the target network. In some embodiments, the parameters of the target network include target end-to-end service performance data, network topology data, network configuration data, and device resource data. Further, the model construction unit 301 can synchronize the service data of the target network to the digital twin model. In some embodiments, the service data may include the source node, target node, current transmission path, service flow parameters, etc. of the service. In some embodiments, when the service data of the target network changes, it is synchronized to the digital twin model in real time. In some embodiments, the model construction unit 301 can perform the operations in any of the above steps 110 and 120.

[0067] Such a path selection device can generate a digital twin model synchronized with the target network, and ensure that the digital twin model is synchronized with the target network in terms of model structure, performance, and current business and network status parameters, so that it can be run in a timely manner when an alternative path needs to be determined, thereby improving the accuracy and timeliness of the target path determination.

[0068] A schematic diagram of the structure of an embodiment of the path selection device disclosed in the present invention is as follows Figure 4As shown. The path selection device includes a memory 401 and a processor 402. Among them: the memory 401 can be a disk, a flash memory or any other non-volatile storage medium. The memory is used to store the instructions in the corresponding embodiment of the path selection method above. The processor 402 is coupled to the memory 401 and can be implemented as one or more integrated circuits, such as a microprocessor or a microcontroller. The processor 402 is used to execute the instructions stored in the memory, which can reduce the trial and error cost when selecting the alternative path and improve the reliability of the determined alternative path.

[0069] In one embodiment, it is also possible to Figure 5 As shown, the path selection device 500 includes a memory 501 and a processor 502. The processor 502 is coupled to the memory 501 via a BUS 503. The path selection device 500 can also be connected to an external storage device 505 via a storage interface 504 to call external data, and can also be connected to a network or another computer system (not shown) via a network interface 506. No further detailed description will be given here.

[0070] In this embodiment, by storing data instructions in a memory and then processing the instructions through a processor, the trial and error cost when selecting an alternative path can be reduced and the reliability of the determined alternative path can be improved.

[0071] In another embodiment, a computer-readable storage medium stores thereon computer program instructions, which, when executed by a processor, implement the steps of the method in the corresponding embodiment of the path selection method. It should be understood by those skilled in the art that the embodiments of the present disclosure may be provided as methods, devices, or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present disclosure may take the form of a computer program product implemented on one or more computer-usable non-transient storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0072] Schematic diagrams of some embodiments of the path selection system disclosed in the present invention are as follows Figure 6 shown.

[0073] The path selection device 60 may be any one of the ones mentioned above.

[0074] The target network 61 is a synchronization object of the digital twin model used in the path selection device 60, and can use the target path determined by the path selection device as an alternative path, and switch the end-to-end service to the alternative path when the path is switched.

[0075] Such a path selection system can use digital twin technology to evaluate end-to-end service quality and recommend paths in a virtual network environment without affecting normal business operations in the actual network and reducing trial and error costs. It can also use comprehensive end-to-end service performance data and SLA indicator data to conduct multi-factor end-to-end service quality evaluation and verification, comprehensively evaluate the quality of each path, improve the reliability of the determined alternative paths, and improve the end-to-end service performance after switching to the alternative paths.

[0076] Schematic diagrams of other embodiments of the path selection system disclosed in the present invention are as follows Figure 7 shown.

[0077] The path selection system includes a path selection device 70 and a target network 71 .

[0078] The path selection device 70 can execute any of the path selection methods mentioned above. The modules in the path selection device 70 in the figure are examples of composition or logical modules based on the method execution steps, and do not constitute an improper limitation on the structure of the path selection device 70.

[0079] The target network is the network on which the digital twin model constructed by the path selection device is based. The end-to-end business flow of the target network is transmitted between router 1 connected to the customer device and router 5 connected to the customer business center. There are multiple optional paths between router 1 and router 5, such as the main path, alternative path 1...alternative path n shown in the figure, where n is an integer greater than 1.

[0080] In some embodiments, the network parameters, business parameters, etc. of the target network 71 are collected by the collection center and integrated and processed by the resource center to generate port configuration data, performance data, etc., which are sent to the digital twin model construction module of the path selection device 70.

[0081] The digital twin model construction module builds an end-to-end business digital twin model based on the target end-to-end business performance data, network topology data, network configuration data, device resource data, etc. In addition, the digital twin model construction module obtains the synchronization data of the target end-to-end business path and synchronizes the data into the end-to-end business digital twin model.

[0082] The data storage module has the ability to efficiently store large amounts of data. It can collect traffic data and equipment data from the digital twin model and perform simple data cleaning. The cleaning operation includes unifying the data units of various performance indicators, processing missing values, and processing outliers.

[0083] The data processing module can process the collected traffic data and analyze the packet loss rate characteristics to obtain data, one-way delay, two-way delay, bandwidth load and other characteristics.

[0084] The calculation formulas for one-way delay, two-way delay, packet loss rate and message length are as follows:

[0085] One-way delay = tail node message receiving time – head node message sending time;

[0086] Two-way delay = (time for receiving the message at the tail node – time for sending the message at the head node) + (time for receiving the message at the head node – time for sending the message at the tail node);

[0087] Packet loss rate = |Number of messages received by the tail node – Number of messages sent by the head node|*100% / Number of messages sent by the head node;

[0088] Bandwidth load = traffic / bandwidth.

[0089] The end-to-end service quality evaluation and verification module can integrate end-to-end service performance data and SLA indicator data (such as maximum packet loss rate, maximum one-way delay, maximum two-way delay, and maximum bandwidth load) to evaluate and verify end-to-end service quality, including SLA limit violation and path congestion evaluation.

[0090] The optimal path recommendation module can select the target path as the best alternative path by combining the end-to-end service quality evaluation results, the comprehensive service flow reception and transmission delay, and the packet loss rate.

[0091] above Figure 2 Taking the weighted evaluation value calculation method in the illustrated embodiment as an example, the single-item delay and packet loss rate of the alternative path are considered when determining the weighted evaluation value. Figure 7 The normalized delay, normalized packet loss rate and weighted evaluation value of each candidate path are shown in Table 1 below as an example. The values ​​in this example are only for explaining the selection method and do not constitute an improper limitation on the present disclosure.

[0092] Table 1 Figure 7 Examples of normalized parameters and evaluation values ​​for multiple candidate paths in

[0093]

[0094] The result notification module can notify the actual target network 71 of the optimal candidate path, for example, sending the path number 3 to the controller of the target network or the related device of the candidate path 3 .

[0095] Such a network system can evaluate the end-to-end service quality and recommend the optimal path in a virtual network environment based on digital twin technology, without affecting the normal operation of the business, effectively ensuring user perception and reducing trial and error costs. It can also conduct multi-factor end-to-end service quality evaluation and verification based on comprehensive end-to-end service performance data and SLA indicator data, and comprehensively evaluate the quality of each path, thereby meeting user needs and further improving network operation and maintenance efficiency.

[0096] The present disclosure is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present disclosure. It should be understood that each process and / or block in the flowchart and / or block diagram and the combination of processes and / or blocks in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0097] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0098] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0099] So far, the present disclosure has been described in detail. In order to avoid obscuring the concept of the present disclosure, some details known in the art are not described. Based on the above description, those skilled in the art can fully understand how to implement the technical solution disclosed here.

[0100] The method and apparatus of the present disclosure may be implemented in many ways. For example, the method and apparatus of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above, unless otherwise specifically stated. In addition, in some embodiments, the present disclosure may also be implemented as a program recorded in a recording medium, which includes machine-readable instructions for implementing the method according to the present disclosure. Therefore, the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.

[0101] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present disclosure rather than to limit it. Although the present disclosure has been described in detail with reference to the preferred embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present disclosure can still be modified or some technical features can be replaced by equivalents without departing from the spirit of the technical solution of the present disclosure, which should be included in the scope of the technical solution for protection of the present disclosure.

Claims

1. A path selection method, comprising: Determining candidate paths based on a digital twin model of a target network includes: determining a poor-quality path according to business performance of the digital twin model and end-to-end service level agreement (SLA) performance indicators; excluding the poor-quality path from feasible end-to-end paths to determine the candidate path, wherein the digital twin model is generated according to parameters of the target network, and the digital twin model dynamically interacts with the target network; Determine a target path according to at least one of a delay, a packet loss rate, or a bandwidth load of the candidate path; sending the target path to the target network so that the target network uses the target path as an alternative path, Among them, determining the poor quality path according to the business performance of the digital twin model and the end-to-end SLA performance indicators includes: determining the congested path according to the business performance and anomaly detection model of the digital twin model; determining the SLA over-limit path according to the business performance and SLA performance indicators of the digital twin model, wherein the SLA performance indicators include one or more of the maximum packet loss rate, the maximum single-way delay, the maximum bilateral delay or the maximum bandwidth load; obtaining the intersection of the congested path and the SLA over-limit path to determine the poor quality path.

2. The method according to claim 1, further comprising: Generating the digital twin model of the target network according to the parameters of the target network, wherein the parameters of the target network include target end-to-end service performance data, network topology data, network configuration data, and device resource data; The business data of the target network is synchronized into the digital twin model.

3. The method according to claim 1, wherein: The digital twin model based on the target network, determining the candidate path also includes: Collecting traffic data and equipment data of the digital twin model; Performing a predetermined cleaning process on the flow data and device data of the digital twin model to obtain cleaned data; The business performance of the digital twin model is determined based on the cleansing data, and the business performance includes at least one of packet loss rate, one-way delay, two-way delay or bandwidth load.

4. The method according to claim 1, wherein: Determining the SLA overlimit path according to the business performance and service level agreement SLA performance indicators of the digital twin model includes: In the case where any service performance does not meet the corresponding SLA performance indicator, it is determined that the path corresponding to the service performance belongs to the SLA over-limit path.

5. The method according to any one of claims 1 to 4, wherein: The determining of the target path according to at least one of the delay, packet loss rate or bandwidth load of the candidate path comprises: When the number of the candidate paths is greater than 1, the service flow is transmitted on each path in the candidate paths respectively, and at least two of the delay, packet loss rate or bandwidth load corresponding to each path in the candidate paths are obtained; Obtaining a weighted evaluation value of a corresponding path according to at least two of the delay, the packet loss rate, or the bandwidth load; The target path is determined according to the weighted evaluation value of each path in the candidate paths.

6. The method according to claim 5, wherein: The obtaining of a weighted evaluation value of a corresponding path according to at least two of the delay, the packet loss rate or the bandwidth load includes: Normalizing the delay, the packet loss rate, and the bandwidth load to obtain a normalized delay parameter, a normalized packet loss rate parameter, and a normalized bandwidth load parameter; Acquire the weighted evaluation value according to the normalized delay parameter, the normalized packet loss rate parameter, the normalized bandwidth load parameter, and corresponding weights, wherein at least one of the weights corresponding to the normalized delay parameter, the normalized packet loss rate parameter, and the normalized bandwidth load parameter is not 0; Determining the target path according to the weighted evaluation value of each path in the candidate paths includes: determining the path with the smallest weighted evaluation value as the target path.

7. The method according to claim 5 or 6, wherein: The delay is a two-way delay, and the target path is a round-trip path; or The delay is a one-way delay, and the target path is a one-way path.

8. The method according to claim 5, wherein: The determining of the target path according to the delay and packet loss rate of the candidate path further comprises: When the number of the candidate paths is 1, the candidate path is used as the target path.

9. A path selection device, comprising: The candidate path determination unit is configured to determine the candidate path based on the digital twin model of the target network, including: determining the poor quality path according to the business performance of the digital twin model and the end-to-end service level agreement SLA performance index; excluding the poor quality path from the feasible end-to-end path to determine the candidate path, wherein the digital twin model is generated according to the parameters of the target network, and the digital twin model dynamically interacts with the target network; a target path determination unit, configured to determine the target path according to at least one of the delay, packet loss rate or bandwidth load of the candidate path; a sending unit, configured to send the target path to the target network so that the target network uses the target path as an alternative path, Among them, determining the poor quality path according to the business performance of the digital twin model and the end-to-end SLA performance indicators includes: determining the congested path according to the business performance and anomaly detection model of the digital twin model; determining the SLA over-limit path according to the business performance and SLA performance indicators of the digital twin model, wherein the SLA performance indicators include one or more of the maximum packet loss rate, the maximum single-way delay, the maximum bilateral delay or the maximum bandwidth load; obtaining the intersection of the congested path and the SLA over-limit path to determine the poor quality path.

10. The apparatus according to claim 9, further comprising a model building unit configured to: The digital twin model of the target network is generated according to the parameters of the target network, wherein: The parameters of the target network include target end-to-end service performance data, network topology data, network configuration data and equipment resource data; The business data of the target network is synchronized into the digital twin model.

11. A path selection device, comprising: Memory; as well as A processor coupled to the memory, the processor being configured to execute the method according to any one of claims 1 to 8 based on instructions stored in the memory.

12. A computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, implement the steps of the method according to any one of claims 1 to 8.

13. A path selection system, comprising: The path selection device according to any one of claims 9 to 11; and The target network is configured to use the target path determined by the path selection device as an alternative path, and switch the end-to-end service to the alternative path when the path is switched.

Images