Link protection method, link protection model, server, system and storage medium applied to all-optical intelligent calculation center

Abstract

The invention relates to the technical field of communication link protection, in particular to a link protection method, a link protection model, a server, a system and a storage medium applied to an all-optical intelligent computing center. According to the method, links are divided according to the fault probability predicted by the support vector machine, link protection is carried out by adopting strategies of different priorities based on sub-networks of different fault levels by taking the sub-networks as units, and the fault positioning time is shortened while the positioning precision is improved. The objective of the invention is to solve the problem of how to shorten the fault positioning time and increase the positioning precision.

Description

Technical Field

[0001] This application relates to the field of communication link protection technology, and in particular to a link protection method, link protection model, server, system and storage medium for use in all-optical intelligent computing centers. Background Technology

[0002] With the development of internet technology, nationwide communication traffic is exploding, leading to an expansion of the all-optical intelligent computing center network and an increase in the number of users. Currently, optical path failures on the data communication network (DCN) within an all-optical intelligent computing center network consist of two parts: optical transmission and optical switching. Optical transmission primarily refers to optical link failures, while optical switching primarily refers to optical switch failures. The entire network communication is composed of these optical paths. If an optical path failure occurs, communication will be interrupted, affecting user experience and network stability. Without optical path protection mechanisms, optical path failures can lead to very serious losses. Therefore, maintaining the stability and security of communication optical paths has become a crucial issue in ensuring the security of all-optical DCNs.

[0003] Traditional link protection technologies are primarily designed for static networks. Common link fault location algorithms experience significantly increased recovery and computation times when dealing with ever-growing network sizes, failing to provide timely survivability protection for network links. Therefore, traditional link fault algorithms struggle to meet the real-time requirements of high-speed, high-bandwidth network scenarios such as intelligent computing center networks.

[0004] In view of this, this application proposes a new link protection method, which aims to shorten the fault location time while increasing the location accuracy. Summary of the Invention

[0005] The main purpose of this application is to provide a link protection method for all-optical intelligent computing centers, which aims to solve the problem of how to shorten the fault location time while increasing the location accuracy.

[0006] To achieve the above objectives, this application provides a link protection method for all-optical intelligent computing centers, the method comprising: S10, obtain the failure probability of each link in the network topology predicted by the support vector machine model, wherein the network topology is constructed with the devices of the all-optical intelligent computing center as nodes and the links as edges; S20, based on the numerical range of the failure probability of each link, determine the failure level corresponding to each link, and divide the network topology into multiple sub-networks based on the failure level; S30, link protection is performed on a per-sub-network basis, wherein the protection priority of sub-networks with high fault levels is higher than that of sub-networks with low fault levels.

[0007] Optionally, before step S10, the method further includes: S40, construct a network topology by using the intelligent computing center equipment as nodes and links as edges, collect the link characteristics of each link in the network topology, and construct a total dataset based on the link characteristics and the network topology; S50, Select a portion of the total dataset as a training dataset, and input the training dataset into the support vector machine model so that the support vector machine model can perform fault probability prediction training based on the training dataset; S60 predicts the failure probability of each link in the network topology based on the trained support vector machine model.

[0008] Optionally, the support vector machine model, when predicting the failure probability, includes the following steps: Read the interval and number of failures for each link; Initialize the experienced time to 0. When the experienced time is less than the specified total time, check if the fault interval is less than the total time. If it is, increment the fault count by one. The experienced time is equal to the current experienced time plus the fault interval. Repeat this calculation until the experienced time is greater than or equal to the total time. The fault count is obtained through the above calculation. Finally, divide the fault count by the total time to obtain the fault probability of the link.

[0009] Optionally, S20 includes: When the failure probability is within the first interval, the failure level corresponding to the link is determined to be a high failure level, and the link with the high failure level is divided into the first sub-network. When the failure probability is within the second interval, the failure level corresponding to the link is determined to be a medium failure level, and the link at the medium failure level is divided into a second sub-network. When the failure probability is within the third interval, the failure level corresponding to the link is determined to be a low failure level, and the link in the low failure level is divided into the third sub-network. Wherein, the first interval is greater than the second interval, which is greater than the third interval.

[0010] Optionally, before step S20, the method further includes: Perform topology verification on links of each level to determine whether links of that level can form a new topology; If the links at this level cannot form a new topology, then the link with the lowest failure probability among the links connected to the nodes in the disconnected area is selected as a "supplementary link" to supplement the sub-topology until a closed loop is formed to form a new network topology. Optionally, step S30 includes: When performing link protection on a sub-network basis, the following constraints must be met: In the formula, N represents the set of all nodes in the network, E represents the set of all links in the network, and W represents the capacity of each link. This represents the requirements of node u. This represents the number of the circle set CSj, which is also the number of the positioning circle. Indicates a link in the network ( u , v Traffic on ) This represents the cost of increasing the capacity per unit on the link. A binary variable, representing if the vector u → v In CS j If the circle is on the left, the value is 1; otherwise, it is 0. This is a binary variable, representing if the node u In CS j If the circle is on the left, the value is 1; otherwise, it is 0. This is a binary variable, representing whether the link can be accessed by CS. j If the circle is protected, the value is 1; otherwise, it is 0. This is a binary variable, representing a value of 1 if node u is the root node of CSj, and 0 otherwise. A non-negative decimal variable, indicating that in constructing CS j During the middle circle process, nodes u voltage value, For a predefined nonnegative decimal constant, it must satisfy... ; This is a binary variable, representing a value of 1 if optical path p is used as the working path for node u on working capacity w, and 0 otherwise.

[0011] Furthermore, to achieve the above objectives, this application also provides a link protection model, which includes: The fault probability prediction module is used to obtain the fault probability of each link in the network topology predicted by the support vector machine model, wherein the network topology is constructed with the devices of the all-optical intelligent computing center as nodes and the links as edges. The network partitioning module is used to determine the fault level corresponding to each link based on the numerical range of the fault probability of each link, and to divide the network topology into multiple sub-networks based on the fault level. The link protection module is used to perform link protection on a per-sub-network basis, wherein the protection priority of sub-networks with high fault levels is higher than that of sub-networks with low fault levels.

[0012] In addition, to achieve the above objectives, this application also provides a server employing the link protection method for all-optical intelligent computing centers as described in any of the preceding claims.

[0013] In addition, to achieve the above objectives, this application also provides a computer system, the computer system comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, it implements the steps of the link protection method for all-optical intelligent computing centers as described in any of the preceding claims.

[0014] In addition, to achieve the above objectives, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the link protection method for all-optical intelligent computing centers as described in any of the preceding claims.

[0015] This application has at least the following beneficial effects: (1) The topology reconstruction was proposed, which divided the original large network topology into three levels of network topology according to the level, thus solving the problem of the difficulty of calculating large network topology; (2) After SVM prediction, a high-precision p-cycle algorithm is used for link maintenance; this achieves 100% link protection while minimizing computation time and improving performance. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the link protection method applied to an all-optical intelligent computing center according to an embodiment of this application; Figure 2 This is a schematic diagram of the reconstructed topology involved in the embodiments of this application; Figure 3 This is a schematic diagram illustrating the link survivability protection based on the positioning circle in the reconstructed sub-topology according to an embodiment of this application. Figure 4This is a comparison chart showing the prediction time of network topology for different nodes involved in the embodiments of this application under different machine learning models; Figure 5 This is a comparison chart showing the prediction time of network topology for different nodes involved in the embodiments of this application under different machine learning models; Figure 6 This is a schematic diagram of the hardware operating environment of the computer system involved in the embodiments of this application.

[0017] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] To better understand the above technical solutions, exemplary embodiments of this disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of this disclosure to those skilled in the art.

[0019] First Embodiment Reference Figure 1 This embodiment provides a link protection method for all-optical intelligent computing centers, which includes the following steps: S10, obtain the failure probability of each link in the network topology predicted by the support vector machine model, wherein the network topology is constructed with the devices of the all-optical intelligent computing center as nodes and the links as edges; In this embodiment, the failure probability of each link in the network topology is first predicted by the support vector machine (SVM) model.

[0020] By treating devices in the intelligent computing center as nodes and links as edges, a network topology is constructed.

[0021] The network topology includes the main characteristics of each link: failure time interval, link speed, latency, packet loss rate, and service type. The service type is mainly determined by the number of link failures and is mainly divided into sensitive, general, and insensitive types. The network topology and edge relationships are transformed into matrix form, and the edge relationship matrix is ​​concatenated with the link characteristics to form the total dataset.

[0022] The essence of SVM model prediction of link failure is to calculate a classification function, or decision function, for all link data, and determine whether a link failure has occurred based on the correct classification of the link categories according to the decision value.

[0023] Furthermore, and optionally, the following steps can be used to train a support vector machine model: S40, construct a network topology by using the intelligent computing center equipment as nodes and links as edges, collect the link characteristics of each link in the network topology, and construct a total dataset based on the link characteristics and the network topology; S50, Select a portion of the total dataset as a training dataset, and input the training dataset into the support vector machine model so that the support vector machine model can perform fault probability prediction training based on the training dataset; S60 predicts the failure probability of each link in the network topology based on the trained support vector machine model.

[0024] In some optional implementations, wT*x+b represents the hyperplane for classification, where x represents the feature vector composed of various features of the link, w is the normal vector perpendicular to the hyperplane, and b is the distance from the sample to the hyperplane. All points when wT*x+b>0 belong to one class, and all points when wT*x+b<0 belong to another class. These two classes are the two classes used to determine whether a fault has occurred in this embodiment. In optical network fault prediction research, because fault data contains complex models and features, linear models may not be able to capture nonlinear relationships. Therefore, we consider constructing complex nonlinear hyperplanes for data partitioning. Non-support vector machines use kernel functions to map the nonlinear features of the original data in low-dimensional space to high-dimensional space for partitioning. After training the training dataset using the above model, the parameters of the test training set are input into the final optimized support vector machine model to predict whether a link has failed.

[0025] Further and optionally, the support vector machine model, when predicting the failure probability, includes the following steps: Read the failure time and number of failures for each link; Initialize the experienced time to 0. When the experienced time is less than the specified total time, check if the fault interval is less than the total time. If it is, increment the fault count by one. The experienced time is equal to the current experienced time plus the fault interval. Repeat this calculation until the experienced time is greater than or equal to the total time. The fault count is obtained through the above calculation. Finally, divide the fault count by the total time to obtain the fault probability of the link.

[0026] S20, based on the numerical range of the failure probability of each link, determine the failure level corresponding to each link, and divide the network topology into multiple sub-networks based on the failure level; In this step, the network topology is divided into multiple subnetworks with different failure levels based on the predicted failure probability of each link.

[0027] Further and optionally, when the failure probability is within the first interval, the failure level corresponding to the link is determined to be a high failure level, and the link in the high failure level is divided into a first sub-network; When the failure probability is within the second interval, the failure level corresponding to the link is determined to be a medium failure level, and the link at the medium failure level is divided into a second sub-network. When the failure probability is within the third interval, the failure level corresponding to the link is determined to be a low failure level, and the link in the low failure level is divided into the third sub-network. Wherein, the first interval is greater than the second interval, which is greater than the third interval.

[0028] In some alternative implementations, the first interval is the fault probability ≤ 1 / 3 quantile; The second interval is the 1 / 3 quantile < failure probability ≤ 2 / 3 quantile; The third interval represents the probability of failure greater than the 2 / 3 quantile.

[0029] S30, link protection is performed on a per-sub-network basis, wherein the protection priority of sub-networks with higher fault levels is higher than that of sub-networks with lower fault levels.

[0030] In this embodiment, after dividing the sub-networks, link protection is performed on a per-sub-network basis, and more resources are allocated to sub-networks with high fault levels, so that their protection priority is higher than that of sub-networks with lower fault levels.

[0031] In the technical solution provided in this embodiment, compared with the traditional full-link fault detection, this application divides the link by the fault probability predicted by the support vector machine, takes the sub-network as the unit, and adopts different priority strategies for link protection based on the sub-networks with different fault levels, thereby shortening the fault location time and increasing the location accuracy.

[0032] Second Embodiment Based on the first embodiment, to prevent links at each level from failing to form a new topology, this embodiment employs a connection guarantee mechanism to avoid situations where there are too many or too few nodes in each sub-network. Specifically, this includes: Perform topology verification on links of each level to determine whether links of that level can form a new topology; If the links at this level cannot form a new topology, then the link with the lowest failure probability among the links connected to the nodes in the disconnected area is selected as a "supplementary link" to supplement the sub-topology until a closed loop is formed to form a new network topology.

[0033] Third Embodiment Based on any of the above embodiments, this embodiment provides a specific implementation method for link protection at the sub-network level, specifically constructing the following objective function or constraint: (1-1) (1-2) (1-3) (1-4) (1-5) (1-6) (1-7) (1-8) (1-9) In the formula, N represents the set of all nodes in the network, E represents the set of all links in the network, and W represents the capacity of each link. This represents the requirements of node u. This represents the number of the circle set CSj, which is also the number of the positioning circle. Indicates a link in the network ( u , v Traffic on ) This represents the cost of increasing the capacity per unit on the link. A binary variable, representing if the vector u → v In CS j If the circle is on the left, the value is 1; otherwise, it is 0. This is a binary variable, representing if the node u In CS j If the circle is on the left, the value is 1; otherwise, it is 0. This is a binary variable, representing whether the link can be accessed by CS. j If the circle is protected, the value is 1; otherwise, it is 0. This is a binary variable, representing a value of 1 if node u is the root node of CSj, and 0 otherwise. A non-negative decimal variable, indicating that in constructing CS j During the middle circle process, nodes u voltage value, For a predefined nonnegative decimal constant, it must satisfy... ; This is a binary variable, representing a value of 1 if optical path p is used as the working path for node u on working capacity w, and 0 otherwise.

[0034] It should be noted that Equation (1-1) is the objective function, which requires minimizing the total cost of all preset loops, reducing the total resource usage, and satisfying the protection requirements.

[0035] Constraint (1-2) represents each link in the positioning circle. Only can be used or It is represented by one of the directions.

[0036] Constraint (1-3) stipulates that each node in the network can have either 2 or 0 vectors on the loop.

[0037] Constraints (1-4) ensure that every link in the network is 100% protected.

[0038] Constraint (1-5) stipulates that there can only be one root node in each positioning circle.

[0039] Constraint (1-6) stipulates that only the root node can be the head of multiple vectors, while other nodes can only be the head of one vector at most. This ensures that there can only be one pair of root node-flip node in each positioning circle.

[0040] The constraint (1-7) stipulates that the pressure value of the tail vector is greater than that of the head vector.

[0041] Constraint (1-8) stipulates that if a link can be protected by a positioning circle, then both of its nodes must be on this circle.

[0042] Constraint (1-9) expresses the service requirements of node u as du. If the predicted failure probability of the optical path available to carry the node's service is high, it is necessary to consider how to meet the service requirements through other optical paths when calculating the service requirements.

[0043] Verification Implementation Examples This embodiment verifies the effectiveness of the proposed method. Figure 2 The diagram illustrates how a large network is divided into three different levels of small topologies based on link failure probability. Figure 3 This diagram illustrates how, after topology reconstruction, each of the 11 nodes and 26 edges achieves link survivability protection based on the p-cycle positioning circle protection algorithm.

[0044] This embodiment demonstrates a comparison of prediction times for network topologies with different node counts (11 nodes, 30 nodes, 50 nodes, and 100 nodes) under different machine learning models (SVM, LSTM, LSSVM, BP, Random Forest, ELM). Figure 4 For example, accuracy Figure 5 .according to Figure 4 and Figure 5The comparison shows that in small networks, different machine learning models have similar prediction times, but SVM and Random Forest models have relatively higher accuracy, while LSSVM and ELM models have relatively lower accuracy. In large networks, the ELM model has the shortest prediction time but the lowest accuracy, the SVM model has the longest computation time but relatively high accuracy, the LSSVM model has a long computation time and low accuracy, the LSTM model has a short computation time and moderate accuracy, the BP model has a moderate computation time and moderate accuracy, and the Random Forest algorithm has a significant advantage in computation time and higher accuracy in large networks compared to small networks. Furthermore, the computation time of different machine learning models combined with the p-cycle algorithm was compared at different nodes. Figure 6 It is clear that the combination of the SVM model and the p-cycle algorithm requires the least computation time. While a single machine learning model has a short fault prediction time but low accuracy, the original p-cycle algorithm has high accuracy but a long computation time. The proposed method of combining the SVM model and the p-cycle algorithm achieves higher accuracy than a single machine learning model while requiring less computation time than the traditional p-cycle algorithm.

[0045] As one implementation scheme, Figure 6 This is a schematic diagram of the hardware operating environment of the computer system involved in the embodiments of this application.

[0046] like Figure 6 As shown, the computer system may include: a processor 1001, such as a CPU; a memory 1005; a user interface 1003; a network interface 1004; and a communication bus 1002. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen or an input unit such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be high-speed RAM or non-volatile memory, such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0047] Those skilled in the art will understand that Figure 6 The computer system architecture shown does not constitute a limitation on the computer system and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0048] like Figure 6As shown, the memory 1005, as a storage medium, may include an operating system, a network communication module, a user interface module, and computer programs. The operating system is a program that manages and controls the hardware and software resources of the computer system, as well as the operation of the computer programs and other software or programs.

[0049] exist Figure 6 In the computer system shown, the user interface 1003 is mainly used to connect to the terminal and communicate with the terminal; the network interface 1004 is mainly used to communicate with the backend server; and the processor 1001 can be used to call the computer program stored in the memory 1005.

[0050] In this embodiment, the computer system includes: a memory 1005, a processor 1001, and a computer program stored in the memory and executable on the processor, wherein: When processor 1001 calls a computer program stored in memory 1005, it performs the following operations: S10, obtain the failure probability of each link in the network topology predicted by the support vector machine model, wherein the network topology is constructed with the devices of the all-optical intelligent computing center as nodes and the links as edges; S20, based on the numerical range of the failure probability of each link, determine the failure level corresponding to each link, and divide the network topology into multiple sub-networks based on the failure level; S30, link protection is performed on a per-sub-network basis, wherein the protection priority of sub-networks with high fault levels is higher than that of sub-networks with low fault levels.

[0051] When processor 1001 calls a computer program stored in memory 1005, it performs the following operations: S40, construct a network topology by using the intelligent computing center equipment as nodes and links as edges, collect the link characteristics of each link in the network topology, and construct a total dataset based on the link characteristics and the network topology; S50, Select a portion of the total dataset as a training dataset, and input the training dataset into the support vector machine model so that the support vector machine model can perform fault probability prediction training based on the training dataset; S60 predicts the failure probability of each link in the network topology based on the trained support vector machine model.

[0052] Optionally, the support vector machine model, when predicting the failure probability, includes the following steps: Read the interval and number of failures for each link; Initialize the experienced time to 0. When the experienced time is less than the specified total time, check if the fault interval is less than the total time. If it is, increment the fault count by one. The experienced time is equal to the current experienced time plus the fault interval. Repeat this calculation until the experienced time is greater than or equal to the total time. The fault count is obtained through the above calculation. Finally, divide the fault count by the total time to obtain the fault probability of the link.

[0053] When processor 1001 calls a computer program stored in memory 1005, it performs the following operations: When the failure probability is within the first interval, the failure level corresponding to the link is determined to be a high failure level, and the link with the high failure level is divided into the first sub-network. When the failure probability is within the second interval, the failure level corresponding to the link is determined to be a medium failure level, and the link at the medium failure level is divided into a second sub-network. When the failure probability is within the third interval, the failure level corresponding to the link is determined to be a low failure level, and the link in the low failure level is divided into the third sub-network. Wherein, the first interval is greater than the second interval, which is greater than the third interval.

[0054] When processor 1001 calls a computer program stored in memory 1005, it performs the following operations: Perform topology verification on links of each level to determine whether links of that level can form a new topology; If the links at this level cannot form a new topology, then the link with the lowest failure probability among the links connected to the nodes in the disconnected area is selected as a "supplementary link" to supplement the sub-topology until a closed loop is formed to form a new network topology. When processor 1001 calls a computer program stored in memory 1005, it performs the following operations: When performing link protection on a sub-network basis, the following constraints must be met: In the formula, N represents the set of all nodes in the network, E represents the set of all links in the network, and W represents the capacity of each link. This represents the requirements of node u. This represents the number of the circle set CSj, which is also the number of the positioning circle. Indicates a link in the network ( u , v Traffic on ) This represents the cost of increasing the capacity per unit on the link. A binary variable, representing if the vector u → v In CS j If the circle is on the left, the value is 1; otherwise, it is 0. This is a binary variable, representing if the node u In CS j If the circle is on the left, the value is 1; otherwise, it is 0. This is a binary variable, representing whether the link can be accessed by CS. j If the circle is protected, the value is 1; otherwise, it is 0. This is a binary variable, representing a value of 1 if node u is the root node of CSj, and 0 otherwise. A non-negative decimal variable, indicating that in constructing CS j During the middle circle process, nodes u voltage value, For a predefined nonnegative decimal constant, it must satisfy... ; This is a binary variable, representing a value of 1 if optical path p is used as the working path for node u on working capacity w, and 0 otherwise.

[0055] Furthermore, this embodiment also proposes a link protection model, which includes: The fault probability prediction module is used to obtain the fault probability of each link in the network topology predicted by the support vector machine model, wherein the network topology is constructed with the devices of the all-optical intelligent computing center as nodes and the links as edges. The network partitioning module is used to determine the fault level corresponding to each link based on the numerical range of the fault probability of each link, and to divide the network topology into multiple sub-networks based on the fault level. The link protection module is used to perform link protection on a per-sub-network basis, wherein the protection priority of sub-networks with high fault levels is higher than that of sub-networks with low fault levels.

[0056] Furthermore, as an implementation scheme, this embodiment also proposes a server that employs the link protection method for all-optical intelligent computing centers as described in any of the above embodiments.

[0057] Furthermore, those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program includes program instructions and can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in a computer system to implement the process steps of the embodiments of the above methods.

[0058] Therefore, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the various steps of the link protection method for an all-optical intelligent computing center as described in the above embodiments.

[0059] The computer-readable storage medium can be any computer-readable storage medium capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory (ROM), magnetic disk, or optical disk.

[0060] It should be noted that, since the storage medium provided in the embodiments of this application is the storage medium used to implement the methods of the embodiments of this application, those skilled in the art can understand the specific structure and variations of the storage medium based on the methods described in the embodiments of this application, and therefore will not be repeated here. All storage media used in the methods of the embodiments of this application fall within the scope of protection of this application.

[0061] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0062] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0063] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0064] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0065] It should be noted that any reference signs placed between parentheses in the claims should not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. This application can be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.

[0066] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0067] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A link protection method applied to an all-optical intelligent computing center, characterized in that, The method includes the following steps: S10, obtain the failure probability of each link in the network topology predicted by the support vector machine model, wherein the network topology is constructed with the devices of the all-optical intelligent computing center as nodes and the links as edges; S20, based on the numerical range of the failure probability of each link, determine the failure level corresponding to each link, and divide the network topology into multiple sub-networks based on the failure level; S30, link protection is performed on a per-sub-network basis, wherein the protection priority of sub-networks with high fault levels is higher than that of sub-networks with low fault levels.

2. The link protection method for all-optical intelligent computing centers as described in claim 1, characterized in that, Before S10, the following are also included: S40, construct a network topology by using the intelligent computing center equipment as nodes and links as edges, collect the link characteristics of each link in the network topology, and construct a total dataset based on the link characteristics and the network topology; S50, Select a portion of the total dataset as a training dataset, and input the training dataset into the support vector machine model so that the support vector machine model can perform fault probability prediction training based on the training dataset; S60 predicts the failure probability of each link in the network topology based on the trained support vector machine model.

3. The link protection method for all-optical intelligent computing centers as described in claim 1 or 2, characterized in that, The support vector machine model, when predicting the failure probability, includes the following steps: Read the interval and number of failures for each link; Initialize the experienced time to 0. When the experienced time is less than the specified total time, check if the fault interval is less than the total time. If it is, increment the fault count by one. The experienced time is equal to the current experienced time plus the fault interval. Repeat this calculation until the experienced time is greater than or equal to the total time. The fault count is obtained through the above calculation. Finally, divide the fault count by the total time to obtain the fault probability of the link.

4. The link protection method for all-optical intelligent computing centers as described in claim 1, characterized in that, S20 includes: When the failure probability is within the first interval, the failure level corresponding to the link is determined to be a high failure level, and the link with the high failure level is divided into the first sub-network. When the failure probability is within the second interval, the failure level corresponding to the link is determined to be a medium failure level, and the link at the medium failure level is divided into a second sub-network. When the failure probability is within the third interval, the failure level corresponding to the link is determined to be a low failure level, and the link in the low failure level is divided into the third sub-network. Wherein, the first interval is greater than the second interval, which is greater than the third interval.

5. The link protection method for all-optical intelligent computing centers as described in claim 4, characterized in that, Before step S20, the method further includes: Perform topology verification on links of each level to determine whether links of that level can form a new topology; If the links at this level cannot form a new topology, then the link with the lowest failure probability among the links connected to the nodes in the disconnected area is selected as a "supplementary link" to supplement the sub-topology until a closed loop is formed to form a new network topology.

6. The link protection method for all-optical intelligent computing centers as described in claim 1, characterized in that, Step S30 includes: When performing link protection on a sub-network basis, the following constraints must be met: ； ； ； ； ； ； ； ； ； In the formula, N represents the set of all nodes in the network, E represents the set of all links in the network, and W represents the capacity of each link. This represents the requirements of node u. This represents the number of the circle set CSj, which is also the number of the positioning circle. Indicates a link in the network ( u , v Traffic on ) This represents the cost of increasing the capacity per unit on the link. A binary variable, representing if the vector u → v In CS j If the circle is on the left, the value is 1; otherwise, it is 0. This is a binary variable, representing if the node u In CS j If the circle is on the left, the value is 1; otherwise, it is 0. This is a binary variable, representing whether the link can be accessed by CS. j If the circle is protected, the value is 1; otherwise, it is 0. This is a binary variable, representing a value of 1 if node u is the root node of CSj, and 0 otherwise. A non-negative decimal variable, indicating that in constructing CS j During the middle circle process, nodes u voltage value, For a predefined nonnegative decimal constant, it must satisfy... ; This is a binary variable, representing a value of 1 if optical path p is used as the working path for node u on working capacity w, and 0 otherwise.

7. A link protection model, characterized in that, The link protection model includes: The fault probability prediction module is used to obtain the fault probability of each link in the network topology predicted by the support vector machine model, wherein the network topology is constructed with the devices of the all-optical intelligent computing center as nodes and the links as edges. The network partitioning module is used to determine the fault level corresponding to each link based on the numerical range of the fault probability of each link, and to divide the network topology into multiple sub-networks based on the fault level. The link protection module is used to perform link protection on a per-sub-network basis, wherein the protection priority of sub-networks with high fault levels is higher than that of sub-networks with low fault levels.

8. A server employing the link protection method for an all-optical intelligent computing center as described in any one of claims 1 to 7.

9. A computer system, characterized in that, The computer system includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps of the link protection method for an all-optical intelligent computing center as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the link protection method for an all-optical intelligent computing center as described in any one of claims 1 to 6.

Images