Link Failure Protection Method, System, Device and Terminal for Hybrid SDN Network

Abstract

The present invention belongs to the technical field of software-defined networks, and discloses a link fault protection method, system, device and terminal for a hybrid SDN network. The system proposes an FP-SCS method for solving the SDN candidate selection problem, which comprehensively considers the node's fault protection ability and the link's fault protection difficulty, so as to deploy the minimum required SDN switches that can protect all links. On the basis of deploying SDN switches, taking advantage of the flexibility of SDN switch routing and relying on the destination node-based tunnel mechanism, an FP-PPS algorithm for protection path selection is further proposed, which can select the most suitable protection path according to the importance of the mutually influencing path parameters. The present invention can effectively and quickly select the minimum number of SDN switches that can protect all links, and at the same time fully consider the mutually influencing path parameters, so as to flexibly select the most suitable protection path.

Description

Technical Field

[0001] The present invention belongs to the technical field of software-defined networks, and particularly relates to a link fault protection method, system, device and terminal for a hybrid SDN network. Background Art

[0002] At present, Software-Defined Networks (SDN) have evolved into a new network model to improve network functions, including flexible routing, traffic engineering, and access control. Due to a large amount of time and economic requirements, organizations are reluctant to completely start over with SDN infrastructure. Therefore, a hybrid SDN network in which SDN switches coexist with traditional routers has emerged.

[0003] Performing fast fault protection on a hybrid SDN network can effectively improve the resilience and robustness of the network. By deploying SDN switches in a traditional network, two goals need to be achieved for fault protection of a hybrid SDN network:

[0004] 1) Deploy the fewest SDN switches to protect all faulty links, which is called SDN Candidate Selection (SCS);

[0005] 2) Select a suitable protection path to eliminate the impact of the faulty link on network performance, thereby ensuring the quality of service of users, which is called Protection Path Selection (PPS).

[0006] Existing research results on fault protection of hybrid SDN networks have extensively investigated and studied the SCS and PPS problems, but there are still some deficiencies that need to be improved urgently. Specifically, existing research only focuses on the node fault protection ability or the difficulty of link fault protection when solving the SCS problem, without comprehensively considering these factors. In addition, existing research on the PPS problem also has limitations, mainly reflected in that the protection path length (PPL) and the maximum link utilization (MLU), as path parameters that interact with each other, do not fully consider their interaction. Therefore, the selected protection path may not fully meet the actual needs of users or is a non-optimal solution.

[0007] Fault protection is an important part of network management, and many studies are dedicated to the SCS and PPS problems in hybrid SDN network fault protection.

[0008] Chu et al. used the Greedy strategy to select nodes with the maximum fault protection ability in turn to deploy SDN switches, so as to protect all links. In addition, the Greedy-LB scheme was further proposed to select a protection path that can achieve load balancing. Its main idea is to select the path with the minimum MLU by means of a link-based tunneling mechanism. However, the Greedy strategy does not consider the impact of the fault protection difficulty of different links on the deployment of SDN switches, and the path selection of Greedy-LB lacks flexibility and consideration of the path length. To improve these defects, Yang et al. proposed that when there are multiple nodes with the maximum protection ability, the node with a shorter average PPL should be preferentially selected to be upgraded to an SDN switch. For the PPS problem, they proposed the RPR-DT algorithm based on the destination node tunneling mechanism to minimize the PPL, and the RPR-LB algorithm to relax the requirement for the PPL to minimize the MLU. Li et al. found all feasible SDN candidate solutions by using a search tree according to the fault protection difficulty of the links, and selected the solution with the highest reliability. Considering the protection path length and network load balancing, the SDN switch with the maximum utility is selected for each link. However, when a certain link fails, only the selected SDN switches can be used for protection, so the selection range of its candidate protection paths has strong limitations.

[0009] SDN decouples the control plane and data plane of forwarding devices, which can effectively improve many network functions including fault protection. In the current Internet architecture, traditional networks are gradually evolving into fully SDN networks to provide flexible services and applications. In this process, hybrid SDN networks are derived. How to select the fewest traditional devices to be upgraded to SDN switches to achieve the goal of protecting all faulty links, and how to select the most suitable protection path that meets user service requirements have become urgent problems to be solved.

[0010] Through the above analysis, the problems and defects of the existing technologies are as follows:

[0011] (1) When the existing technologies solve the SCS problem of fault protection in hybrid SDN networks, if the node fault protection ability is used to select the nodes to be upgraded to SDN switches, the impact of the link fault protection difficulty on the required SDN switches is ignored, and introducing it can reduce and optimize the process or running time of SCS;

[0012] (2) In the methods of SCS from the perspective of link fault protection difficulty, the impact of node fault protection ability on solving the SCS problem is not considered, which may lead to an increase in the number of required SDN switches. The existing technologies do not comprehensively analyze the impact of the node fault protection ability and the link fault protection difficulty on the required SDN switch solutions when performing SCS;

[0013] (3) After determining the deployment plan of the SDN switch, the prior art does not fully utilize the flexible routing capabilities of the SDN switch when solving the PPS problem and does not adequately consider the performance parameters of the protection path. It only optimizes one of these parameters. However, the protection path has multiple parameters that simultaneously affect network performance, including PPL and MLU, etc.;

[0014] (4) When selecting the protection path, the interaction between path parameters such as PPL and MLU is not fully considered. Therefore, the selected protection path may not fully meet the actual needs of users or may not be able to obtain the optimal solution. Summary of the Invention

[0015] In view of the problems existing in the prior art, the present invention provides a link fault protection method, system, device, and terminal for a hybrid SDN network.

[0016] The present invention is implemented as follows. A link fault protection method for a hybrid SDN network includes: The system proposes an FP-SCS method for solving the SDN candidate selection problem, which comprehensively considers the node's fault protection ability and the link's fault protection difficulty, thereby deploying the minimum number of SDN switches required to protect all links. On the basis of deploying the SDN switch, using the flexibility of the SDN switch routing and relying on the tunnel mechanism based on the destination node, an FP-PPS algorithm for protecting path selection is further proposed, which can select the most suitable protection path according to the importance of the mutually influencing path parameters.

[0017] Furthermore, the link fault protection method for a hybrid SDN network includes the following steps:

[0018] Step 1, according to the goals and constraints of the actual service requirements of users, establish an optimization model for the link fault protection problem of the hybrid SDN network;

[0019] Step 2, propose an FP-SCS method for the SCS problem involved, which aims to minimize the number of SDN switches required to protect all links;

[0020] Step 3, for the PPS problem, use the flexibility of the SDN switch routing and rely on the tunnel mechanism based on the destination node to propose an FP-PPS algorithm to select a suitable protection path and ultimately achieve fault protection for the hybrid SDN network.

[0021] Furthermore, the establishment of the optimization model in Step 1 includes:

[0022] The network topology is represented by a directed graph G=(V, E), where V is the set of nodes and E is the set of links. The bidirectional link between node i and node j includes two directed links, e1 = <i, j> and e2 = <j, i>. For the network topology of a 2-edge connected graph that can ensure the existence of a protection path, when link e fails, the path that cannot transmit data packets is called the affected path, and it is replaced by a backup protection path that does not pass through link e.

[0023] According to the goals and constraints of the actual service requirements of users, an optimization model is established for the link failure protection problem of the hybrid SDN network; among them, the optimization model is shown as follows:

[0024] Minimize:

[0025]

[0026] The objective function is to minimize the number of SDN switches required for fault protection of all links in the hybrid SDN network;

[0027] Subject to:

[0028]

[0029] Its constraint is that at least one designated SDN switch can be found by the end routers of the failed link e to transmit data packets;

[0030]

[0031] It ensures that only one designated SDN switch is configured for the end routers when link e fails;

[0032] At the same time, the node configured as the designated SDN switch must have been upgraded from a traditional router to an SDN switch;

[0033]

[0034] In the formula, is a binary symbol. When node x is the end router of link e otherwise is a binary symbol. When the path from node s to d passes through link e otherwise is a binary symbol. When node k is selected as the designated SDN switch of node x when link e fails otherwise is a binary symbol. When node m is a one-hop neighbor of node k otherwise u k is a binary symbol. When node k is selected to be upgraded to an SDN switch, u k = 1; otherwise, u k = 0.

[0035] Furthermore, the FP - SCS method for SDN candidate selection in step two includes:

[0036] Based on the deployment cost of evolving to a fully SDN network and the impact of device downtime on network user services, the SCS problem in hybrid SDN network fault protection aims to minimize the number of SDN switches required to protect all links, and the FP - SCS method is proposed.

[0037] The FP - SCS method specifically includes: constructing an SDN candidate table tSDN to identify the locations of suitable candidate SDN switches for each link e; if node δ τ protects link e, then mark tSDN(e, τ) as 1, otherwise mark it as 0; based on tSDN, select important nodes and store them in sSDN, which is responsible for protecting the links with the highest protection difficulty and only protected by one node; delete the links already protected by sSDN from tSDN to update the candidate table; reduce the candidate node set to be pre - upgraded to an SDN switch to csSDN according to the link with the highest protection difficulty among the unprotected links; with the idea of the greedy strategy, select the node in csSDN with the greatest protection ability for the unprotected links and add it to sSDN; until the selected nodes sSDN have the ability to protect all links, which is the final set of nodes to be upgraded to SDN switches, and the FP - SCS method ends.

[0038] Furthermore, the FP - PPS algorithm for protection path selection in step three includes:

[0039] After determining the minimum number of SDN switches using the FP - SCS method, use the flexible routing of SDN switches to select appropriate protection paths to finally achieve fault protection for the hybrid SDN network. For the PPS problem, the FP - PPS algorithm is proposed with the tunnel mechanism based on the destination node, analyze the path parameters of the mutually influential protection path length (PPL) and maximum link utilization (MLU), and select the most appropriate protection path according to the importance of each parameter.

[0040] The FP - PPS algorithm specifically includes: based on the traffic matrix T N between network nodes, after normal traffic transmission on the unaffected paths, update T LThe traffic data of each link in; Based on the influence of the different protection path selection orders of multiple affected paths on the overall traffic distribution of the network after protection, the affected paths are sorted in descending order according to the required tunnel traffic size on the affected paths, and then appropriate protection paths are selected for them in turn; Use the tunnel mechanism based on the destination node to construct all available protection paths for the affected paths, and set path scores to evaluate the performance of each path; According to the property that the smaller the path parameters PPL and MLU, the better the path performance, select the protection path with the smallest path score; If there is a tie and α is large, it means that PPL dominates the user's needs, then select the protection path with a smaller PPL; Otherwise, select the protection path with a smaller MLU.

[0041] Furthermore, according to the following formula, the path score of the protection path from node s to d when link e fails is calculated to evaluate the path performance:

[0042]

[0043] In the formula, α and β are variable weights set according to the actual needs of the user for PPL and MLU, and α + β = 1; and is the length of the protection path from node s to d and the normalized result of the maximum link utilization rate of the path to unify all path parameters to the same order of magnitude.

[0044] Another object of the present invention is to provide a link fault protection system for a hybrid SDN network applying the link fault protection method of the hybrid SDN network. The link fault protection system for the hybrid SDN network includes:

[0045] An optimization model construction module for establishing an optimization model for link fault protection of a hybrid SDN network according to the objectives and constraints of actual service requirements;

[0046] An SDN candidate selection module for using the FP-SCS method to solve the SDN candidate selection problem to minimize the number of SDN switches required to protect all links;

[0047] A protection path selection module for selecting an appropriate protection path by means of the routing flexibility of the SDN switch and the tunnel mechanism based on the destination node, and using the FP-PPS algorithm to achieve fault protection for the hybrid SDN network.

[0048] Another object of the present invention is to provide a computer device. The computer device includes a memory and a processor. When the computer program stored in the memory is executed by the processor, the processor executes the steps of the link fault protection method of the hybrid SDN network.

[0049] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to execute the steps of the link fault protection method for the hybrid SDN network.

[0050] Another object of the present invention is to provide an information data processing terminal for implementing the link fault protection system for the hybrid SDN network.

[0051] Combined with the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solutions to be protected by the present invention are as follows:

[0052] First, in view of the technical problems existing in the above-mentioned prior art and the difficulty of solving this problem, closely combining the technical solutions to be protected by the present invention and the results and data in the research and development process, etc., analyze in detail and deeply how the technical solutions of the present invention solve the technical problems and the creative technical effects brought after solving the problems. The specific description is as follows:

[0053] On the premise of comprehensively considering the link fault protection difficulty and the node fault protection ability, the present invention proposes an FP-SCS method for SDN candidate selection (SCS), which can deploy the minimum number of SDN switches required to protect all links. Based on the deployed SDN switches, the present invention further proposes an FP-PPS algorithm for protection path selection (PPS). This algorithm utilizes the flexibility of SDN switch routing and, with the help of the tunnel mechanism based on the destination node, can flexibly select the most suitable protection path according to the importance of the mutually influencing path parameters. The simulation results show that the proposed solution of the present invention has better performance compared with the existing solutions.

[0054] In order to improve the deficiencies in the prior art and further solve the SCS and PPS problems, the present invention proposes a novel link fault protection scheme for the hybrid SDN network. The main contributions are as follows:

[0055] (1) By comprehensively considering the link fault protection difficulty and the node fault protection ability, an FP-SCS method is proposed, which can select the minimum number of SDN switches to protect all links;

[0056] (2) Based on the SDN switches selected by the FP-SCS method, for the PPS problem, an FP-PPS algorithm is proposed by utilizing the flexibility of SDN switch routing and with the help of the tunnel mechanism based on the destination node; FP-PPS fully considers the mutually influencing path parameters such as PPL and MLU, and can flexibly select the most suitable protection path according to the importance of these parameters;

[0057] (3) The performance of the proposed FP-SCS and FP-PPS methods of the present invention was evaluated through simulation experiments; in addition, comparative experiments with existing solutions showed that the proposed solution of the present invention is superior to existing solutions.

[0058] Second, considering the technical solution as a whole or from the perspective of the product, the technical effects and advantages of the technical solution to be protected by the present invention are specifically described as follows:

[0059] Aiming at the existing problems, the present invention respectively proposes optimized solutions such as FP-SCS and FP-PPS for the SCS and PPS problems.

[0060] Regarding the SCS and PPS problems of hybrid SDN network fault protection, the present invention comprehensively considers the difficulty of link fault protection and the node's fault protection ability, and respectively proposes optimized solutions such as FP-SCS and FP-PPS, which make up for the deficiencies of the existing link fault protection solutions for hybrid SDN networks and greatly promote the development of future new networks; among them, FP-SCS comprehensively considers the protection difficulty of the link and the node's protection ability, and can effectively and quickly select the fewest SDN switches that can protect all links; based on these SDN switches, FP-PPS fully considers the mutually influential path parameters, so as to flexibly select the most suitable protection path. The results of a large number of simulation experiments show that the hybrid SDN network fault protection solution of the present invention is superior to existing solutions, and has important research significance for improving the robustness and stability of hybrid SDN networks and the development of future new networks.

[0061] Third, as the creative auxiliary evidence of the claims of the present invention, it is also reflected in the following important aspects:

[0062] (1) The expected benefits and commercial value after the transformation of the technical solution of the present invention are:

[0063] The occurrence of network failures poses great challenges to the robustness, stability, and adaptability of the network. Link failures are also inevitable in future new networks, especially in hybrid SDN networks. Protecting hybrid SDN networks with link failures can effectively reduce the impact of failures on the normal operation of the network and the provision of user services. However, there are still many deficiencies in the means of protecting link failures in hybrid SDN networks at present. In this invention, when selecting SDN candidates, the protection difficulty of links and the protection ability of nodes are comprehensively considered to reduce the network deployment overhead and equipment downtime. It can select the fewest number of nodes to be upgraded to SDN switches in the shortest possible time to protect all links. At the same time, on the deployed SDN switches, the proposed protection path selection algorithm in this invention can select appropriate protection paths with multiple path parameters meeting the actual needs of users, making up for the deficiencies of the existing technologies to a certain extent. The technical solution of this invention has obvious advantages and good application prospects compared with other existing solutions, providing an effective solution for the requirements of hybrid SDN network fault protection.

[0064] (2) The technical solution of this invention fills the technical gaps in the domestic and international industries:

[0065] Protecting link failures in hybrid SDN networks not only requires solving the SDN candidate selection problem to reduce network deployment overhead and equipment downtime, but also it is equally important to select multiple alternative candidate protection paths for the affected paths using the flexibility of its routing in the network where SDN switches are deployed. Existing technical solutions are only limited to focusing on and researching only one of the SDN candidate selection or protection path selection problems. However, this invention proposes the FP-SCS method and the FP-PPS algorithm for the SDN candidate selection and protection path selection problems respectively, thus providing a new and unified optimization solution that can completely solve the hybrid SDN network link failure protection problem and optimize the performance of the existing technical solutions.

[0066] (3) The technical solution of this invention solves the technical problems that people have always been eager to solve but have never been successful in obtaining:

[0067] At present, traditional communication networks are characterized by privatization and rigidity. In order to meet the needs of existing network users, future new networks such as SDN networks have emerged as the times require. Fault protection for hybrid SDN networks derived during the transition to a fully SDN network is an important part of its fault management, which can effectively enhance the robustness and resilience of the network. However, existing technologies lack sufficient consideration of the optimal SDN switch deployment scheme during SDN candidate selection and various parameters affecting network performance during protection path selection. Based on this, the present invention introduces the difficulty of link fault protection into SDN candidate selection in a greedy strategy aiming at maximizing the node fault protection ability, and adjusts various path parameters according to user requirements during protection path selection. Through simulation experiments, it is proved that the technical solution of the present invention is superior to existing solutions, can effectively reduce the number of required SDN switches or running time, and select appropriate protection paths, thus fully meeting the actual needs of network users.

[0068] (4) The technical solution of the present invention overcomes technical prejudice:

[0069] With the continuous development of future new networks, the link fault protection technology of hybrid SDN networks has gradually advanced and tended to be mature, but there are still a large number of problems to be solved and optimization directions. It mainly includes that existing technologies do not comprehensively analyze the node fault protection ability and the difficulty of link fault protection during SDN candidate selection, and cannot obtain the optimal SDN candidate selection scheme; during protection path selection, only a single parameter of the path is considered and optimized, such as PPL or MLU, but the protection path has multiple parameters that simultaneously affect network performance and are ignored. Therefore, the present invention introduces the difficulty of link fault protection into SDN candidate selection in a greedy strategy aiming at maximizing the node fault protection ability, effectively reducing the number of required SDN switches or running time, and adjusting various mutually influencing path parameters according to user requirements during protection path selection to select appropriate protection paths, and fully realizing the link fault protection of hybrid SDN networks on the premise of fully meeting the actual needs of network users. Brief Description of the Drawings

[0070] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the embodiments of the present invention. Obviously, the following described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0071] Figure 1 It is a flowchart of the link fault protection method for a hybrid SDN network provided by an embodiment of the present invention;

[0072] Figure 2It is a schematic diagram of the FP-SCS method framework provided by an embodiment of the present invention;

[0073] Figure 3 It is a schematic diagram of the influence of network scale on SDN candidate selection provided by an embodiment of the present invention;

[0074] Figure 4A It is a schematic diagram of the applicability comparison of the average protection path length APPL provided by an embodiment of the present invention;

[0075] Figure 4B It is a schematic diagram of the applicability comparison of the average maximum link utilization rate AMLU provided by an embodiment of the present invention. Detailed implementation manners

[0076] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention.

[0077] Aiming at the problems existing in the prior art, the present invention provides a link failure protection method, system, device and terminal for a hybrid SDN network, and the present invention will be described in detail below with reference to the accompanying drawings.

[0078] As Figure 1 shown, the link failure protection method for a hybrid SDN network provided by an embodiment of the present invention includes the following steps:

[0079] S101, according to the objectives and constraint conditions of the actual service requirements of users, establish an optimization model for the link failure protection problem of the hybrid SDN network;

[0080] S102, propose the FP-SCS method for the SCS problem involved, which aims to minimize the number of SDN switches required to protect all links;

[0081] S103, for the PPS problem, utilize the flexibility of SDN switch routing and propose the FP-PPS algorithm with the help of the tunnel mechanism based on the destination node to select a suitable protection path and finally achieve the failure protection of the hybrid SDN network.

[0082] As a preferred embodiment, as Figure 2 shown, the link failure protection method for a hybrid SDN network provided by an embodiment of the present invention specifically includes the following steps:

[0083] 1. Model construction

[0084] The network topology can be represented by a directed graph G = (V, E), where V is the set of nodes and E is the set of links. The bidirectional link between node i and node j includes two directed links, namely e1 = <i, j> and e2 = <j, i>. In addition, it is assumed that the network topology is a 2-edge connected graph to ensure the existence of protection paths. When link e fails, the path that cannot transmit data packets is called the affected path and needs to be replaced by a backup protection path that does not pass through e.

[0085] Table 1 Symbol Explanation

[0086]

[0087] According to the goals and constraints such as the actual service requirements of users, an optimization model (1)-(4) for link fault protection in a hybrid SDN network is established. The symbols involved in this model are shown in Table 1.

[0088] Minimize:

[0089]

[0090] Subject to:

[0091]

[0092]

[0093]

[0094] In addition, the actual specific requirements quantitatively represented by the constraint conditions of the model are as follows:

[0095] 1) As shown in formula (2), at least one designated SDN switch for configuring the protection path can be found in the end routers of the faulty link e for data packet transmission.

[0096] 2) As shown in formula (3), when link e fails, only one designated SDN switch is configured for its end routers.

[0097] 3) As shown in formula (4), the nodes configured as designated SDN switches must have been upgraded from traditional routers to SDN switches.

[0098] 2. Fault Protection Scheme for Hybrid SDN Network

[0099] The present invention details the proposed fault protection scheme for a hybrid SDN network, including the FP-SCS method proposed for the SCS problem and the FP-PPS algorithm proposed for the PPS problem.

[0100] 2.1 SDN Candidate Selection Method

[0101] Considering the deployment overhead of evolving to a fully SDN network and the impact of device downtime on network user services, the SCS problem in fault protection aims to minimize the number of SDN switches required to protect all links. The framework of the FP-SCS method provided by the embodiments of the present invention is as Figure 2 shown.

[0102] First, construct the SDN candidate table tSDN, whose purpose is to identify the locations of suitable candidate SDN switches for each link e. If node δ τ can protect link e, then mark tSDN(e,τ) as "1"; otherwise, mark it as "0". Based on tSDN, first select some important nodes and store them in sSDN. These nodes are responsible for protecting the links with the highest protection difficulty and protected by only one node. Then delete the links protected by sSDN from tSDN to update the candidate table. Further, reduce the candidate node set to be pre-upgraded to SDN switches to csSDN according to the link with the highest protection difficulty among the unprotected links. Then, FP-SCS selects the node with the maximum protection ability for the unprotected links in csSDN and adds it to sSDN by means of the idea of the greedy strategy. Until the selected node sSDN can protect all links, which is the final node set that needs to be upgraded to SDN switches, the FP-SCS method ends.

[0103] 2.2 Protection Path Selection Algorithm

[0104] After the FP-SCS method determines the minimum required number of SDN switches, it is necessary to further select appropriate protection paths by using the routing flexibility of these switches to finally achieve fault protection for the hybrid SDN network. Specifically, for the PPS problem, the FP-PPS algorithm is proposed by using the flexible routing ability of SDN switches and relying on the tunnel mechanism based on the destination node. It fully considers the mutually influential path parameters such as PPL and MLU, and aims to flexibly select the most suitable protection path according to the importance of these parameters. The detailed description of the FP-PPS algorithm is shown in Table 2.

[0105] Table 2 Detailed Description of the FP-PPS Algorithm

[0106]

[0107] · Based on the traffic matrix T N , after the traffic is normally transmitted on the unaffected paths in the 3rd row, update the traffic data of each link in T L .

[0108] · Considering that different protection path selection orders for multiple affected paths may affect the overall traffic distribution of the network after protection, the 4th row sorts the affected paths, and then selects appropriate protection paths for these paths in turn.

[0109] · Lines 7 - 8 construct all available protection paths for the affected paths and set path scores to evaluate the performance of each path, which is calculated according to formula (5).

[0110]

[0111] Among them, α and β are variable weights set according to the actual needs of users for PPL and MLU (α + β = 1), which lays a foundation for fully considering the importance of the path parameters PPL and MLU respectively. In addition, and is the length of the protection path from node s to d and the normalized result of the maximum link utilization of the path to unify all path parameters to the same order of magnitude.

[0112] According to the attributes of the path parameters, finally, the protection path with the smallest path score is selected in line 10. If there is a tie and α is larger, it means that PPL dominates the user's needs at this time, then the protection path with a smaller PPL is selected. If β is larger, then the protection path with a smaller MLU is selected.

[0113] The link fault protection system of the hybrid SDN network provided by the embodiment of the present invention includes:

[0114] An optimization model construction module, configured to establish an optimization model for link fault protection of a hybrid SDN network according to the goals and constraints of actual service requirements;

[0115] An SDN candidate selection module, configured to use the FP - SCS method to solve the SDN candidate selection problem to minimize the number of SDN switches required to protect all links;

[0116] A protection path selection module, configured to select a suitable protection path by means of the routing flexibility of the SDN switch and the tunnel mechanism based on the destination node, and use the FP - PPS algorithm to achieve fault protection for the hybrid SDN network.

[0117] To prove the creativity and technical value of the technical solution of the present invention, this part is an application embodiment of the technical solution of the claims on a specific product or related technology.

[0118] The future new network is effective in solving problems such as the privatization and rigidity of traditional networks. Protecting the link failures of the hybrid SDN network belonging to the future new network can ensure the normal operation of the network, improve the robustness and elasticity of the network. The FP-SCS method for SDN candidate selection based on an optimized greedy strategy and the protection path selection FP-PPS algorithm that utilizes the flexible routing ability of SDN switches and considers various protection path parameters can be applied to the complete hybrid SDN network link failure protection process and have been widely applied in fields such as network optimization, electronic information, and industrial detection.

[0119] The technologies adopted in the present invention have applications in solving optimization problems, complex multi-objective programming problems, etc., mainly including applications in robotics such as mobile robot path planning and robot structure planning, control aspects such as pipeline routing control, planning problems such as production planning and task allocation, network optimization aspects such as communication network design, combinatorial optimization aspects such as knapsack problems and graph partitioning problems, and aspects such as image processing and signal processing.

[0120] Performance evaluation: In order to evaluate the performance of the FP-SCS and FP-PPS algorithms provided by the embodiments of the present invention, a large number of simulation experiments were carried out under various network topologies and network parameters.

[0121] 1. Simulation settings

[0122] For the SCS problem in hybrid SDN network fault protection, the FP-SCS method provided by the present invention was compared with the existing Greedy, SCS-ST, and Search-tree methods in terms of the number of required SDN switches. In addition, the FP-PPS algorithm of the present invention was compared with the existing Greedy-LB, RPR-DT, RPR-LB, and MASSS algorithms in terms of the path parameter performance involved in the PPS problem.

[0123] The effectiveness of the solution of the present invention was verified using Python simulation software on a Windows 7 operating system with a 3.60GHz Intel Core i7 CPU. The real network topologies adopted include the Internet2 network (I2), Abilene network (Ab), US network, Germany network (GM), scale-free network with 30 nodes (SF(30)), and ER random network with 50 nodes and a node connection probability of 0.1 (ER(0.1)).

[0124] In order to verify the universality of the algorithms provided by the embodiments of the present invention for different network traffic distributions, random traffic matrices were used in the simulation experiments. In this matrix, when nodes i≠j, the traffic from node i to j follows a uniform distribution in the range of [1, 100] Mbytes / s; otherwise, the traffic is 0.

[0125] 2. Number of SDN switches

[0126] Regarding the number of SDN switches required for fault protection, the performance of the FP-SCS method provided in the embodiments of the present invention is deeply analyzed under different topology types and network scales.

[0127] 2.1 Network topology adaptability

[0128] The number of SDN switches required to implement SCS on different network topologies and the running time are shown in Tables 3 and 4. Among these four algorithms, the FP-SCS method can select the fewest SDN switches for all network topologies, and it only takes 0.0028 s to 0.1045 s more than the Greedy method. However, it shortens 0.0047 s to 0.3842 s and 0.0323 s to 0.476 s compared with the SCS-ST and Search-tree methods respectively.

[0129] Table 3 Number of SDN switches required for SDN candidate selection

[0130]

[0131] Table 4 Running time (s) of SDN candidate selection

[0132]

[0133] 2.2 Influence of network scale

[0134] The influence of network scale on the number of SDN switches required is as Figure 3 shown. As the network scale expands, the required SDN switches also gradually increase. At the same time, on average, FP-SCS requires 9.62%, 9.46%, and 1.19% fewer SDN switches than Greedy, SCS-ST, and Search-tree respectively, and it can also achieve the protection of all links.

[0135] 3. Average protection path length and maximum link utilization rate

[0136] The present invention explores the applicability of the FP-PPS algorithm to obtain protection paths, and mainly evaluates it using two indicators: the average protection path length (APPL) and the average maximum link utilization rate (AMLU). The specific evaluation methods are shown in Formulas (6) and (7).

[0137]

[0138]

[0139] Among them, ε eis the number of paths affected by the failed link e. is the length of the protection path from node s to d, MLU e is the maximum link utilization of the network after protecting the failed link e.

[0140] The comparison results between the algorithm provided by the embodiment of the present invention and the existing algorithms are as Figures 4A to 4B shown. If PPL is used as the only evaluation basis for protection path selection, the APPL of FP-PPS (α = 1, β = 0) is similar to that of RPR-DT, but the AMLU of FP-PPS is 0.34% - 6.38% lower than that of RPR-DT. From the perspective of the MLU of the protection path, the AMLU of FP-PPS (α = 0, β = 1) is slightly lower than that of Greedy-LB and RPR-LB by 0.13% - 4.83% and 0.11% - 7.74% respectively. In addition, compared with these two algorithms, the APPL of FP-PPS provided by the embodiment of the present invention is significantly reduced by 1.33% - 24.4% and 0.14% - 6.13% respectively.

[0141] If both PPL and MLU are considered, taking the example that these two parameters have the same importance (α = 0.5, β = 0.5), it is compared with the MASSS algorithm that also considers these two parameters at the same time. MASSS uses a link-based tunnel mechanism to construct candidate protection paths, which is independent of the affected destination nodes, and may lead to a limited selection range of candidate protection paths. Different from MASSS, the FP-PPS provided by the embodiment of the present invention uses a destination-node-based tunnel mechanism, which can dynamically adjust the weights of path parameters to prompt users to select more appropriate protection paths. Therefore, compared with MASSS, FP-PPS can reduce the APPL by 9.34% - 39.78% and at the same time reduce the AMLU by 2.72% - 13.03%.

[0142] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated designed hardware. Those of ordinary skill in the art can understand the above-mentioned devices and methods, which can be implemented using computer-executable instructions and / or included in processor control code. For example, such code is provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuits of programmable hardware devices such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, or can be implemented by a combination of the above hardware circuits and software, such as firmware.

[0143] As described above, the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention should be covered by the protection scope of the present invention.

Claims

1. A link failure protection method for a hybrid SDN network, characterized in that, The link fault protection method for a hybrid SDN network includes the following steps: Step 1: Establish an optimization model for the link fault protection problem of the hybrid SDN network according to the goals and constraints of the actual service requirements of users; Step 2: Propose the FP-SCS method for the SCS problem involved, which aims to minimize the number of SDN switches required to protect all links; Step 3: For the PPS problem, utilize the flexibility of SDN switch routing and propose the FP-PPS algorithm with the help of the destination node-based tunneling mechanism to select appropriate protection paths and ultimately achieve fault protection for the hybrid SDN network; The establishment of the optimization model in Step 1 includes: Use the directed graph G = (V, E) to represent the network topology, where V is the set of nodes and E is the set of links; the bidirectional link between node i and node j includes two directed links, e1 = <i, j> and e2 = <j, i>; for the network topology of a 2-edge connected graph that can ensure the existence of protection paths, when link e fails, the path that cannot transmit data packets is called the affected path, and it is replaced by a backup protection path that does not pass through link e; Establish an optimization model for the link fault protection problem of the hybrid SDN network according to the goals and constraints of the actual service requirements of users; among them, the optimization model is shown as follows: Minimize: Take minimizing the number of SDN switches required for fault protection of all links in the hybrid SDN network as the objective function; Subject to: Its constraint is that at least one designated SDN switch can be found for the end routers of the faulty link e to transmit data packets; It ensures that only one designated SDN switch is configured for the end router when link e fails; At the same time, the node configured as the designated SDN switch must have been upgraded from a traditional router to an SDN switch; Wherein, is a binary symbol. When node x is the end router of link e otherwise is a binary symbol. When the path from node s to d passes through link e otherwise is a binary symbol. When node k is selected as the designated SDN switch of node x when link e fails otherwise is a binary symbol. When node m is a one-hop neighbor of node k otherwise u k is a binary symbol. When node k is selected to be upgraded to an SDN switch, u k = 1, otherwise u k = 0; The SDN candidate selection method in Step 2 includes: Based on the deployment cost of evolving to a fully SDN network and the impact of device downtime on network user services, the SCS problem in hybrid SDN network fault protection aims to minimize the number of SDN switches required to protect all links, and the FP-SCS method is proposed; The FP-SCS method specifically includes: constructing the SDN candidate table tSDN to identify the locations of suitable candidate SDN switches for each link e; if the node δ τ protects the link e, then record tSDN(e,τ) as 1, otherwise record it as 0; based on tSDN, select important nodes and store them in sSDN, which is responsible for protecting the link with the highest protection difficulty and only protected by one node; delete the links already protected by sSDN from tSDN to update the candidate table; reduce the candidate node set to be pre-upgraded to SDN switches to csSDN according to the link with the highest protection difficulty among the unprotected links; by means of the idea of the greedy strategy, select the node in csSDN with the greatest protection ability for the unprotected links and add it to sSDN; until the selected node sSDN has the ability to protect all links, which is the final node set to be upgraded to SDN switches, and the FP-SCS method ends; The protection path selection algorithm in Step 3 includes: After determining the minimum number of SDN switches using the FP-SCS method, use the flexible routing of SDN switches to select appropriate protection paths to ultimately achieve fault protection for the hybrid SDN network; for the PPS problem, propose the FP-PPS algorithm with the help of the destination node-based tunneling mechanism, analyze the path parameters of the mutually influencing protection path length (PPL) and maximum link utilization (MLU), and select the most appropriate protection path according to the importance of each parameter; The FP-PPS algorithm specifically includes: based on the traffic matrix T between network nodes N , after the traffic is normally transmitted on the unaffected paths, update the traffic data of each link in T L ; based on the influence of different protection path selection orders for multiple affected paths on the overall traffic distribution of the protected network, sort the affected paths in descending order according to the required tunnel traffic size on the affected paths, and then select appropriate protection paths for them in turn; use the tunnel mechanism based on the destination node to construct all available protection paths for the affected paths, and set path scores to evaluate the performance of each path; according to the attribute that the smaller the path parameters PPL and MLU, the better the path performance, select the protection path with the smallest path score; if there is a tie and α is large, it means that PPL dominates the user requirements, then select the protection path with a smaller PPL; otherwise, select the protection path with a smaller MLU; Calculate the path score of the protection path from node s to d when link e fails according to the following formula to evaluate the path performance: Wherein, α and β are variable weights set according to the actual requirements of the user for PPL and MLU, and α + β = 1; and is the length of the protection path from node s to d and the maximum link utilization rate of the path The normalization result of is used to unify all path parameters to the same order of magnitude.

2. A link failure protection system for a hybrid SDN network applying the link failure protection method of the hybrid SDN network as described in claim 1, characterized in that, The link fault protection system for a hybrid SDN network includes: An optimization model construction module for establishing an optimization model for link fault protection of a hybrid SDN network according to the goals and constraints of actual service requirements; An SDN candidate selection module for using the FP-SCS method to solve the SDN candidate selection problem to minimize the number of SDN switches required to protect all links; The protection path selection module is used to select a suitable protection path by means of the routing flexibility of the SDN switch and the tunnel mechanism based on the destination node, and utilize the FP-PPS algorithm to achieve fault protection for the hybrid SDN network.

3. A computer device, characterized in that, The computer device includes a memory and a processor. When the computer program stored in the memory is executed by the processor, the processor is caused to execute the steps of the link fault protection method for the hybrid SDN network as described in claim 1.

4. A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the processor is caused to execute the steps of the link fault protection method for the hybrid SDN network as described in claim 1.

5. An information data processing terminal, characterized in that, The information data processing terminal is used to implement the link fault protection system for the hybrid SDN network as described in claim 2.

Images