A method for prioritizing SRv6 node-based hybrid network backup routing

Abstract

This invention discloses a method for prioritizing backup routes in a hybrid network based on SRv6 nodes. The method includes: obtaining one primary path and K backup paths for each service; combining the primary path with each backup path to obtain K primary / backup route combinations; performing segment processing on each primary / backup route combination; determining the objective function for each primary / backup route combination after segment processing; solving for the Pareto solution set based on the objective function; and obtaining the primary path and backup path for each service based on the Pareto solution set. This invention solves the problem of low service stability caused by link failures in existing networks.

Description

Technical Field

[0001] This invention relates to the field of computer networks, and more particularly to a method for prioritizing the selection of backup routes in a hybrid network based on SRv6 nodes. Background Technology

[0002] With the continuous development of the Internet, networks not only need to provide a wider range of services but also more diverse and customized applications. For operators, this requires continuous adjustments to network architecture to meet the demands of the new environment, while the underlying network needs to provide services to upper-layer applications in a more flexible manner. Segment routing (SR) is an emerging source routing architecture. Segment routing technology inserts a segment list into packets, specifying the forwarding path of the packet at the source node. Other nodes do not need to maintain any information about the forwarding path; they only need to forward the data according to the specified path after receiving the packet. Currently, segment routing architectures can be deployed on both Multi-Protocol Label Switching (MPLS) and Internet Protocol Version 6 (IPv6) data planes. SR based on the MPLS forwarding plane is called SR-MPLS, and SR based on the IPv6 forwarding plane is called SRv6.

[0003] Numerous papers and patents have demonstrated the application potential of SR (Surveyor Linear Technology). However, for Internet Service Providers (ISPs), migrating from traditional IP networks to a complete SR network is not an easy task, presenting numerous economic and technical challenges. Firstly, the routers currently operating in the network are not designed to support SR technology, necessitating upgrades to existing physical equipment, which require substantial financial and technical support. Even though some advanced routers can be upgraded to run SR, large-scale upgrades to network equipment may lead to network service instability. Therefore, upgrading only a subset of routers may be a feasible solution for ISPs initially, and a hybrid SR network, partially deploying SR nodes within the IP network, represents a possible transitional network scenario.

[0004] Improving the availability of network services is one of the goals that network service providers have always pursued. The ability to detect and respond to failures in physical devices within the network, ensuring uninterrupted service, is an important indicator for evaluating network availability.

[0005] To understand the current state of technological development, existing papers and patents were searched, compared, and analyzed, and the following technical information with high relevance to this invention was selected:

[0006] Paper 1: Published in January 2022 in the journal "Computer Communications" by Anbiah, Anix, et al., "Anbiah A, Sivalingam KM. Efficient failure recovery techniques for segment-routed networks[J]. Computer Communications, 2022(Jan.):182.DOI:10.1016 / j.comcom.2021.10.033.", this paper proposes a segment-level failure recovery technique. It uses the Bandar algorithm to calculate two non-intersecting paths and segments these paths into network segments. Protection is applied to a network segment of the end-to-end path, rather than a single link or node. Simulation results show that this scheme outperforms previous algorithms in terms of recovery capability from multiple failures and the resources required for recovery.

[0007] Paper 2: The paper "Li X, Yeung KL. Fast Reroute in Hybrid Segment Routing Network[C]. 2020 IEEE 17th Annual Consumer Communications & Networking Conference (CCNC). IEEE, 2020. DOI: 10.1109 / CCNC46108.2020.9045129" presented by Li, XQ et al. at the 17th CCNC conference in 2020 considers a hybrid segment routing (SR) network composed of IP routers and SR-MPLS routers. It focuses on utilizing the source routing capabilities of SRs and proposes two integer linear programming (ILP) formulas to find the optimal repair path for rerouted packets in order to maximize the percentage of links protected by backup routes. The two ILP formulas aim to minimize the repair path length and balance traffic load, respectively.

[0008] Among existing backup routing planning algorithms for SR networks, Paper 1 is based on the premise of a pure SR device network. If the backup routing algorithm under a pure SR network is applied to an SRv6+IPv6 hybrid network, some nodes lack SR capabilities and cannot complete the re-forwarding of packets after a failure. Furthermore, since the non-shortest path is often used when calculating the forwarding path of packets, routing loops may occur when a failure occurs. Paper 2 addresses the scenario of a hybrid segment routing (SR) network composed of IP routers and SR-MPLS routers. In the SR-MPLS hybrid network scenario, the backup routing algorithm requires SR nodes to form SR domains and connected subgraphs, which is not perfectly suitable for the distributed deployment characteristics of SRv6 nodes. Summary of the Invention

[0009] This invention provides a method for prioritizing the selection of backup routes in a hybrid network based on SRv6 nodes, which solves the problem of low service stability caused by link failures in existing networks.

[0010] A method for prioritizing SRv6 nodes in a hybrid network backup route selection, wherein the hybrid network is a network including SRv6 nodes and IPv6 nodes, comprising:

[0011] Obtain one primary path and K backup paths for each service;

[0012] The primary path is combined with each backup path to obtain K primary and backup route combinations;

[0013] Segment processing is performed separately for each primary / backup route combination;

[0014] Determine the objective function for each primary / backup route combination after segment processing;

[0015] The Pareto solution set is obtained by solving the objective function, and the main path and backup path of each service are obtained from the Pareto solution set.

[0016] In one embodiment of the present invention, the primary and backup routing combinations are segmented, specifically including: in each of the K primary and backup routing combinations for each service, the primary and backup paths are divided into segment paths using an improved Segment Based Protection (SBP) algorithm, and non-SR nodes are filtered out when selecting the head and tail nodes of the segment paths.

[0017] In one embodiment of the present invention, the step of dividing the primary and backup paths into segmented paths using the improved SBP algorithm in the K primary and backup route combinations for each service specifically includes: determining the shared nodes between the primary and backup nodes; filtering the non-segmented routing nodes among the shared nodes; sorting the filtered segmented routing nodes in descending order of betweenness centrality, and then truncating a corresponding number of nodes according to the segment label depth; and creating a segment between each truncated shared node starting from the source node.

[0018] In one embodiment of the present invention, the objective function includes: maximum link utilization, link revenue, and network reliability.

[0019] In one embodiment of the present invention, determining the objective function for each primary / backup route combination after segment processing specifically includes: calculating the maximum link utilization U of all links in the network according to the following formula. max :

[0020]

[0021]

[0022]

[0023] x e (t)=PP e (t)*r(t)+BP e (t)*r(t)

[0024] In the formula, u e This represents the utilization rate of each link, x. e c represents the total traffic volume on the edges of link set E. e The maximum capacity of link e is defined as the link e being an element in the link set E; x e (t) represents the traffic volume traversed by service t on edge E, PP e (t) indicates whether link set E is on the main path of service t, BP e (t) indicates whether edge E is on the backup path of service t.

[0025] In one embodiment of the present invention, determining the objective function for each primary / backup route combination after segment processing specifically includes: calculating the maximum link revenue β of all links in the network according to the following formula. p(t) :

[0026]

[0027] In the formula, U represents the benefit of link e. eSR represents the utilization rate of each link, where ω1 and ω2 are the link revenue weight and node revenue weight, respectively. v This indicates whether node v is an SR node, and p(t) represents the path of service t.

[0028] In one embodiment of the present invention, the objective function for determining each primary / backup route combination after segment processing specifically includes:

[0029] Calculate the network reliability α of all links in the network using the following formula:

[0030]

[0031]

[0032]

[0033]

[0034]

[0035]

[0036] In the formula, NF represents the number of faults in the network, |E| represents the number of edges in the network links, and f e The SLD represents the average failure rate of each link; SLD represents the segment label depth, which is a constant, h p (t) represents the number of hops on the main path, h b (t) represents the number of hops in the backup path; This represents the probability of a failure in a segment of the main path. f represents the probability of a segment failure in the backup path; p (t) represents the main link failure rate, f b f(t) represents the backup link failure rate; f(t) represents the probability of both primary and backup failures; w(t) represents the weight of each service t.

[0037] An apparatus for prioritizing SRv6 nodes to select backup routes in a hybrid network, wherein the hybrid network includes SRv6 nodes and IPv6 nodes, comprising:

[0038] The primary and backup route combination acquisition module is used to acquire one primary path and K backup paths for each service; and to combine the primary path with each backup path to obtain K primary and backup route combinations.

[0039] The segment processing module is used to perform segment processing on the primary and backup routing combination;

[0040] The objective function generation module is used to determine the objective function for each primary / backup route combination after segment processing;

[0041] The solution module is used to solve for the Pareto solution set based on the objective function, and to obtain the main path and backup path of each service based on the Pareto solution set.

[0042] A device for prioritizing SRv6 node selection of hybrid network backup routes includes:

[0043] At least one processor; and,

[0044] The memory is communicatively connected to the at least one processor via a bus; wherein,

[0045] The memory stores instructions that can be executed by the at least one processor to implement the method as described in any of the above embodiments.

[0046] A non-volatile storage medium storing computer-executable instructions, which are executed by a processor to implement the method as described in any of the above embodiments.

[0047] This invention provides a method for prioritizing backup routes in a hybrid network based on SRv6 nodes, offering at least the following advantages: Simulation experiments were conducted using Java, comparing the proposed algorithms to a segment-level routing recovery algorithm (SBP) for pure SR networks ("Efficient failure recovery techniques for segment-routed networks") and a fast rerouting algorithm (FRR) for SR-MPLS+IPv4 networks ("Fast Reroute in Hybrid Segment Routing Network"). Simulation results show that when network failures occur, the proposed routing scheme has a smaller impact on services and better protects services from failures; while ensuring service stability, it can balance the load between links and avoid link congestion. Attached Figure Description

[0048] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings:

[0049] Figure 1 A diagram illustrating the encapsulation format of the SRv6 extension head provided in an embodiment of the present invention;

[0050] Figure 2 A flowchart of the backup route selection algorithm based on SRv6 node priority provided in an embodiment of the present invention;

[0051] Figure 3 This invention provides a schematic diagram illustrating the steps of a method for prioritizing the selection of backup routes in a hybrid network based on SRv6 nodes, as provided in an embodiment of the invention.

[0052] Figure 4 This is a schematic diagram illustrating the influence of parameter K on the FDR results provided in this embodiment of the invention;

[0053] Figure 5 This is a schematic diagram illustrating the influence of parameter SLD on FDR results provided in an embodiment of the present invention.

[0054] Figure 6 Comparative experimental diagrams of FDR in KDL topology provided in embodiments of the present invention;

[0055] Figure 7 Comparison test diagram of FDR in Cogentco topology provided in the embodiments of the present invention;

[0056] Figure 8 Comparison test diagram of FDR in RingOfRing topology provided in the embodiments of the present invention;

[0057] Figure 9 A comparative experimental diagram of FDR in the BCube topology provided in the embodiments of the present invention;

[0058] Figure 10 A comparative experimental diagram of Umax in the KDL topology provided in the embodiments of the present invention;

[0059] Figure 11 A comparative experimental diagram of Umax in the Cogentco topology provided in the embodiments of the present invention;

[0060] Figure 12 A comparative experimental diagram of Umax in the RingOfRing topology provided in the embodiments of the present invention;

[0061] Figure 13 This is a comparative experimental diagram of Umax in the Bcube topology provided in the embodiments of the present invention;

[0062] Figure 14 A schematic diagram of an apparatus for prioritizing the selection of hybrid network backup routes based on SRv6 nodes is provided in an embodiment of the present invention;

[0063] Figure 15 This is a schematic diagram of a device for prioritizing the selection of hybrid network backup routes based on SRv6 nodes, provided as an embodiment of the present invention. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described clearly and completely below in conjunction with specific embodiments. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0065] It should be noted that those skilled in the art will understand, explicitly and implicitly, that the embodiments described in this invention can be combined with other embodiments without conflict. Unless otherwise defined, the technical or scientific terms used in this invention should be understood in their ordinary sense by those skilled in the art. The terms "a," "an," "an," "the," etc., used in this invention do not indicate quantity limitation and can represent singular or plural. The terms "comprising," "including," "having," and any variations thereof used in this invention are intended to cover non-exclusive inclusion; the terms "first," "second," "third," etc., used in this invention are merely to distinguish similar objects and do not represent a specific ordering of objects.

[0066] A network that combines SRv6 nodes and regular IPv6 nodes is a network architecture that inherits both SRv6 technology and the IPv6 protocol standard. The main characteristics of this hybrid network architecture are as follows:

[0067] First, the hybrid network architecture offers high flexibility and resilience. SRv6 networks can achieve granular management and control through technologies and policies on the control plane, while supporting advanced routing functions such as multi-path forwarding and extended forwarding tables to help optimize network performance and resource utilization.

[0068] Secondly, the hybrid deployment network architecture is well-suited for large-scale distributed applications. SRv6 networks not only support service requirements such as network segment isolation and Quality of Service (QoS) guarantees, but also offer excellent scalability and upgradeability.

[0069] In addition, compared with SR-MPLS hybrid deployment, SRv6 networks have the following significant advantages: First, compared with SR-MPLS, SRv6 uses IPv6 addresses as path identifiers, which not only provides a larger routing space, but also enables more fine-grained path control; Second, compared with SR-MPLS fragment isolation technology, SRv6 network fragment technology is more flexible and customizable, supporting the provision of more network assurance and service functions.

[0070] The biggest difference between SRv6 nodes and SR-MPLS nodes during node deployment is that SRv6 nodes do not need to form a connected subgraph. In SRv6, packet paths are driven by adding SRv6 headers to endpoints instead of using label switching. Therefore, SRv6 nodes can flexibly and efficiently control packet forwarding paths, while ordinary IPv6 nodes can perform routing and addressing through standard IPv6 protocols. This hybrid deployment network architecture does not require SRv6 nodes to form a connected subgraph; SRv6 nodes and ordinary IPv6 nodes can be interconnected through standard IPv6 routing. The encapsulation format of the SRv6 extension header is as follows: Figure 1 As shown.

[0071] In hybrid SRv6 and IPv6 networks, to better utilize the source routing capabilities of SR nodes, this invention proposes a method for selecting backup routes in the hybrid network based on SRv6 nodes selected-first (SNSF-BR). Figure 2 The diagram shows the overall flowchart of the SNSF-BR algorithm. The algorithm's inputs include the network topology G and the traffic matrix T, while its outputs include the primary path and backup path for each service. A detailed explanation follows.

[0072] The relevant symbols involved in the algorithm flow are shown in Table 1 below:

[0073]

[0074]

[0075] Table 1

[0076] The objective function and the meanings of the various symbols mentioned below are shown in Table 2:

[0077]

[0078]

[0079] Table 2

[0080] Figure 3 A schematic diagram illustrating the steps of a method for prioritizing the selection of hybrid network backup routes based on SRv6 nodes, provided in an embodiment of the present invention, may include the following steps:

[0081] S310: Obtain one primary path and K backup paths for each service; combine the primary path with each backup path to obtain K primary / backup route combinations.

[0082] Specifically, Dijkstra's algorithm is used to calculate a primary path for each service, and then Yen_K algorithm is used to calculate K backup paths for each service. The primary path can be combined with each backup path to form a primary-backup route combination, and each service has a total of K primary-backup route combinations.

[0083] S320: Perform segment processing for each primary / backup route combination.

[0084] In one embodiment of the present invention, segment processing is performed on the primary and backup route combination, specifically including: in each of the K primary and backup route combinations for each service, the primary and backup paths are cut into segment paths using the improved SBP algorithm, and non-SR nodes are filtered out when selecting the head and tail nodes of the segment paths.

[0085] Specifically, the modified SBP algorithm is used to segment the primary and backup paths in the K primary and backup route combinations for each service, while non-SR nodes need to be filtered out when selecting the head and tail nodes of the segments.

[0086] In one embodiment of the present invention, in each of the K primary and backup routing combinations for each service, the primary and backup paths are divided into segmented paths using an improved SBP algorithm. Specifically, this includes: determining the shared nodes between the primary and backup nodes; filtering the non-SR nodes among the shared nodes; sorting the filtered segmented routing nodes in descending order of betweenness centrality, and then truncating a corresponding number of nodes according to the segment label depth; and creating a segment between each truncated shared node starting from the source node.

[0087] Specifically, the process of improving the SBP algorithm is as follows:

[0088] Input: main path, edge-disjoint backup path, and segment label depth (SLD).

[0089] Output: Segment sequence.

[0090] 1. Identify the shared nodes between the primary and backup nodes; these nodes are potential segment endpoints. These shared nodes are denoted as `endPoints`.

[0091] 2: Determine if the shared nodes contain non-SR nodes. If yes, delete and filter out the non-SR nodes in the shared nodes; otherwise, proceed to the next step.

[0092] 3: If there are multiple shared nodes, determine whether the number of shared nodes is greater than SLD. If so, sort them in descending order by betweenness centrality and then extract SLD nodes. If not, proceed to the next step.

[0093] 4: Create a segment for each endPoint, starting from the source node:

[0094]

[0095] S330: Determine the objective function for each primary / backup route combination after segment processing.

[0096] In one embodiment of the present invention, the objective function includes: maximum link utilization, link revenue, and network reliability.

[0097] Specifically, the objective function for each primary / backup route combination is calculated as: maximum link utilization U. max Link revenue β p(t) Network reliability α 。 The objective function for optimizing the SNSF-BR backup route planning algorithm is as follows:

[0098] In one embodiment of the present invention, determining the objective function for each primary / backup routing combination after segment processing specifically includes: calculating the maximum link utilization U of all links in the network according to the following formula. max :

[0099]

[0100]

[0101]

[0102] x e (t)=PP e (t)*r(t)+BP e (t)*r(t)

[0103] In the formula, u e This represents the utilization rate of each link, x. e c represents the total traffic volume on the edges of link set E. e The maximum capacity of link e is given by link e, where link e is an element in the link set E; x e (t) represents the traffic volume traversed by service t on edge E, PP e (t) indicates whether link set E is on the main path of service t, BP e (t) indicates whether edge E is on the backup path of service t.

[0104] Specifically, the formula for calculating the minimum maximum link utilization is as follows:

[0105] x e (t)=PP e (t)*r(t)+BP e (t)*r(t) (1-1)

[0106]

[0107]

[0108]

[0109] Equation (1-1) is used to calculate the traffic volume of service t on edge E; Equation (1-2) is used to calculate that the traffic on link e is equal to the sum of all traffic passing through that link; Equation (1-3) is used to express the calculation formula for the utilization rate of each link; Equation (1-4) is used to calculate the maximum link utilization rate of all links in the network. In order to ensure the load balance of each link in the network, the ultimate optimization goal is to minimize the maximum link utilization rate.

[0110] In one embodiment of the present invention, determining the objective function for each primary / backup routing combination after segment processing specifically includes: calculating the maximum link revenue β of all links in the network according to the following formula. p(t) :

[0111]

[0112] In the formula, U represents the benefit of link e. e SR represents the utilization rate of each link, where ω1 and ω2 are the link revenue weight and node revenue weight, respectively. v This indicates whether node v is an SR node, and p(t) represents the path of service t.

[0113] Specifically, the maximum link revenue is calculated using the following formula:

[0114]

[0115]

[0116] In an SRv6+IPv6 hybrid network, there are two different types of physical nodes: SRv6 nodes and traditional IP nodes. Because SRv6 technology has network programmability, it can promptly redirect affected packets to backup paths by adjusting SRH segment labels in response to network failures, achieving the goal of rapid backup path switching; this avoids the routing loops that may occur with traditional IP nodes when handling network failures. Therefore, links with more SRv6 nodes should be prioritized when planning paths, hence a link revenue target is defined; simultaneously, to avoid link congestion, link utilization is used as a penalty when calculating link revenue.

[0117] Equation (1-5) defines the formula for calculating edge revenue, and a non-linear penalty method is used here to avoid link congestion. Equation (1-6) calculates the combined node revenue and link revenue on a business path. The first term of the formula first normalizes the node revenue to avoid the influence of path length on the result. The second term of the formula calculates the link revenue and performs normalization.

[0118] In one embodiment of the present invention, determining the objective function for each primary / backup routing combination after segment processing specifically includes:

[0119] Calculate the network reliability α of all links in the network using the following formula:

[0120]

[0121]

[0122]

[0123]

[0124]

[0125]

[0126] In the formula, NF represents the number of faults in the network, |E| represents the number of edges in the network links, and f e The SLD represents the average failure rate of each link; SLD represents the segment label depth, which is a constant, h p (t) represents the number of hops on the main path, h b (t) represents the number of hops in the backup path; This represents the probability of a failure in a segment of the main path. f represents the probability of a segment failure in the backup path; p (t) represents the main link failure rate, f b f(t) represents the backup link failure rate; f(t) represents the probability of both primary and backup failures; w(t) represents the weight of each service t.

[0127] Specifically, the formula for maximizing network reliability is as follows:

[0128]

[0129]

[0130]

[0131]

[0132]

[0133]

[0134] In a network with fault handling capabilities, it is often necessary to perform fault analysis on each service and allocate backup resources for it. However, in actual systems, the failure probability of each service path is different, so it is necessary to consider the failure probability of different paths as an optimization objective.

[0135] In equation (1-9), NF represents the number of faults in the network, |E| represents the number of edges in the network links, and f e The average failure rate of each link was calculated, and it can be treated as a constant in actual calculations.

[0136] In equation (1-10), SLD represents the segment label depth, which is a constant, h p (t),h b (t) represent the number of hops in the primary and backup paths, respectively. The average number of hops on a segment of the primary and backup paths can be calculated using equation (1-10).

[0137] In equation (1-11) A formula for calculating the failure probability of a segment in the primary backup path is defined.

[0138] Equation (1-12)f p (t),f b (t) represents the calculation method of the failure rate of the primary and backup links respectively. Equation (1-13) calculates the probability of all failures of the primary and backup links.

[0139] Equation (1-14) calculates the probability that a certain service can operate stably, and the ultimate optimization goal is to obtain the highest possible network reliability.

[0140] S340: Solve the Pareto solution set according to the objective function, and obtain the main path and backup path of each service based on the Pareto solution set.

[0141] Specifically, the Pareto solution set is obtained using the multi-objective optimization algorithm NSGAⅡ, and the main path and backup path of each service are obtained based on the Pareto solution set.

[0142] This invention proposes a backup routing algorithm based on SRv6 node priority selection for SRv6+IPv6 hybrid networks: In the process of calculating backup routes for services in SRv6+IPv6 hybrid networks, by defining the benefits of link paths, the algorithm can prioritize paths containing more SR nodes when planning backup routes for services.

[0143] This invention considers fault probability based on segment-level recovery: Considering that the fault probability of each service path is different in actual systems, it is necessary to analyze the fault probability of different services. Therefore, fault probability based on segment level is considered, and a corresponding optimization objective function is defined so that the path with a lower fault probability can be selected when planning service paths.

[0144] The solution of this invention can balance the network load: when calculating the end-to-end service path, the Yen_K algorithm is introduced to calculate multiple different alternative paths, increasing path diversity, thereby reducing the maximum link utilization and reducing the probability of link congestion.

[0145] To evaluate the performance of the algorithm proposed in this invention, corresponding simulation experiments were conducted. The simulation conditions are as follows:

[0146] 1) Network topology:

[0147] KDL Topology: The actual network topology structure from the dataset "The Internet Topology Zoo" (Kentucky Datalink) consists of 754 nodes and 899 links.

[0148] Cogentco topology: The Cogent network topology is derived from the dataset "The Internet Topology Zoo", which contains 197 nodes and 245 links.

[0149] Ring-Of-Rings topology: 8 nodes per ring, 3 rings, forming a topology with a total of 456 nodes and 1040 links.

[0150] BCube topology: With parameter n=8, a network with a total of 754 nodes and 1536 links is formed.

[0151] 2) Evaluation indicators:

[0152] Flow Drop Rate (FDR): In a network, due to the failure of a link or node, services cannot be rerouted and data packets cannot reach their destination correctly. The specific meaning of FDR is the percentage of services affected by the failure.

[0153] Umax: Maximum Link Utilization (Umax) refers to the maximum percentage of data transmitted on a network's available bandwidth over a given period. It is a crucial indicator of network bottlenecks and performance, typically used to assess network transmission capacity, plan network bandwidth, and optimize network performance. A Umax of 100% indicates that the link is fully utilized; exceeding 100% can lead to network congestion and performance degradation. Therefore, monitoring and controlling Umax can effectively improve network stability and reliability.

[0154] Hardware environment:

[0155] The experimental host hardware configuration is an Intel Core i5-7300HQ CPU, 2.50GHz, 12GB RAM; the operating system is Windows 10 Professional; and the Java language version is 1.8.0_361.

[0156] First, we will introduce two important hyperparameters involved in this invention:

[0157] When calculating backup paths for a service, the Yen_K algorithm is used to calculate K alternative paths, where K represents the number of non-shortest paths to calculate. In segment routing, the number of segments in an end-to-end path is limited by the length of the segment list. This limitation is called the Segment Label Depth (SLD) of the SR network. In the scheme proposed in this invention, SLD plays a crucial role, as it relates to the segmentation of the end-to-end path of a flow. For example, if SLD is only 1, the entire end-to-end path must be treated as a single segment. In this case, any failure anywhere on the primary path requires an end-to-end backup path to function properly. However, if SLD is 3, the end-to-end path of a flow can be divided into 3 segments.

[0158] Figure 4 and Figure 5 The effects of the parameters K and SLD on the FDR results are shown separately. Experimental results indicate that:

[0159] exist Figure 4 The vertical axis represents the FDR (Functional Path Reduction), and the horizontal axis represents different K values. The effects of SLD values ​​of 2, 3, 4, 5, and 6 on the results were investigated. The results show that changes in the value of K have no significant impact on the FDR results. Considering that each increase of K by 1 requires running Dijkstra's algorithm once, increasing the running time, and to maintain path diversity, K=3 was subsequently chosen for subsequent experiments.

[0160] Figure 5Experimental results show that FDR tends to decrease as the SLD value increases, but this trend becomes less pronounced after SLD=3. This is also supported by formula (1-13), which shows that the probability of service failures gradually decreases as SLD increases. However, the additional overhead caused by increasing SLD needs to be considered. Since SRv6 technology defines an SRH field in IPv6 packets, pushing an explicit IPv6 address stack onto the SRH, node-to-node forwarding is achieved by intermediate nodes continuously updating the destination address and offset address stack. As SLD increases, the length of the SRH field also gradually increases, affecting the data transmission rate. In practical SRv6 applications, SRv6 routers (taking the Huawei NE8000 M1A router as an example) can support a maximum Segment List length of 10. If fully utilized, the size of this content alone accounts for approximately 10.6% of the entire data packet, introducing additional overhead to data transmission. Therefore, subsequent experiments used SLD=3.

[0161] This invention compares its algorithm with the segment-level route recovery (SBP) algorithm from the paper "Efficient failure recovery techniques for segment-routed networks" and the fast reroute (FRR) algorithm from the paper "Fast Reroute in Hybrid Segment Routing Network" in four network topologies: KDL, Cogentco, Ring-Of-Rings, and BCube. Experiments were conducted with different proportions of SR nodes, and the results are as follows. Figure 6 , Figure 7 , Figure 8 , Figure 9 As shown. Figure 6 The comparison experiment diagram is shown for KDL topology FDR; Figure 7 A comparative test diagram of FDR in Cogentco; Figure 8 A comparative experimental diagram of FDR in RingOfRing; Figure 9 This is a comparative test diagram of FDR in BCube.

[0162] Previous studies on hybrid SR and IP node networks have shown that when SR nodes account for 30%-50% of all nodes, the source routing capabilities of SR nodes can be fully utilized to orchestrate the network and achieve the same capabilities as a pure SR network.

[0163] The experimental results show that, in simulations of the KDL and Cogentco internet topologies, this invention achieves a lower FDR (Free Response Rate) under most conditions compared to the other two algorithms, thus better protecting services. Furthermore, the advantage of this algorithm in selecting preferred SR node paths gradually becomes apparent as the SR ratio decreases. Taking the SR ratios of 30%, 40%, and 50% in the KDL topology as examples, this invention reduces FDR by 34.98%, 22.18%, and 15.83% respectively compared to the SBP algorithm, and by 19.77%, 30.69%, and 31.32% respectively compared to the FRR algorithm.

[0164] In the Ring-Of-Rings and BCube network topologies, due to structural differences from the previous two topologies, most nodes act as edge nodes, with a smaller number of nodes serving as central nodes connecting these edge nodes. Therefore, when deploying SR nodes, these central nodes are deployed first, and service path planning also tends to prioritize these central nodes, thus sacrificing some service path diversity. Experimental results show that this invention does not offer a significant overall improvement over the SBP algorithm, but it is still better than the FRR algorithm in most cases.

[0165] Since the FRR algorithm is based on Point of Local Repair, it does not reserve resources specifically for services when planning paths. In contrast, both the SBP algorithm and the SNSF-BR proposed in this invention reserve resources in advance for services. Therefore, this invention only compares the Umax metric with the SBP algorithm. Figure 10 , Figure 11 , Figure 12 , Figure 13 The experimental results of this algorithm compared to the SBP algorithm are shown. Figure 10 The graph shows a comparative experiment of Umax in KDL; Figure 11 The graph shows a comparative experiment of Umax in Cogentco. Figure 12 A comparative experimental diagram of Umax in RingOfRing; Figure 13 This is a comparative experimental diagram of Umax in Bcube.

[0166] Experimental results show that the proportion of SR nodes has little impact on Umax. Regardless of the value of the SR nodes, the SNSF-BR algorithm outperforms the SBP algorithm. In the four topologies, the algorithm proposed in this invention reduces the Umax index by 3.08%, 4.89%, 3.73%, and 11.44%, respectively. The results indicate that the SNSF-BR algorithm can reduce the maximum link utilization and decrease the probability of link congestion.

[0167] The above describes a method for prioritizing SRv6 node selection in a hybrid network backup route according to an embodiment of the present invention. Based on the same inventive concept, the present invention also provides a corresponding apparatus for prioritizing SRv6 node selection in a hybrid network backup route, such as... Figure 14 As shown.

[0168] The primary and backup route combination acquisition module 1402 is used to acquire one primary path and K backup paths for each service; and to combine the primary path with each backup path to obtain K primary and backup route combinations.

[0169] Segment processing module 1404 is used to perform segment processing on the primary and backup route combination;

[0170] The objective function generation module 1406 is used to determine the objective function for each primary / backup route combination after segment processing.

[0171] The solver module 1408 is used to solve for the Pareto solution set based on the objective function, and obtain the main path and backup path of each service based on the Pareto solution set.

[0172] This invention also provides a corresponding device for prioritizing SRv6 node selection in a hybrid network backup routing environment, such as... Figure 15 As shown, it includes:

[0173] The system includes at least one processor 1502, a communication interface 1504, a memory 1506, and a communication bus 1508; wherein the processor 1502, the communication interface 1504, and the memory 1506 communicate with each other through the communication bus 1508; the processor 1502 can call logical instructions stored in the memory 1506 to cause at least one processor 1502 to execute the method of the above embodiment.

[0174] Based on the same idea, some embodiments of the present invention also provide media corresponding to the above methods.

[0175] Some embodiments of the present invention provide a storage medium storing computer-executable instructions, which are executed by a processor to implement the methods of the above embodiments.

[0176] The various embodiments in this invention are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device and medium embodiments are relatively simple in description because they are fundamentally similar to the method embodiments; relevant parts can be referred to the descriptions in the method embodiments.

[0177] The devices, media, and methods provided in the embodiments of the present invention are one-to-one correspondences. Therefore, the devices and media also have similar beneficial technical effects as their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media will not be repeated here.

[0178] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process method or product that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process method or product. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process method or product that includes that element.

[0179] The above are merely embodiments of the present invention and are not intended to limit the invention. Although the present invention has been described in detail above with general descriptions and specific embodiments, modifications or improvements can be made to it, which will be obvious to those skilled in the art. Therefore, all such modifications or improvements made without departing from the spirit of the present invention fall within the scope of protection claimed by the present invention.

Claims

1. A method for prioritizing the selection of backup routes in a hybrid network based on SRv6 nodes, characterized in that, The hybrid network is a network including SRv6 nodes and IPv6 nodes, including: Obtain one primary path and K backup paths for each service; The primary path is combined with each backup path to obtain K primary and backup route combinations; Segment processing is performed separately for each primary / backup route combination; Determine the objective function for each primary / backup route combination after segment processing; The Pareto solution set is obtained by solving the objective function, and the main path and backup path of each service are obtained from the Pareto solution set. The segment processing of the primary and backup route combination specifically includes: in each of the K primary and backup route combinations for each service, the primary and backup paths are divided into segment paths using the improved Segment Protection SBP algorithm, and non-SR nodes are filtered out when selecting the head and tail nodes of the segment path. In each of the K primary and backup route combinations for each service, the improved SBP algorithm is used to cut the primary and backup paths into segmented paths. Specifically, this includes: determining the shared nodes between the primary and backup nodes; filtering the non-segmented route SR nodes in the shared nodes; sorting the filtered segmented route nodes in descending order of betweenness centrality, and then truncating a corresponding number of nodes according to the segment label depth; and creating a segment between each truncated shared node starting from the source node.

2. The method according to claim 1, characterized in that, The objective function includes: Maximum link utilization, link revenue, and network reliability.

3. The method according to claim 2, characterized in that, The objective function for determining each primary / backup route combination after segment processing specifically includes: Calculate the maximum link utilization of all links in the network using the following formula. : ； ； ； ； In the formula, This indicates the utilization rate of each link. This represents the total traffic volume on the edges of link set E. The maximum capacity of link e is defined as the link e being an element in the link set E. This represents the amount of traffic that service t passes through on edge E. This indicates whether link set E is on the main path of service t. This indicates whether edge E is on the backup path of service t. This represents the size of the demand for business t.

4. The method according to claim 2, characterized in that, The objective function for determining each primary / backup route combination after segment processing specifically includes: Calculate the maximum link revenue for all links in the network using the following formula. : ； In the formula, This represents the benefit of link e. This indicates the utilization rate of each link. These are the node revenue weight and the link revenue weight, respectively. Indicates whether node v is an SR node. This represents the path of business t.

5. The method according to claim 2, characterized in that, The objective function for determining each primary / backup route combination after segment processing specifically includes: Calculate the network reliability of all links in the network using the following formula. : ； ； ； ； ； ； In the formula, NF represents the number of faults in the network, and |E| represents the number of edges in the network links. This represents the average failure rate of each link; SLD represents the segment label depth, which is a constant. This indicates the number of hops on the main path. Indicates the number of hops in the backup path; This represents the probability of a failure in a segment of the main path. This indicates the probability of failure for a segment of the backup path; Indicates the main link failure rate. Indicates the backup link failure rate; This indicates the probability of both primary and backup systems failing. This represents the weight of each business segment t.

6. A device for prioritizing the selection of backup routes in a hybrid network based on SRv6 nodes, characterized in that, The hybrid network is a network including SRv6 nodes and IPv6 nodes, including: The primary and backup route combination acquisition module is used to acquire one primary path and K backup paths for each service; and to combine the primary path with each backup path to obtain K primary and backup route combinations. The segment processing module is used to perform segment processing on the primary and backup route combinations. Specifically, it includes: in each of the K primary and backup route combinations for each service, the primary and backup paths are cut into segment paths using the improved Segment Protection SBP algorithm, and non-SR nodes are filtered out when selecting the head and tail nodes of the segment paths. The steps of cutting the segment path specifically include: determining the shared nodes between the primary and backup nodes; filtering the non-segmented routing SR nodes in the shared nodes; sorting the filtered segmented routing nodes in descending order of betweenness centrality, and then truncating a corresponding number of nodes according to the segment label depth; creating a segment between each truncated shared node starting from the source node. The objective function generation module is used to determine the objective function for each primary / backup route combination after segment processing; The solution module is used to solve for the Pareto solution set based on the objective function, and to obtain the main path and backup path of each service based on the Pareto solution set.

7. A device for prioritizing SRv6 node selection of hybrid network backup routes, characterized in that, include: At least one processor; as well as, The memory is communicatively connected to the at least one processor via a bus; wherein, The memory stores instructions executable by the at least one processor, which are executed to implement the method as described in any one of claims 1-5.

8. A non-volatile storage medium storing computer-executable instructions, characterized in that, The computer-executable instructions are executed by a processor to implement the method as described in any one of claims 1-5.

Images