Data transmission method and device based on VXLAN, storage medium and electronic equipment

Abstract

The invention discloses a VXLAN (Virtual Extensible Local Area Network)-based data transmission method and device, a storage medium and electronic equipment, and the method comprises the steps: after a plurality of physical links between two user terminals are virtualized into a virtual link through a VXLAN technology, the priorities of the plurality of physical links are dynamically adjusted through a dynamic routing protocol, identifying the plurality of physical links as equivalent paths; and based on the equivalent path, dispersing service traffic to the plurality of physical links for transmission through a load sharing algorithm, and monitoring the state of the physical link in real time, and switching the service flow to an available target physical link through the load sharing algorithm when detecting that any physical link has a fault. According to the scheme, the limitation that a traditional routing protocol only supports double-link loads is broken through, a plurality of physical links cooperatively work logically, the link resource utilization rate is improved, automatic switching during faults is achieved through dynamic priority adjustment, and the network reliability is enhanced.

Description

Technical Field

[0001] This application relates to the field of computer communications. Specifically, it relates to a VXLAN-based data transmission method, apparatus, storage medium, and electronic device. Background Art

[0002] In traditional network architectures, common dynamic routing protocols such as OSPF and BGP are mainly designed for small and medium-sized network scenarios. In the case of dual-link configurations, these protocols can better achieve link load sharing and redundancy protection, distribute traffic to two physical links, and switch to ensure continuous business operation when a link fails.

[0003] However, when there are three or more physical links in the network, the limitations of existing protocols become apparent. The underlying algorithms of these protocols usually only support a fixed number of transmission paths, generally defaulting to two, and the extra physical links cannot be regarded as effective transmission paths by dynamic routing protocols. This is because the protocols simplify the design of path calculation and state synchronization. For example, the SPF algorithm of OSPF tends to select the optimal path.

[0004] This results in a large amount of physical link resources being idle, unable to be allocated on demand or dynamically adjusted in weight. This not only reduces bandwidth utilization but may also cause congestion due to traffic concentration on a few links. In scenarios with dense links such as cloud computing and multi-data center interconnection, this problem is more prominent, so there is an urgent need for technological innovation to break the path capacity limitations of traditional routing protocols. Summary of the Invention

[0005] This application provides a VXLAN-based data transmission method, apparatus, storage medium, and electronic device, which can solve the problem of multiple physical links being idle due to only supporting dual-link load.

[0006] The specific technical solutions are as follows:

[0007] In a first aspect, an embodiment of this application provides a VXLAN-based data transmission method, and the method includes:

[0008] After virtualizing multiple physical links between two user terminals into a virtual link through Virtual eXtensible Local Area Network (VXLAN) technology, dynamically adjust the priorities of the multiple physical links through a dynamic routing protocol to identify the multiple physical links as equivalent paths;

[0009] Based on the equivalent paths, disperse service traffic to the multiple physical links for transmission through a load sharing algorithm;

[0010] Monitor the physical link status in real time. When any physical link fails, switch the service traffic to the available target physical link through the load balancing algorithm.

[0011] In a possible implementation, dispersing the service traffic to the multiple physical links for transmission through the load balancing algorithm includes:

[0012] Obtain the link weights assigned to each of the multiple physical links based on the weight allocation model;

[0013] Calculate the minimum integer ratio of all link weights among the multiple physical links, and calculate the sum of each value in the minimum integer ratio as the total weight;

[0014] According to the total weight and the minimum integer ratio, configure a corresponding weight interval for the link weight of each physical link, such that the length of the weight interval is equal to the corresponding integer in the minimum integer ratio;

[0015] Calculate the hash value of the target information in each packet in the service traffic respectively, and perform normalization processing on each hash value to obtain the normalized hash value;

[0016] Determine the weight interval where each normalized hash value is located, and allocate the corresponding packet to the physical link corresponding to the weight interval for transmission.

[0017] In a possible implementation, the method for allocating link weights to each of the multiple physical links based on the weight allocation model includes:

[0018] Obtain the current link status metrics of each of the multiple physical links and the service characteristics of the service traffic;

[0019] Input the current link status metrics of each physical link and the service characteristics of the service traffic into the weight allocation model to obtain the link weights allocated to each of the multiple physical links.

[0020] In a possible implementation, dynamically adjust the priorities of the multiple physical links through a dynamic routing protocol to identify the multiple physical links as equivalent paths, including:

[0021] Based on the Open Shortest Path First (OSPF) protocol, set the cost values of the multiple physical links to the same value, and enable the Equal-Cost Multi-Path Routing (ECMP) function of the OSPF protocol, such that the multiple physical links are identified as equivalent paths.

[0022] In a possible implementation, when a failure occurs in any physical link, the service traffic is switched to an available target physical link through the load balancing algorithm, including:

[0023] When a failure occurs in any physical link, based on a failure prediction model, perform failure prediction on the remaining physical links to obtain the failure prediction results of each remaining physical link within a preset future duration, where the remaining physical links are the other physical links among the multiple physical links except those with failures;

[0024] Regard all the physical links with failure prediction results of no failure risk among the remaining physical links as available target physical links;

[0025] Switch the service traffic to the available target physical link through the load balancing algorithm.

[0026] In a second aspect, an embodiment of the present application provides a VXLAN-based data transmission device, and the device includes:

[0027] An adjustment unit, configured to, after virtualizing multiple physical links between two user terminals into a virtual link through the Virtual eXtensible Local Area Network (VXLAN) technology, dynamically adjust the priorities of the multiple physical links through a dynamic routing protocol to identify the multiple physical links as equivalent paths;

[0028] A load distribution unit, configured to, based on the equivalent paths, disperse service traffic to the multiple physical links for transmission through a load balancing algorithm;

[0029] A switching unit, configured to monitor the physical link status in real time, and when a failure occurs in any physical link, switch the service traffic to an available target physical link through the load balancing algorithm.

[0030] In a possible implementation, the load distribution unit includes:

[0031] An acquisition module, configured to acquire the link weights assigned to each physical link among the multiple physical links based on a weight distribution model;

[0032] A calculation module, configured to calculate the smallest integer ratio of all link weights among the multiple physical links, and calculate the sum of each value in the smallest integer ratio as the total weight;

[0033] A configuration module, configured to, according to the total weight and the smallest integer ratio, configure a corresponding weight interval for the link weight of each physical link such that the length of the weight interval is equal to the corresponding integer in the smallest integer ratio;

[0034] The calculation module is further configured to calculate the hash value of the target information in each packet of the service traffic respectively, and perform normalization processing on each hash value to obtain the normalized hash values.

[0035] The determination module is configured to determine the weight interval where each normalized hash value is located.

[0036] The allocation module is configured to allocate the corresponding packets to the physical links corresponding to the weight intervals for transmission.

[0037] In a possible implementation manner, the method for allocating link weights to each of the multiple physical links based on a weight allocation model includes: obtaining the current link state metrics of each of the multiple physical links and the service characteristics of the service traffic; inputting the current link state metrics of each physical link and the service characteristics of the service traffic into the weight allocation model to obtain the link weights allocated to each of the multiple physical links.

[0038] In a possible implementation manner, the adjustment unit is configured to set the cost values of the multiple physical links to the same value based on the Open Shortest Path First (OSPF) protocol, and enable the Equal-Cost Multi-Path Routing (ECMP) function of the OSPF protocol, so that the multiple physical links are identified as equivalent paths.

[0039] In a possible implementation manner, the switching unit includes:

[0040] The fault prediction module is configured to, when detecting that any physical link fails, perform fault prediction on the remaining physical links based on a fault prediction model to obtain the fault prediction results of each of the remaining physical links within a preset future time period, where the remaining physical links are the other physical links among the multiple physical links except the one that has failed;

[0041] The selection module is configured to use all the physical links among the remaining physical links with fault prediction results indicating no fault risk as the available target physical links;

[0042] The switching module is configured to switch the service traffic to the available target physical links through the load balancing algorithm.

[0043] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the method described in any possible implementation manner of the first aspect is implemented.

[0044] In a fourth aspect, an embodiment of the present application provides an electronic device, where the electronic device includes:

[0045] One or more processors;

[0046] The processor is coupled to a storage device for storing one or more programs;

[0047] When the one or more programs are executed by the one or more processors, the electronic device implements the method according to any possible implementation manner of the first aspect.

[0048] In a fifth aspect, an embodiment of the present application provides a computer program product containing instructions, which, when running on a computer or a processor, cause the computer or the processor to execute the method according to any possible implementation manner of the first aspect.

[0049] As can be seen from the above solutions, the data transmission method, apparatus, storage medium, and electronic device provided by the embodiments of the present application based on VXLAN (Virtual Extensible Local Area Network) can virtualize multiple physical links into a logically unified virtual link through the VXLAN technology, breaking through the limitation of the traditional dynamic routing protocol on the number of links, enabling three or more physical links to be recognized as equivalent transmission paths through dynamic priority adjustment, realizing dynamic load sharing of service traffic on multiple physical links, significantly improving the utilization rate of link resources, and avoiding idle waste; at the same time, through real-time link status monitoring and the traffic automatic switching mechanism in case of faults, it ensures that traffic quickly migrates to available physical links in case of fault scenarios, guarantees service continuity, and effectively solves the problem that multiple physical links cannot work together in the traditional architecture, especially suitable for high-density link scenarios such as cloud computing and multi-data center interconnection, realizing the efficient utilization and reliable transmission of network resources.

[0050] In addition, the technical effects that can be achieved by the embodiments of the present application may further include:

[0051] 1. In the embodiments of the present application, a weight distribution model is used to assign different weights to each physical link, and the weights are converted into quantifiable minimum integer ratios and total weights to construct a weight interval matching the link performance, so that traffic distribution can accurately reflect the actual carrying capacity of the link. By performing hash and normalization processing on the target information of the packet, the traffic is dispersed to the corresponding physical links according to the weight ratio, which not only supports physical links with high bandwidth to carry more traffic to fully utilize resources, but also allows low-performance physical links to share the load proportionally, avoiding resource waste or congestion risks in the traditional fixed allocation method. This mechanism realizes dynamic balance and refined control of traffic in the multi-physical link scenario, supports the collaborative work of heterogeneous links (such as a mixture of links with different bandwidths and priorities), significantly improves the flexibility of network resource scheduling and the fairness of traffic distribution, and is especially suitable for complex network environments with requirements for link utilization rate and service differentiation guarantee.

[0052] 2. In the embodiments of the present application, the cost values of multiple physical links can be uniformly set to the same value through the OSPF (Open Shortest Path First) protocol, so that the shortest path algorithm of the protocol (such as the Dijkstra algorithm) determines these links as equivalent paths with the same priority. By enabling the ECMP (Equal-cost multi-path routing) function of OSPF, the limitation of the traditional protocol on dual-link load is broken through, and three or more physical links are supported to participate in traffic transmission simultaneously. This mechanism enables network devices to evenly disperse service traffic to all equivalent links, making full use of the bandwidth resources of each link and avoiding resource idling or traffic congestion caused by link priority differences. When a certain link fails, OSPF automatically eliminates the failed link through dynamic routing convergence, and the remaining equivalent links can still maintain the traffic sharing state, ensuring service continuity. This technical solution significantly improves the routing efficiency and network reliability in multi-link scenarios, and is particularly suitable for scenarios with extremely high requirements for link resource utilization, such as high-bandwidth interconnection inside data centers and multi-outlet load balancing, achieving a technical breakthrough from dual-link to multi-link collaborative work.

[0053] 3. In the embodiments of the present application, by introducing a fault prediction model, when a physical link fails, not only the current failed link is excluded in real time, but also based on the real-time status indicators of the remaining physical links (such as bandwidth, latency, packet loss rate, etc.), the fault risk within a preset duration in the future is predicted, and only the links without fault risk are included in the available target physical link set. This mechanism avoids the secondary interruption problem caused by potential faults of the remaining physical links in traditional fault switching, achieving a double guarantee from "instantly available" to "short-term reliable". By accurately switching traffic to the physically predicted stable links, the overall reliability of the multi-physical link system is significantly improved, and network oscillations caused by frequent switching or blind allocation are reduced, especially suitable for scenarios with extremely high requirements for service continuity (such as industrial control and financial transactions). This technology breaks through the limitations of traditional fault handling. Through forward-looking risk assessment and intelligent screening, passive fault response is transformed into active reliability guarantee, providing a more flexible and predictable solution for traffic scheduling in complex network environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0055] Figure 1 A schematic flowchart of a data transmission method based on VXLAN provided by an embodiment of the present application;

[0056] Figure 2 A schematic flowchart of a load distribution method provided by an embodiment of the present application;

[0057] Figure 3 A block diagram of a data transmission device based on VXLAN provided by an embodiment of the present application. Specific embodiments

[0058] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0059] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments may be combined with each other. The terms "including" and "having" and any variations thereof in the embodiments of the present application and the accompanying drawings are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but optionally further includes steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products, or devices.

[0060] Figure 1 A schematic flowchart of a data transmission method based on VXLAN provided by an embodiment of the present application. This method can be applied to the switch side. VXLAN is deployed in the switch, and the switch with VXLAN deployed can be called a VTEP (VXLAN Tunnel Endpoints) node. The method includes:

[0061] S110: After virtualizing multiple physical links between two user terminals into a virtual link through the Virtual Extensible LAN (VXLAN) technology, dynamically adjust the priorities of the multiple physical links through a dynamic routing protocol to make the multiple physical links recognized as equivalent paths.

[0062] Through the VXLAN technology, multiple physical links between two user terminals can be logically integrated into one "virtual link", enabling Layer 2 service traffic to traverse a Layer 3 physical network (similar to building a unified logical channel on different physical paths). On this basis, the dynamic routing protocol adjusts the priority parameters of each physical link, making the protocol consider these physical links to have the same "quality level" (i.e., equivalent paths). The meaning of equivalent paths is that they all have the right to transmit data, but the amount of data they transmit is not necessarily equal.

[0063] Dynamic routing protocols include the OSPF protocol, the IS-IS (Intermediate system to intermediate system) protocol, etc. When the dynamic routing protocol is the OSPF protocol, the switch can set the cost values of multiple physical links to the same value based on the OSPF protocol and enable the ECMP function of the OSPF protocol, making multiple physical links recognized as equivalent paths. When the dynamic routing protocol is the IS-IS protocol, the switch can set the Metric values of multiple physical links to the same value and enable the ECMP (Equal-cost multi-path routing) function of the IS-IS protocol, making multiple physical links recognized as equivalent paths.

[0064] In the dynamic routing protocol, the Cost value (overhead value) and the Metric value (metric value) are the core parameters for measuring the link priority, used to judge the "quality" of the data transmission path.

[0065] The Cost value is commonly found in the OSPF protocol and represents the transmission cost of the link, usually inversely proportional to the bandwidth (e.g., the Cost value of a 10G link is set to 1, and that of a 1G link is set to 10). The smaller the value, the "better" the link.

[0066] The Metric value is a more general concept and can be used in multiple protocols (such as IS-IS, BGP (Border Gateway Protocol)). In essence, it is a quantitative evaluation of the link state (bandwidth, latency, reliability, etc.). The definitions of different protocols vary slightly (for example, the Metric of IS-IS can be manually configured, and the MED value of BGP is used for cross-autonomous system path comparison).

[0067] In the embodiments of the present application, the core function of both is to enable the routing protocol to calculate the optimal path through an algorithm (such as the Dijkstra shortest path algorithm): when the Cost or Metric values of multiple physical links are configured to be the same, the protocol determines these links as "equivalent paths" and allows traffic to be transmitted in parallel on them, thereby achieving multi-link load sharing. For example, if the Cost values of 3 physical links are all set to 5, OSPF will consider them to have the same priority and then evenly distribute the traffic to these 3 links, making full use of the bandwidth resources of each link.

[0068] S120: Based on the equivalent paths, disperse the service traffic to multiple physical links for transmission through a load sharing algorithm.

[0069] There are various load sharing algorithms, which can be equal distribution or load allocation according to the status of each physical link. The embodiments of the present application provide a load sharing algorithm, as Figure 2 shown, and the specific implementation process includes steps S121 - S125:

[0070] S121: Obtain the link weights assigned to each physical link in multiple physical links based on a weight assignment model.

[0071] Among them, the weight assignment model is a neural network model, which can be deployed in an SDN controller (the SDN controller can be deployed in the cloud) or in a switch.

[0072] The weight assignment model can be obtained by training based on the historical link state metrics of a large number of physical links and the service characteristics of the corresponding service traffic. After the weight assignment model is trained, when it is necessary to assign link weights to physical links, the current link state metrics and service characteristics of service traffic of each physical link in multiple physical links can be obtained; the current link state metrics and service characteristics of service traffic of each physical link are input into the weight assignment model to obtain the link weights assigned to each physical link in multiple physical links.

[0073] Among them, the link state metrics include at least one of bandwidth utilization rate, latency duration, packet loss rate, and jitter degree, and the service characteristics of service traffic include traffic type, priority label, etc.

[0074] The traffic type can classify network traffic according to the transmission characteristics, real-time requirements, or application scenarios of service data, etc., and is used to distinguish the quality of service (QoS) requirements of different services. For example:

[0075] Real-time interactive traffic: such as voice calls, video conferences, and remote controls. This type of traffic is extremely sensitive to latency (end-to-end delay) and jitter (delay fluctuation), allows a certain degree of packet loss but needs to ensure low latency;

[0076] Bulk transfer traffic: such as file downloads, data backups, and email attachments. This type of traffic requires high bandwidth utilization but allows for relatively high latency (e.g., on the order of seconds), and is more concerned with throughput rather than real-time performance.

[0077] Bursty transaction traffic: such as database queries and web browsing. It is characterized by short connections and high burstiness, with a certain tolerance for latency but requiring fast response times.

[0078] Multicast / broadcast traffic: such as video live streaming and internet radio. It requires efficient multicast routing support to avoid overloading individual links.

[0079] Priority tags are level identifiers assigned to service traffic, used to indicate the importance or service level of the traffic, so that network devices can perform differential processing during resource contention (such as preferential forwarding, bandwidth reservation). Common implementation methods include:

[0080] IP layer priority: such as the Type of Service (ToS) field in IPv4 or the Traffic Class field in IPv6. Among them, DSCP (Differentiated Services Code Point) can define 64 priorities (such as CS6, EF marking high-priority services);

[0081] Layer 2 priority: such as the 3-bit priority tag defined by the Ethernet 802.1p protocol (a total of 8 levels), used for traffic priority marking within a local area network;

[0082] Custom tags: In the VXLAN scenario, service priorities can be carried through the VNI (Virtual Network Identifier) or extended fields, and combined with QoS policies to achieve priority transmission across three-layer networks.

[0083] In addition, the loss function of the weight allocation model can be:

[0084]

[0085] where L represents the loss value, N represents the number of physical links, K represents the number of service traffic types, ω i represents the link weight assigned by the model to the i-th physical link, ω i,k represents the weight of the k-th type of service on the i-th physical link, U i represents the bandwidth utilization rate of the i-th physical link, D i represents the latency duration of the i-th physical link, L1 represents the packet loss rate of the i-th physical link, J i represents the jitter degree of the i-th physical link, Priority k represents the priority of the k-th type of service (e.g., real-time video = 3, file transfer = 1), and α, β, γ, δ, ∈ are hyperparameters that control the importance of each component.

[0086] α·Var(U i ) represents the bandwidth balancing term, which is used to avoid overloading a certain physical link and promote the balance of bandwidth utilization. It calculates the variance (Var(U i )) of the bandwidth utilization of all links. The smaller the variance, the more balanced the allocation.

[0087] represents the delay penalty term, which is used to assign a lower weight to high-delay links. By accumulating the product of the weight of each link and its delay, the higher the delay, the greater the penalty.

[0088] represents the packet loss penalty term, which is used to reduce the weight of high-packet-loss links. By accumulating the product of the weight of each link and its packet loss rate, the higher the packet loss rate, the greater the penalty

[0089] represents the jitter penalty term, which is used to reduce the path jitter of jitter-sensitive services. By accumulating the product of the weight of each link and its jitter, the greater the jitter, the greater the penalty.

[0090] represents the service characteristic matching term, which is used to avoid high-delay and high-packet-loss links for high-priority services. By calculating the weighted sum of the delay and packet loss of the allocated path for each type of service, the higher the priority, the more sensitive it is to this term.

[0091] During the model training process, for each round of training, the loss value of the above loss function can be calculated. When the current loss value is greater than the preset loss threshold, after adjusting the model parameters, the model continues the next round of training until the current loss value is less than or equal to the preset loss threshold, and the final required weight allocation model is obtained.

[0092] Through this loss function, the model can learn a weight allocation strategy that can both improve network resource utilization and meet service quality requirements.

[0093] S122: Calculate the smallest integer ratio of the weights of all links in multiple physical links, and calculate the sum of each value in the smallest integer ratio as the total weight.

[0094] For example, the link weight values of a single physical link output by the weight allocation model are: 30% for physical link 1, 50% for physical link 2, and 20% for physical link 3. Then physical link 1: physical link 2: physical link 3 = 3:5:2, and the total weight = 3 + 5 + 2 = 10.

[0095] S123: According to the total weight and the smallest integer ratio, configure the corresponding weight interval for the link weight of each physical link, so that the length of the weight interval is equal to the corresponding integer in the smallest integer ratio.

[0096] Continuing with the above example, when Physical Link 1: Physical Link 2: Physical Link 3 = 3:5:2 and the total weight = 10, the weight range corresponding to Physical Link 1 can be [0, 3), the weight range corresponding to Physical Link 2 can be [3, 8), and the weight range corresponding to Physical Link 2 can be [8, 10).

[0097] S124: Calculate the hash value of the target information in each packet of the service traffic respectively, and perform normalization processing on each hash value to obtain the normalized hash value.

[0098] Among them, the target information can include the five-tuple of the packet, including source IP, destination IP, source port, destination port, and protocol, or can include other information.

[0099] The method of performing normalization processing on each hash value to obtain the normalized hash value includes: dividing each hash value by the total weight respectively, and the remainder obtained is the normalized hash value. For example, if the hash value corresponding to Packet 1 is 12, then the normalized hash value obtained after normalization is 2, and the corresponding weight range is [0, 3), and it can be allocated to Physical Link 1. Another example, if the hash value corresponding to Packet 1 is 25, then the normalized hash value obtained after normalization is 5, and the corresponding weight range is [3, 8), and it can be allocated to Physical Link 2.

[0100] S125: Determine the weight range where each normalized hash value is located, and allocate the corresponding packet to the physical link corresponding to the weight range for transmission.

[0101] Since each physical link has its corresponding weight range, after normalizing the hash value corresponding to each packet to determine the weight range where it is located, the packet can be allocated to the physical link corresponding to the weight range for transmission.

[0102] In the embodiment of the present application, a differential weight is assigned to each physical link through a weight allocation model, and the weight is converted into a quantifiable minimum integer ratio and total weight, and a weight range matching the link performance is constructed, so that the traffic allocation can accurately reflect the actual bearing capacity of the link. By performing hash and normalization processing on the target information of the packet, the traffic is dispersed to the corresponding physical link according to the weight ratio, which not only supports the physical link with high bandwidth to bear more traffic to fully utilize resources, but also allows the low-performance physical link to share the load proportionally, avoiding resource waste or congestion risk of the traditional fixed allocation method. This mechanism realizes dynamic balance and fine control of traffic in the multi-physical link scenario, supports the collaborative work of heterogeneous links (such as mixed links with different bandwidths and priorities), significantly improves the flexibility of network resource scheduling and the fairness of traffic allocation, and is especially suitable for complex network environments with requirements for link utilization rate and service differentiation guarantee.

[0103] It should be added that before the service traffic is distributed to each physical link through the load balancing algorithm, independent encryption keys (such as AES-GCM or SM4) can be generated in advance for each physical link, and the encryption algorithm and key index are carried through the extended field of the VXLAN header (or custom TLV); when the service traffic is distributed to a certain link through the load balancing algorithm, the corresponding encryption tunnel of the physical link is automatically called for encapsulation (encryption of the outer IP packet), ensuring traffic key isolation for different links.

[0104] Among them, the following information is embedded in the extended field of the VXLAN header (such as a newly added TLV field): encryption algorithm identifier (such as AES-GCM, SM4), key index (pointing to the key corresponding to the link), so as to inform the receiving end of the algorithm and key position required for decryption without transmitting the key itself.

[0105] When the service traffic is distributed to a certain physical link through the load balancing algorithm according to the policy (such as hash, weight), the corresponding encryption key of the physical link is called, and after encrypting the outer IP packet (the outer encapsulation of the VXLAN tunnel) of the transmission, the transmission is carried out, so that the service traffic is transmitted through the physical link in ciphertext form, and the encryption keys of different physical links do not interfere with each other.

[0106] In the embodiment of the present application, by generating independent encryption keys for each physical link and dynamically carrying the encryption algorithm and key index by using the VXLAN header extension field or custom TLV, fine-grained security protection of "one link, one key" in a multi-link environment is realized. When the service traffic is distributed to a specific link through the load balancing algorithm, the system automatically calls the exclusive encryption tunnel of the link to encrypt the outer IP of the packet, ensuring that the traffic of different links is encapsulated with independent keys, and cutting off the risk of cross-link key association from the bottom layer.

[0107] S130: Monitor the physical link status in real time. When it is detected that any physical link fails, the service traffic is switched to the available target physical link through the load balancing algorithm.

[0108] In addition to deploying the OSPF module in the switch, the BFD module can also be deployed. During the communication process between two user terminals through their respective directly connected switches based on multiple physical links, when the BFD module in any switch detects that at least one of the multiple physical links has been interrupted, the BFD module in that switch sends a link interruption notification to the local OSPF module; when the OSPF module in the switch receives the link interruption notification, it skips waiting for the Dead timer and directly switches the service traffic to the available target physical link through the load balancing algorithm. Among them, the service traffic can be directly switched to all the remaining physical links through the load balancing algorithm, or the physical links that will not fail within a preset future duration can be selected from all the remaining physical links first, and then the service traffic can be switched to the selected physical links through the load balancing algorithm for transmission, so as to avoid the phenomenon that a new failure occurs just after the traffic is switched in a short time and needs to be switched again.

[0109] Specifically, when any physical link failure is detected, based on the fault prediction model, the remaining physical links are predicted for faults, and the fault prediction results of each remaining physical link within the preset future duration are obtained. The remaining physical links are the other physical links among the multiple physical links except those with failures; all the physical links with fault prediction results of no fault risk among the remaining physical links are used as the available target physical links; the service traffic is switched to the available target physical links through the load balancing algorithm.

[0110] Among them, using the pre-trained fault prediction model, the real-time status indicators of the remaining physical links are input, and the fault probability of each physical link within the preset future duration (such as the next 30 minutes) is output. Only the physical links with a fault probability less than or equal to the preset probability threshold are included in the set of available target physical links.

[0111] The fault prediction model can adopt supervised learning algorithms (such as random forest, LSTM (Long Short-Term Memory) neural network). Among them, the tree model is suitable for processing structured time series data, and LSTM is suitable for capturing the time series dependencies of link states. When training the fault prediction model, the collected training samples include the historical status indicators of each physical link and the corresponding fault records. Through the gradient descent optimization algorithm, the prediction error (such as the cross-entropy loss function) is minimized to learn the mapping relationship between the link state characteristics and the occurrence of faults. The status indicators include bandwidth utilization rate, delay duration, packet loss rate, jitter degree, temperature, voltage and other indicators. The fault records include the link disconnection time and can also include the fault type, such as hardware failure, traffic congestion, configuration error.

[0112] In addition, new physical link state data can be regularly imported for incremental training to adapt to network environment changes (such as hardware aging and business traffic pattern changes), ensuring long-term stability of prediction accuracy.

[0113] Through the above training process, the fault prediction model can dynamically evaluate the reliability of the remaining physical links based on real-time data, providing a scientific basis for traffic switching, upgrading the traditional "post-response" fault handling mode to "prevention + real-time optimization" intelligent scheduling, and significantly improving the overall availability of the multi-link system.

[0114] The VXLAN-based data transmission method provided by the embodiments of this application can virtualize multiple physical links into a logically unified virtual link through VXLAN technology, breaking through the limitation of the traditional dynamic routing protocol on the number of links, enabling three or more physical links to be identified as equivalent transmission paths through dynamic priority adjustment, realizing dynamic load sharing of service traffic on multiple physical links, significantly improving link resource utilization, and avoiding idle waste; at the same time, through real-time link state monitoring and traffic automatic switching mechanism during faults, it ensures that traffic quickly migrates to available physical links in case of faults, guarantees service continuity, effectively solves the problem that multiple physical links cannot work together in the traditional architecture, and is especially suitable for high-density link scenarios such as cloud computing and multi-data center interconnection, realizing efficient utilization and reliable transmission of network resources.

[0115] Based on the above method embodiments, another embodiment of this application provides a VXLAN-based data transmission device, as Figure 3 shown, the device includes:

[0116] An adjustment unit 210, configured to, after virtualizing multiple physical links between two user terminals into one virtual link through Virtual eXtensible Local Area Network (VXLAN) technology, dynamically adjust the priorities of the multiple physical links through a dynamic routing protocol, so that the multiple physical links are identified as equivalent paths;

[0117] A load distribution unit 220, configured to, based on the equivalent paths, disperse service traffic to the multiple physical links for transmission through a load sharing algorithm;

[0118] A switching unit 230, configured to monitor the physical link state in real time, and when it detects that any physical link fails, switch the service traffic to an available target physical link through the load sharing algorithm.

[0119] In a possible implementation manner, the load distribution unit 220 includes:

[0120] An acquisition module, configured to acquire the link weights assigned to each of the multiple physical links based on a weight distribution model;

[0121] A calculation module, configured to calculate the minimum integer ratio of all link weights in the multiple physical links, and calculate the sum of each value in the minimum integer ratio as the total weight;

[0122] A configuration module, configured to configure a corresponding weight interval for the link weight of each physical link according to the total weight and the minimum integer ratio, so that the length of the weight interval is equal to the corresponding integer in the minimum integer ratio;

[0123] The calculation module is further configured to calculate the hash value of the target information in each packet in the service traffic respectively, and perform normalization processing on each hash value to obtain the normalized hash value;

[0124] A determination module, configured to determine the weight interval where each normalized hash value is located;

[0125] An allocation module, configured to allocate the corresponding packet to the physical link corresponding to the weight interval for transmission.

[0126] In a possible implementation manner, the method for allocating link weights to each physical link in the multiple physical links based on a weight allocation model includes: obtaining the current link state index of each physical link in the multiple physical links and the service characteristics of the service traffic; inputting the current link state index of each physical link and the service characteristics of the service traffic into the weight allocation model to obtain the link weights allocated to each physical link in the multiple physical links.

[0127] In a possible implementation manner, the adjustment unit 210 is configured to set the cost values of the multiple physical links to the same value based on the Open Shortest Path First (OSPF) protocol, and enable the Equal-Cost Multi-Path Routing (ECMP) function of the OSPF protocol, so that the multiple physical links are identified as equivalent paths.

[0128] In a possible implementation manner, the switching unit 230 includes:

[0129] A fault prediction module, configured to, when detecting that any physical link fails, perform fault prediction on the remaining physical links based on a fault prediction model to obtain the fault prediction results of each remaining physical link within a preset duration in the future, where the remaining physical links are the other physical links except the failed one among the multiple physical links;

[0130] A selection module, configured to use the physical links with fault prediction results of no fault risk among the remaining physical links as available target physical links;

[0131] A switching module, configured to switch the service traffic to an available target physical link through the load balancing algorithm.

[0132] The VXLAN-based data transmission device provided by the embodiments of the present application can virtualize multiple physical links into a logically unified virtual link through VXLAN technology, breaking through the limitation of the traditional dynamic routing protocol on the number of links, enabling three or more physical links to be recognized as equivalent transmission paths through dynamic priority adjustment, realizing dynamic load balancing of service traffic on multiple physical links, significantly improving the utilization rate of link resources, and avoiding idle waste. At the same time, through real-time link status monitoring and automatic traffic switching mechanism during faults, it ensures that traffic can be quickly migrated to available physical links in case of faults, guarantees service continuity, and effectively solves the problem that multiple physical links cannot work together in the traditional architecture. It is particularly suitable for high-density link scenarios such as cloud computing and multi-data center interconnection, realizing efficient utilization and reliable transmission of network resources.

[0133] Based on the above method embodiments, another embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the method described in any of the above embodiments.

[0134] Based on the above method embodiments, another embodiment of the present application provides an electronic device, the electronic device includes:

[0135] One or more processors;

[0136] The processor is coupled to a storage device, and the storage device is used to store one or more programs;

[0137] When the one or more programs are executed by the one or more processors, the electronic device implements the method described in any of the above embodiments.

[0138] Based on the above embodiments, another embodiment of the present application provides a computer program product, which contains instructions, and when the instructions run on a computer or a processor, the computer or the processor executes the method described in any of the above embodiments.

[0139] The above device embodiments and system embodiments correspond to the method embodiments respectively, and have the same technical effects as the method embodiments. For specific descriptions, please refer to the method embodiments. The device embodiments and system embodiments are obtained based on the method embodiments. For specific descriptions, please refer to the method embodiment part, and details are not repeated here. Those of ordinary skill in the art can understand that the drawings are only schematic diagrams of one embodiment, and the modules or processes in the drawings are not necessarily essential for implementing the present application.

[0140] Those of ordinary skill in the art can understand that the modules in the devices in the embodiments can be distributed in the devices in the embodiments as described in the embodiments, or can be correspondingly changed and located in one or more devices different from the present embodiment. The modules in the above embodiments can be combined into one module, or can be further split into multiple sub-modules.

[0141] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A data transmission method based on VXLAN, characterized in that The method includes: After virtualizing multiple physical links between two user terminals into one virtual link through the Virtual eXtensible Local Area Network (VXLAN) technology, dynamically adjust the priorities of the multiple physical links through a dynamic routing protocol to identify the multiple physical links as equivalent paths; Based on the equivalent paths, disperse the service traffic to the multiple physical links for transmission through a load balancing algorithm; Monitor the physical link status in real time. When any physical link fails, switch the service traffic to an available target physical link through the load balancing algorithm.

2. The method according to claim 1, wherein Dispersing the service traffic to the multiple physical links for transmission through a load balancing algorithm includes: Obtain the link weights assigned to each of the multiple physical links based on a weight assignment model; Calculate the minimum integer ratio of all link weights among the multiple physical links, and calculate the sum of each value in the minimum integer ratio as the total weight; According to the total weight and the minimum integer ratio, configure a corresponding weight interval for the link weight of each physical link, such that the length of the weight interval is equal to the corresponding integer in the minimum integer ratio; Calculate the hash value of the target information in each packet in the service traffic respectively, and perform normalization processing on each hash value to obtain the normalized hash values; Determine the weight interval where each normalized hash value is located, and allocate the corresponding packets to the physical link corresponding to the weight interval for transmission.

3. The method according to claim 2, wherein The method for assigning link weights to each of the multiple physical links based on a weight assignment model includes: Obtain the current link status metrics of each of the multiple physical links and the service characteristics of the service traffic; Input the current link status metrics of each physical link and the service characteristics of the service traffic into the weight assignment model to obtain the link weights assigned to each of the multiple physical links.

4. The method according to claim 1, wherein Dynamically adjusting the priorities of the multiple physical links through a dynamic routing protocol to identify the multiple physical links as equivalent paths includes: Based on the Open Shortest Path First (OSPF) protocol, set the cost values of the multiple physical links to the same value, and enable the Equal Cost Multi-Path (ECMP) function of the OSPF protocol, such that the multiple physical links are identified as equivalent paths.

5. The method according to any one of claims 1 to 4, characterized in that, When any physical link fails, switching the service traffic to an available target physical link through the load balancing algorithm includes: When any physical link fails, perform a fault prediction on the remaining physical links based on a fault prediction model to obtain the fault prediction results of each remaining physical link within a preset time period in the future, where the remaining physical links are the other physical links among the multiple physical links except the one that has failed; Regard the physical links among the remaining physical links with fault prediction results of no fault risk as the available target physical links; Switch the service traffic to the available target physical links through the load balancing algorithm.

6. A data transmission device based on VXLAN, characterized in that The device includes: An adjustment unit, configured to, after virtualizing multiple physical links between two user terminals into one virtual link through Virtual eXtensible Local Area Network (VXLAN) technology, dynamically adjust the priorities of the multiple physical links through a dynamic routing protocol, so that the multiple physical links are recognized as equivalent paths; A load distribution unit, configured to, based on the equivalent paths, disperse service traffic to the multiple physical links for transmission through a load sharing algorithm; A switching unit, configured to monitor the physical link status in real time, and when any physical link is detected to have a fault, switch the service traffic to an available target physical link through the load sharing algorithm.

7. The device according to claim 6, characterized in that, The load distribution unit includes: An acquisition module, configured to acquire the link weights assigned to each of the multiple physical links based on a weight distribution model; A calculation module, configured to calculate the minimum integer ratio of all the link weights in the multiple physical links, and calculate the sum of each value in the minimum integer ratio as the total weight; A configuration module, configured to configure a corresponding weight interval for the link weight of each physical link according to the total weight and the minimum integer ratio, so that the length of the weight interval is equal to the corresponding integer in the minimum integer ratio; The calculation module is further configured to calculate the hash value of the target information in each packet in the service traffic respectively, and perform normalization processing on each hash value to obtain a normalized hash value; A determination module, configured to determine the weight interval where each normalized hash value is located; An allocation module, configured to allocate the corresponding packet to the physical link corresponding to the weight interval for transmission.

8. The device according to claim 6, characterized in that The adjustment unit is configured to set the cost values of the multiple physical links to the same value based on the Open Shortest Path First (OSPF) protocol, and enable the Equal-Cost Multi-Path Routing (ECMP) function of the OSPF protocol, so that the multiple physical links are recognized as equivalent paths.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by a processor, the method described in any one of claims 1-5 is implemented.

10. An electronic device, characterized in that, The electronic device includes: One or more processors; The processor is coupled to a storage device, and the storage device is used to store one or more programs; When the one or more programs are executed by the one or more processors, the electronic device implements the method described in any one of claims 1-5.

Images