Network methods, network devices, and network apparatuses

Abstract

The present disclosure relates to a network method, a network device and a network apparatus. The network method is for a network comprising a plurality of logical network nodes connected by physical network links, the network method being performed by a controller of the network and comprising: determining a link quality of each physical link between a first logical network node and a second logical network node of the plurality of logical network nodes; and based on the determined link quality of each physical link, link quality labelling the physical links.

Description

Technical Field

[0001] This disclosure relates to the field of network communications, and more specifically, to a method, network device, and network apparatus for a network comprising multiple logical network nodes. Background Technology

[0002] Virtual Extended Local Area Network (VXLAN) is a Layer 3 network virtualization (NVO3) technology defined by the Internet Engineering Task Force (IETF), and is an extension of the traditional Extended Local Area Network (VLAN) protocol. VXLAN establishes a logical VXLAN tunnel between the source and destination network devices, using "MAC in UDP (User Datagram Protocol)" encapsulation. This means that the raw Ethernet packets sent by the virtual machine are completely encapsulated within a UDP packet, and then further encapsulated using the physical network's Internet Protocol (IP) header and Ethernet header. In this way, the encapsulated packets behave like ordinary IP packets and can be forwarded through the routing network.

[0003] VXLAN comprises a logical network (overlay) and a physical network (underlay). The physical network consists of switches, routers, physical links, IP routes, etc., and achieves IP reachability through traditional routing protocols such as Open Shortest Path First (OSPF), Intermediate System to Intermediate System (IS-IS), and Border Gateway Protocol (BGP). The logical network is a virtual network built on top of the physical network (e.g., virtual Layer 2 and Layer 3 networks). It consists of logical entities such as VXLAN Tunnel Endpoints (VTEPs), Virtual Network Identifiers (VNIs), and Virtual Network Policies. VXLAN tunnel information is automatically distributed through control plane protocols such as Ethernet Virtual Private Networks (EVPN), or forwarding rule tables are directly distributed to switches by the Software Defined Networking (SDN) controller. Summary of the Invention

[0004] Based on the above, this disclosure provides methods, network devices, network apparatuses, computer-readable storage media, and program products for networks comprising multiple logical network nodes.

[0005] In one aspect, this disclosure provides a method for a network comprising a plurality of logical network nodes connected via a physical network, the method being executed by a controller of the network and comprising: determining the link quality of each physical link between a first logical network node and a second logical network node among the plurality of logical network nodes; and tagging each physical link based on the determined link quality of each physical link.

[0006] In one aspect, this disclosure provides a method for a network comprising a plurality of logical network nodes connected via a physical network, wherein a first logical network node and a second logical network node are connected via at least one physical link, the method being performed by the first logical network node and comprising: sending a message for establishing a communication channel between the first logical network node and the second logical network node, the message including information indicating quality of service (QoS) requirements for data to be transmitted; and using the established communication channel for data transmission, the communication channel being established based on a physical link among the at least one physical link whose link quality matches the QoS requirements, each of the at least one physical link being tagged with its link quality.

[0007] In one aspect, this disclosure provides a method for a network comprising a plurality of logical network nodes connected via a physical network, wherein a first logical network node and a second logical network node are connected via at least one physical link, the method being performed by the second logical network node and comprising: receiving a message for establishing a communication channel between the first logical network node and the second logical network node, the message including information indicating quality of service (QoS) requirements for data to be transmitted; and, based on the QoS requirements, selecting a physical link among the at least one physical link whose link quality matches the QoS requirements to establish the communication channel between the first logical network node and the second logical network node, each of the at least one physical link being tagged with its link quality.

[0008] In one aspect, this disclosure provides a network device including: a memory storing instructions, and a processor that executes the instructions to implement a network method according to embodiments of this disclosure.

[0009] In one aspect, this disclosure provides a network apparatus including: components for performing steps of a network method according to embodiments of this disclosure.

[0010] In one aspect, this disclosure provides a computer-readable storage medium having instructions stored thereon that, when executed by a processor, implement a networking method according to embodiments of this disclosure.

[0011] In one aspect, this disclosure provides a computer program product including instructions that, when executed by a processor, implement a networking method according to embodiments of this disclosure.

[0012] The technical solution according to embodiments of this disclosure uses the link quality of each physical link connecting each logical network node to mark the link quality of each physical link. When establishing a communication channel between any two logical network nodes, the quality of service (QoS) requirements of the data to be transmitted between these two logical network nodes are appended to the message used to establish the communication channel. This allows the establishment of the communication channel by selecting a physical link whose link quality matches the QoS requirements of the data to be transmitted from among multiple physical links between the two logical network nodes, based on the QoS mark of the physical link and the QoS requirements in the message. In this way, the QoS requirements of logical network services can be matched with the actual link quality of the physical links in the physical network, achieving cross-network awareness between the logical and physical networks and improving the data transmission performance of the network. Attached Figure Description

[0013] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments will be briefly described below. Obviously, the drawings described below only relate to some embodiments of this disclosure and are not intended to limit this disclosure.

[0014] Figure 1 An example network system to which network methods according to embodiments of this disclosure can be applied is shown.

[0015] Figure 2 This is an example flowchart of a network method according to an embodiment of the present disclosure.

[0016] Figure 3A The frame structure of an existing EVPN route update message is shown.

[0017] Figure 3B The frame structure of a routing update message that indicates the quality of service requirements of data to be transmitted, according to an embodiment of the present disclosure, is shown.

[0018] Figure 4 An example flowchart of the VXLAN tunnel establishment process in a VXLAN network according to an embodiment of the present disclosure is shown.

[0019] Figure 5 An example flowchart of a physical link dynamic adjustment process in a VXLAN network according to an embodiment of the present disclosure is shown.

[0020] Figure 6 An example flowchart illustrates a physical link resource scheduling adjustment process in a VXLAN network based on changes in the quality of service requirements of logical network-based data, according to embodiments of the present disclosure.

[0021] Figure 7An example configuration of a network device according to an embodiment of the present disclosure is shown. Detailed Implementation

[0022] The technical solutions of this disclosure will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are part of, but not all of, the embodiments of this disclosure. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without any creative effort are within the protection scope of this disclosure.

[0023] In the description of this disclosure, it should be noted that the directions or positional relationships indicated by terms such as “center,” “upper,” “lower,” “left,” “right,” “vertical,” “horizontal,” “inner,” and “outer” are based on the directions or positional relationships shown in the figures and are used only for convenience and simplification of the description of this disclosure, and do not indicate or imply that the indicated device or element must have a specific orientation. Furthermore, terms such as “first,” “second,” and “third” are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Similarly, words such as “a,” “an,” or “the” do not represent a limitation of quantity but rather indicate the presence of at least one. Words such as “comprising” or “including” mean that the element or object preceding the word includes those elements or objects listed after the word and their equivalents, without excluding other elements or objects. Words such as “connected” or “linked” are not limited to physical or mechanical connections but can include direct or indirect electrical connections.

[0024] In the description of this disclosure, it should be noted that, unless otherwise expressly specified and limited, terms such as “installation,” “link,” and “connection” should be interpreted broadly. For example, these terms may refer to a fixed connection, a detachable connection, or an integral connection; they may refer to a mechanical connection or an electrical connection; they may refer to a direct connection, an indirect connection via an intermediate medium, or a connection within two components. For those skilled in the art, the specific meaning of the above terms in this disclosure can be understood according to the specific circumstances.

[0025] Furthermore, the technical features described in the different embodiments of this disclosure below can be combined with each other, as long as they do not conflict with each other. Additionally, the accompanying drawings are for illustrative purposes only and have been simplified for brevity, and therefore may not be exactly the same as actual implementations. For example, in the figures, device processing delays may be omitted.

[0026] As described in the background section, a VXLAN network comprises a logical network and a physical network. The physical network achieves IP reachability through traditional routing protocols such as OSPF, IS-IS, and BGP. The logical network automatically distributes VXLAN tunnel information through control plane protocols such as EVPN or directly distributes forwarding rule tables to switches via the SDN controller. However, this disclosure finds that in current VXLAN network operations, the allocation of physical and logical network resources is fragmented. Specifically, in existing VXLAN application scenarios, the physical network is responsible for the connectivity of the underlying network but does not consider the quality of the connectivity paths, while the logical network is responsible for service path selection but does not consider the actual capabilities of the physical network. In other words, the physical and logical networks are completely decoupled and share physical resources. While this greatly improves flexibility and enables multi-tenancy support, this architecture also brings some problems. For example, the VXLAN network architecture may lead to an imbalance in resource utilization. For instance, the logical network may allocate multiple VNIs and tunnels for a high-traffic service, but these tunnels may be hashed to the same congested physical link, while other physical links remain idle. For example, VXLAN network architecture may lead to difficulties in guaranteeing service experience. Specifically, because the quality of service requirements (such as low latency and high bandwidth) cannot be directly translated into physical network links and resource reservation policies, the network quality of services is often uncertain. This is especially true for applications sensitive to network performance (such as AI training and financial transactions), which struggle to meet their network quality requirements. Furthermore, VXLAN network architecture does not allow for flexible resource adjustments. Specifically, physical network expansion or changes (such as adding a core (Spine) switch) are time-consuming and cautious, while adjustments to the logical network are quick and flexible. This mismatch in pace limits the overall flexibility of network resource adjustments.

[0027] Based on the above problems, some manufacturers propose to achieve the linkage between the logical network and the physical network through a centralized controller. Specifically, after determining the physical links for data transmission based on the existing methods, periodically collect the network telemetry data of each physical link, determine the optimal VXLAN physical link for data transmission based on the telemetry data, and when there are resource conflicts, trigger the re-selection of physical links. When re-selecting physical links, re-select physical links through static configuration or relying on a predefined policy template. For example, the determined optimal VXLAN physical link can be selected. Some other manufacturers propose to optimize resource utilization through multi-link detection and dynamic switching. Specifically, intercept the encapsulated packets at the VTEP node, select the optimal path according to the link bandwidth, packet rate and other states, and modify the outer IP header to implement traffic scheduling. However, there are some deficiencies in the above solutions. For example, achieving the linkage between the logical network and the physical network through a centralized controller requires relying on the global computing power of the controller, there may be decision-making delays in a large-scale network, and complex negotiation mechanisms are required for load balancing in a multi-controller scenario. Multi-link detection and dynamic switching rely on local decisions of the VTEP and are difficult to optimize globally.

[0028] Based on the above, the present disclosure provides a network method, network device, network apparatus, computer-readable storage medium and program product that can solve one or more of the above problems. According to the technical solution of the embodiment of the present disclosure, the link quality of each physical link connecting each logical network node is used to mark (i.e., tag) the link quality of each physical link, and when establishing a communication channel between any two logical network nodes, the quality of service requirements of the data to be transmitted between these two logical network nodes is attached to the message used to establish the communication channel, so that when establishing the communication channel, the physical link whose link quality matches the quality of service requirements of the data to be transmitted can be selected from multiple physical links between the two logical network nodes based on the service quality mark of the physical link and the quality of service requirements in the message to establish the communication channel between the two logical network nodes. In this way, the quality of service requirements of the logical network service can be matched with the actual link quality of the physical links of the physical network, realizing cross-network awareness between the logical network and the physical network and improving the data transmission performance of the network.

[0029] To better understand the present disclosure, some terms used in the present disclosure are first described herein. [[ID=...]] [[ID=...]]

[0030] VXLAN: VXLAN is a Layer 3 network virtualization technology defined by the IETF, and it is an extension of traditional extended local area network protocols. A key feature of VXLAN is that it encapsulates L2 Ethernet frames into UDP packets and transmits them over the L3 network. Essentially, VXLAN is a tunneling technology that establishes a logical tunnel over the IP network between the source and destination network devices, allowing user-side packets to be forwarded through this tunnel after specific encapsulation.

[0031] VTEP: A VTEP is an edge device in a VXLAN network, serving as both the start and end point of a VXLAN tunnel. Both the encapsulation and decapsulation of raw user data frames are performed on the VTEP. A VTEP can be a standalone network device (e.g., a switch) or a virtual switch within a server. Raw data frames sent from the source server are encapsulated into VXLAN format packets on a VTEP and then transmitted over the IP network to another VTEP.

[0032] VNI: A VNI is a user identifier similar to a VLAN ID. A VNI represents a tenant, and virtual machines belonging to different VNIs cannot directly communicate at Layer 2. When encapsulating VXLAN packets, a 24-bit length space is allocated to the VNI. In distributed gateway deployment scenarios, VNIs can also be divided into Layer 2 VNIs and Layer 3 VNIs with different functions. Layer 2 VNIs are ordinary VNIs, mapped 1:1 to the broadcast domain (BD) to enable forwarding of VXLAN packets within the same subnet; Layer 3 VNIs are associated with Virtual Private Network (VPN) instances for forwarding VXLAN packets across subnets.

[0033] BGP EVPN Protocol: In VXLAN networks, the BGP EVPN protocol is used as the control plane protocol of the VXLAN network. It can help VTEPs learn host IP and MAC (media access control) addresses and automatically establish VXLAN tunnels between two VTEPs.

[0034] Figure 1 An example network system 100 to which network methods according to embodiments of the present disclosure can be applied is illustrated. Network system 100 may be a VXLAN network system. Network system 100 may include VTEPs 110a-110c. Each of VTEPs 110a-110c may be connected to a server for data transmission between the respective servers. Figure 1In this system, VTEPs 110a-110c can each connect to servers 120a-120c. VTEPs 110a-110c can be interconnected via a physical network (e.g., an IP network). More specifically, VTEPs can be interconnected via one or more physical links in the physical network. For example, VTEP 110a can be connected to VTEP 110b via multiple physical links. Different physical links can be associated with different interfaces of network devices (e.g., switches) or different spine switches (not shown). A VXLAN tunnel can be established between two VTEPs in VTEPs 110a-110c. The network system 100 may also include a network controller (not shown), which includes a portion that controls the logical network composed of the individual VTEPs and a portion that controls the physical network that interconnects the individual VTEPs.

[0035] It should be understood that Figure 1 The network system 100 described is merely an example and not a limitation of this disclosure. For example, although Figure 1 A VXLAN network system is illustrated, but the network system to which the network methods according to embodiments of this disclosure can be applied can be any other network system, as long as the network system includes multiple logical network nodes and these multiple logical network nodes are interconnected via a physical network. For example, although... Figure 1 The network system 100 includes three VTEPs, but a network system to which the network methods according to embodiments of the present disclosure can be applied may include more or fewer VTEPs.

[0036] Figure 2 This is an example flowchart of a network method 200 according to embodiments of the present disclosure. Network method 200 can be used in a network comprising multiple logical network nodes. As previously described, these multiple logical network nodes can be connected via a physical network (e.g., an IP network). Method 200 can be executed by, for example, a network controller. (As in conjunction with...) Figure 1 The network controller may include a portion that controls the physical network and a portion that controls the logical network. Whether the various steps of method 200 are specifically performed by the portion of the controller that controls the physical network or the portion that controls the logical network may depend on the specific steps.

[0037] Method 200 may begin at step S210. At step S210, the link quality of each physical link between the first and second logical network nodes of the plurality of logical network nodes is determined. Exemplarily, determining the link quality of each physical link may include collecting link quality data for each physical link in real time or periodically. For example, link quality data can be collected using an in-band network telemetry (INT) scheme or a border gateway protocol link state (BGP-LS) scheme. The INT scheme can embed an INT header in the network tunnel traffic, collecting metrics such as link latency, packet loss rate, and queue depth hop-by-hop, and then the data is aggregated by the spine node. The BGP-LS scheme can broadcast OSPF / IS-IS topology and service level agreement (SLA) metrics to the SDN controller. Link quality data can be any data indicating link quality, such as one or more of the following: link availability, link data transmission latency, link data transmission jitter, link data transmission packet loss rate, link bandwidth, link data throughput, and link utilization.

[0038] In step S220, based on the determined link quality of each physical link, a link quality label is applied to each physical link. This link quality label is used to determine at least one physical link from among the physical links that matches the service quality requirements of the data to be transmitted when establishing a communication channel between the first logical network node and the second logical network node. Alternatively, the link quality label enables the logical network node to select a physical link whose link quality matches the service quality of the data when subsequently selecting a physical link to establish a communication channel between the two logical network nodes (e.g., a VXLAN tunnel in a VXLAN network). In this way, the service quality requirements of logical network services can be matched with the actual link quality of the physical links in the physical network, achieving cross-network awareness between the logical and physical networks and improving the data transmission performance of the network.

[0039] In one embodiment, physical links can be directly labeled with their link quality data collected. That is, link quality can be labeled using one or more of the following: link availability, link data transmission latency, link data transmission jitter, link data transmission packet loss rate, link bandwidth, link data throughput, and link utilization. In another embodiment, the quality of service (QoS) level of each physical link can be determined based on its link quality data, and then the corresponding physical link can be labeled using the determined QoS level. This simplifies the labeling of physical links, and when leasing an access server, the controller can transmit the physical link's QoS label to the logical network node with lower signaling overhead. For example, the QoS level of a physical link, also known as an SLA label, can be determined based on link quality data, as shown in Table 1.

[0040]

[0041] Table 1

[0042] After link quality marking (LQT) is applied to the physical links between logical network nodes in the network, when establishing a communication channel between two logical network nodes (i.e., a VXLAN tunnel in VXLAN), the physical links used to establish the communication channel can be selected based on the LQT of the physical links. Specifically, when a tenant accesses a server, they can configure a Quality of Service (QoS) requirement policy for different services (or VNIs), i.e., specify their QoS requirements. When a logical network entity (e.g., a VTEP) publishes server-related information (e.g., IP / MAC Type 2 routing) through a network technology (e.g., EVPN), the QoS requirements can be appended to the message used to establish a communication channel between two logical network nodes (i.e., a channel establishment message, such as a route update in VXLAN). Then, any source logical network node (i.e., the first logical network node) that wants to send data can send a channel establishment message with the QoS requirements appended. After receiving the channel establishment message, the destination logical network node (i.e., the second logical network node) can select an available physical link between the first and second logical network nodes whose link quality matches the quality of service (QoS) requirements of the data to be transmitted to establish a communication channel between them. In a VXLAN network, the source logical network node can send a channel establishment message to a high-performance device (which is also a logical network node). This high-performance device acts as a route reflector, advertising the channel establishment message from the source logical network node throughout the VXLAN network, thus enabling the transmission of the channel establishment message to the destination logical network node. The QoS requirements of the data to be transmitted can be indicated, for example, by latency, bandwidth, packet loss rate, etc. Alternatively, the QoS requirements of the data to be transmitted can be indicated by a Quality of Service (QoS) level.

[0043] For example, suppose there are three physical links between the first logical network node and the second logical network node. The link quality of the physical links is marked by high service quality, medium service quality, and low service quality as shown in Table 1. And suppose the service quality levels of the three physical links are as follows: physical link 1: medium service quality, physical link 2: high service quality, physical link 3: low service quality. The service quality requirements of the data to be transmitted are: link latency < 12ms, link jitter < 2ms, and link packet loss rate < 0.01%. Then it can be determined that physical link 2 in the first logical network node and the second logical network node matches the link quality requirements of the data to be transmitted. Therefore, physical link 2 is selected to establish the communication channel between the first logical network node and the second logical network node.

[0044] Regarding the Quality of Service (QoS) requirements for the data to be transmitted appended to the channel establishment message, this disclosure proposes two approaches for VXLAN networks. One approach is to reuse existing standard attributes, namely, borrowing the Cost Community or Priority Color. However, this approach has limited support for new extensions and, in scenarios supporting standard communities, suffers from semantic inconsistencies, requiring strict internal definition of mapping relationships.

[0045] The second approach is to add a new Extended Community attribute. Specifically, this involves adding a Quality of Service (QoS) requirement Extended Community attribute to the existing Extended Community attributes in the BGP Extended Community attributes list. This approach provides clear semantics and allows for customizable structure. Figure 3A This shows an existing EVPN route update message 300a. When... Figure 3A When the value of the type field 310a-2 in the path attribute field 310a of the EVPN route update message 300a is 16, it indicates that the path attribute for the EVPN route update is an EVPN extended community attribute. The current EVPN extended community attributes include 17 attributes (i.e., sub-type attributes). Table 2 shows the addition of a Quality of Service (QoS) requirement extended community attribute to the current EVPN extended community attributes, where the newly added QoS requirement extended community attribute is indicated by an underscore. That is, when... Figure 3A When the value of subclass field 310a-4 in the message is 18, the path attribute of route update 300a indicates the quality of service requirement extended community attribute, that is, route update message 300a includes the quality of service requirement indicating the data to be transmitted.

[0046]

[0047] Table 2

[0048] Table 3 shows an exemplary structural design for the extended community attributes of service quality requirements.

[0049]

[0050] Table 3

[0051] Figure 3B The frame structure 300b of a route update message based on the path attributes in Tables 2 and 3 above, which are Quality of Service Requirements Extended Community Attributes, is shown. The value of the type field 310b-2 of the path attribute field 310b is 16, indicating that the path attribute is an EVPN Extended Community Attribute, and the value of the subtype field 310b-4 of the path attribute field 310b is 18, indicating that the EVPN Extended Community Attribute is a Quality of Service Requirements Extended Community Attribute.

[0052] It should be understood that Tables 1, 2, and 3 above... Figure 3B This is merely an example and not a limitation of this disclosure. For example, the field names in Table 3 above can be replaced with other names without departing from the scope of this disclosure. As another example, although Table 3 above includes a latency field, a jitter field, a packet loss rate field, and a bandwidth field, other embodiments may include more or fewer fields, such as one or more of the latency field, jitter field, packet loss rate field, and bandwidth field.

[0053] Once a communication channel is established between the first and second logical network nodes using multiple physical links that match the quality of service (QoS) requirements of the data to be transmitted, this established communication channel can be used to transmit data between the first and second logical network nodes.

[0054] To better understand this disclosure, Figure 4 An example flowchart of a VXLAN tunnel establishment process 400 in a VXLAN network according to an embodiment of the present disclosure is shown. Figure 4 The WXLAN tunnel establishment process 400 according to an embodiment of this disclosure may include three parts: 410: Quality of Service (QoS) marking of physical links in the physical network, 420: BGP EVPN protocol extension, and 430: VXLAN tunnel establishment. The QoS marking of physical links in the physical network 410 includes steps S410-2 to S410-8. In step S410-2, physical link data is collected. In step S410-4, the link quality of the physical link is determined based on the collected physical link data. In step S410-6, based on the determined link quality of the physical link, a QoS level is generated for the physical link, such as the high QoS level, medium QoS level, and low QoS level described above. In step S410-8, the physical link is marked with the QoS level. The BGP EVPN protocol extension 420 may include steps S420-2 to S420-8. In step S420-2, an extension attribute scheme is selected: whether to reuse existing standard attributes or add custom attributes. In step S420-4, service quality requirement extended community attribute coding is performed, that is, based on the selected extended attribute scheme, the service quality requirement format is defined, for example, it could be... Figure 3BThe frame format is shown. At step S420-4, the Quality of Service (QoS) requirement is encapsulated. After encapsulating the QoS requirement, the control plane protocol extension is completed (S420-8). VXLAN tunnel establishment 430 may include steps S430-2 to S420-12. At step S430-2, tenant service access occurs. At step S430-4, EVPN route advertising is performed based on the VNI and QoS level policy configured for the tenant's service and the BGP EVPN protocol extension. At step S430-6, QoS requirement extension community attributes are attached to the type 2 route. At step S430-8, the peer VTEP receives the route. At step S430-10, the peer VTEP selects a matching physical link by matching the route target (RT) and QoS requirement. At step S430-12, a VXLAM tunnel is established using the selected physical link. In this way, the link quality of the physical link used to establish the VXLAN tunnel matches the QoS requirement of the data to be transmitted, which facilitates data transmission.

[0055] It should be understood that Figure 4 The VXLAN tunnel establishment process 400 shown is merely an example and not a limitation of this disclosure. For example, although Figure 4 The VXLAN tunnel establishment process 400 shown is based on a VXLAN network, but the communication channel establishment process according to embodiments of this disclosure can be applied to networks similar to VXLAN networks.

[0056] After establishing a communication channel between two logical network nodes as described above, data transmission can then occur on this channel. During subsequent data transmission, the physical links in the network can be monitored in real-time or periodically, and physical link resources can be scheduled based on their actual communication status. In this way, dynamic scheduling of physical link resources during data transmission can be achieved, ensuring that data is transmitted on physical links that best match its quality of service requirements, thereby improving the overall performance of data transmission in the network.

[0057] Specifically, the network controller (specifically, the part of the controller that controls the physical network) can monitor the quality of one or more of the current physical links used for data transmission; determine whether the quality of the current physical link has degraded relative to a marked link quality; and, in response to the degraded quality of the current physical link, determine a new physical link for data transmission. In one embodiment, a new physical link for data transmission can be determined whenever the quality of the current physical link degrades (e.g., from a high quality of service level to a medium or low quality of service level). In another embodiment, a new physical link for data transmission can be determined only when the current physical link is congested. In this way, frequent switching of physical links for data transmission can be avoided. In yet another embodiment, a new physical link for data transmission can be determined only when the quality of a physical link degrades to a predetermined level. For example, in an embodiment where the quality of service level of physical links is marked by high, medium, and low quality of service, a new physical link for data transmission is determined only when the quality of a high quality of service physical link degrades to low quality of service. That is, if the quality of a high quality of service physical link degrades to medium quality of service, no physical link adjustment may be required. In this way, a good trade-off can be achieved between the switching frequency of physical links and data transmission performance. Additionally, the link quality label of the current physical link can be changed to a reduced link quality. For example, in an embodiment where physical links are labeled by their corresponding Quality of Service (QoS) level, the QoS level of the physical link can be updated to the QoS level corresponding to the reduced link quality.

[0058] In embodiments where physical links are tagged by quality of service (QoS) levels (e.g., high, medium, and low QoS as described above), determining a new physical link may include determining whether an available physical link with the same QoS level as the current physical link exists. If such a physical link exists, it can be identified as the new physical link. If not, an available physical link with a lower QoS level than the current physical link can be identified as the new physical link.

[0059] To better understand this disclosure, Figure 5An example flowchart of a physical link dynamic adjustment process 500 in a VXLAN network according to an embodiment of the present disclosure is shown. Process 500 may begin at step S502. At step S502, the link quality of the current physical link used for data transmission is monitored. At step S504, it is determined whether the link quality of the current physical link used for data transmission has degraded. If the link quality of the current physical link has degraded, process 500 may proceed to steps S506 and S508; otherwise, it returns to step S502. At step S506, the link quality label of the current physical link is updated, i.e., the link quality label of the current physical link is updated with the degraded link quality. At step S508, it is determined whether the current physical link is congested. If congested, it proceeds to step S510; otherwise, it returns to step S502. At step S510, it is determined whether there is an available physical link with the same quality of service (QoS) level as the current physical link. If there is, it proceeds to step S512; otherwise, it proceeds to step S516. In step S512, data transmission is scheduled to an available physical link with the same Quality of Service (QoS) level as the current physical link. In step S514, an available physical link with the same QoS level as the current physical link is used to establish a VXLAN tunnel. In step S516, it is determined whether an available physical link with a lower QoS level than the current physical link exists. If it does, proceed to step S518; otherwise, return to step S502. In step S518, data transmission is scheduled to an available physical link with a lower QoS level than the current physical link. In step S520, an available physical link with a lower QoS level than the current physical link is used to establish a VXLAN tunnel.

[0060] It should be understood that, similar to Figure 4 , Figure 5 The physical link dynamic adjustment process 500 shown is merely an example and not a limitation of this disclosure. For example, although Figure 5 The illustrated physical link dynamic adjustment process 500 is based on a VXLAN network, but the physical link dynamic adjustment process according to embodiments of this disclosure can be applied to networks similar to VXLAN networks. For example, although in Figure 5 In this context, determining whether a physical link is congested is done in response to a decrease in the link quality of the physical link used for data transmission. However, in other embodiments, the congestion of the physical link can be determined directly based on the monitored link quality.

[0061] The physical link dynamic adjustment process described above is a self-managed adjustment of physical link resources by the physical network. In some embodiments, the quality of service requirements of services in the logical network may change. However, current physical link resources are based on static configurations, such as fixed bandwidth resources, and therefore cannot perceive real-time changes in the service requirements of the logical network. This may lead to resource mismatch, such as idle communication channels in the logical network but congested physical links or wasted resource reservations in the physical network. Based on this, this disclosure proposes a method for achieving a "passive response" of physical link resources of the physical network to the needs of the logical network, so as to achieve a good fit between logical network resources and physical network resources.

[0062] Specifically, when the demands of the logical network undergo a predetermined change (e.g., a change in quality of service requirements meeting predetermined conditions), the logical network can declare its demands to the physical network. This allows the logical network to translate its dynamic demands (e.g., bandwidth bursts, priority changes, etc.) into standardized instructions understandable to the physical network. Predetermined changes might include the network controller (specifically, the part of the controller that controls the logical network) detecting a VNI traffic change reaching a threshold (e.g., a 50% traffic surge) or a new service deployment requirement. For example, the part of the controller that controls the logical network can interact with an API to translate the logical network's dynamic demands into standardized instructions understandable to the physical network. For instance, structured messaging in a JSON format can be used to transmit the logical network's dynamic demands to the physical network (i.e., the part of the controller that controls the physical network).

[0063] {

[0064] "request_id": "req_20251117001",

[0065] "timestamp": "2025-11-17T10:00:00Z",

[0066] "overlay_element": {

[0067] "vni": 100, / / Service identifier (VXLAN network identifier)

[0068] "vtep_pair": ["10.1.1.1", "10.1.1.2"] / / Source-Destination VTEP IP

[0069] },

[0070] "resource_demand": {

[0071] "type": "bandwidth", / / Resource type (bandwidth / queue depth / caching)

[0072] "value": "2Gbps", / / Demand (supports dynamic values ​​such as "+50%)"

[0073] "duration": "10m", / / Duration (in minutes)

[0074] "priority": "gold" / / Business priority (predefined level)

[0075] }

[0076] }

[0077] After the controller's physical network control section receives a request from the logical network, it can perform resource scheduling adjustments to the physical links based on the request. For example, the controller's physical network control section can determine the link quality of the current physical link used for transmitting data with changed quality of service (QoS) requirements, and, based on the mismatch between the changed QoS requirements and the current physical link's link quality, change the current physical link to an available physical link whose link quality matches the changed QoS requirements.

[0078] For example, suppose there are three physical links between the first and second logical network entities: physical link 1, physical link 2, and physical link 3, with link quality labels of medium, high, and low service quality, respectively. When establishing a communication channel between the first and second logical network entities, the service quality requirement of the data to be transmitted (e.g., the data corresponding to VNI=100 in the above JSON format example) is medium service quality. Therefore, physical link 1, with a link quality label of medium service quality, is selected to establish the communication channel between the first and second logical network entities. After a period of data transmission, the service quality requirement of the data to be transmitted changes to high service quality. At this time, the change in service quality requirement (i.e., the changed service quality requirement) can be transmitted from the logical network to the physical network through API interaction, as shown above. After receiving the message of the changed service quality requirement, if physical link 2, which matches the changed service quality requirement, is available, the physical link used for transmitting the data with the changed service quality requirement (e.g., the data corresponding to VNI=100) can be changed from physical link 1 to physical link 2.

[0079] Alternatively, after receiving the request from the logical network, the controller's physical network portion can adjust the resource scheduling of physical links based on the utilization of each physical link. For example, this can include determining the utilization of each physical link between the first and second logical network entities. If there are physical links with utilization below a first threshold (e.g., 30%) and the current physical link used for data transmission due to changed quality of service requirements is not the least utilized physical link, then the current physical link can be changed to the least utilized physical link.

[0080] If no physical links have utilization rates below the first threshold, it can be determined whether the utilization rates of all physical links are greater than a second threshold (e.g., 80%). If the utilization rates of all physical links are greater than the second threshold, a portion of the bandwidth used for data transmission with low quality of service (QoS) requirements can be used for data transmission with high QoS requirements. This can further include determining the QoS requirements of the data based on the QoS requirements of the data. The QoS requirements of the data can be determined using a method similar to that used to determine the QoS requirements of physical links. For example, the QoS requirements of the data can be determined as high QoS requirements, medium QoS requirements, and low QoS requirements based on the QoS requirements of the data. In this embodiment, a portion of the bandwidth used for data transmission with medium and low QoS requirements can be used for data transmission with high QoS requirements. A portion of the bandwidth used for data transmission with low QoS requirements can be used for data transmission with medium and / or high QoS requirements. Testing has shown that setting different utilization thresholds and adjusting the resource scheduling of physical links according to embodiments of this disclosure can achieve a very good fit between logical network resources and physical network resources, greatly improving the overall data transmission performance of the network, such as the total throughput of network data transmission.

[0081] Furthermore, in this disclosure, the proportion of bandwidth available for data transmission of different low-quality-of-service (QoS) levels that can be used for data transmission of the same high-QoS level can be the same or different. For example, suppose the QoS requirements of data are divided into three QoS levels: high, medium, and low. And suppose the proportion of bandwidth available for high-QoS data transmission of medium-QoS data is s1, while the proportion of bandwidth available for high-QoS data transmission of low-QoS data is s2. In one embodiment, s1 can be the same as s2. In another embodiment, s1 can be different from s2; for example, s1 can be less than s2. For example, s1 can be 50%, and s2 can be 80%. Additionally or alternatively, the proportion of bandwidth available for different high-QoS levels of data transmission of the same low-QoS level can be the same or different. As described above, for example, suppose the service quality level requirements for data are divided into three service quality levels: high, medium, and low. And suppose that the proportion of bandwidth available for medium service quality data transmission within the bandwidth used for low service quality data is s3, while the proportion available for high service quality data transmission within the bandwidth used for low service quality data transmission is s4. In one embodiment, s3 can be the same as s4. In another embodiment, s3 can be different from s4; for example, s3 can be less than s4. For instance, s3 can be 60%, and s4 can be 80%. The different proportions of bandwidth available for the same high service quality level data transmission within the bandwidth used for different low service quality level requirements, and / or the different proportions available for different high service quality level data transmission within the bandwidth used for the same low service quality level requirement, can ensure that low service quality level data can also be transmitted without being blocked, while still allowing high service quality level data transmission. This promotes the transmission of data with various service quality levels and improves the user experience for server tenants.

[0082] After the physical network resource scheduling adjustments have been determined as described above, the network controller, such as the BGP EVPN acting as the network control plane in a VXLAN network, can dynamically adjust the physical link connections of the physical network by modifying the next-hop attribute of the routes. For example, in a VXLAN network, this can include the following steps:

[0083] Step 1: Calculate the target path

[0084] The controller determines, based on the resource scheduling adjustment method described above, that data transmission needs to be scheduled to the target physical link. For example, it switches traffic with VNI=100 (i.e., data) from physical link A (high load) to physical link B (low load). Physical link B corresponds to one or more VTEP addresses, such as the IP address of the Spine node: 10.2.1.1.

[0085] Step 2: Generate BGP EVPN route update

[0086] The controller sends BGP EVPN update messages to VTEP devices (such as leaf switches in the physical network) via NETCONF to modify the next hop of the VNI, for example, by modifying the routing attributes of the VNI as follows:

[0087] - Original EVPN route: "VNI=100, MAC / IP=00:11:22:33:44:55 / 10.1.1.1, Next-Hop=10.1.1.1"

[0088] - Updated route: "VNI=100, MAC / IP=00:11:22:33:44:55 / 10.1.1.1, Next-Hop=10.2.1.1"

[0089] - By extending the community attribute of BGP Quality of Service (QoS) requirements, the QoS requirements of traffic are marked, enabling the selection of physical links that match the QoS requirements to establish VXLAN tunnels.

[0090] Step 3: Traffic Switching

[0091] After receiving the route update, the source VTEP device modifies the encapsulated packet of VNI, for example, changing the destination IP of the encapsulated packet of VNI 100 from 10.1.1.1 to 10.2.1.1, so that the packet can be forwarded through physical link B.

[0092] To better understand this disclosure, Figure 6 An example flowchart illustrates a physical link resource scheduling adjustment process 600 in a VXLAN network according to an embodiment of the present disclosure, based on changes in the quality of service (QoS) requirements of logical network-based data. Figure 6The process 600 can include three parts: 610: physical link resource scheduling decision, 620: physical link determination and route update, and 630: physical link switching. The physical link resource scheduling decision 610 can include steps S610-2 to S610-14. In step S610-2, the controller's control of the physical network portion receives the Quality of Service (QoS) requirement, i.e., the QoS requirement after the logical network change. In step S610-4, link quality data for each physical link is collected. In step S610-6, it is determined whether there is a physical link between the source VTEP and the destination VTEP with a utilization rate lower than 30%. If so, proceed to step S610-8; otherwise, proceed to step S610-10. In step S610-8, the physical link with the lowest utilization rate is selected as the target physical link. In step S610-10, it is determined whether the utilization rate of all physical links between the source VTEP and the destination VTEP is greater than 80%. If the value is greater than the specified value, proceed to step S610-12; otherwise, proceed to step S610-14. In step S610-12, services with low quality-of-service (QoS) requirements release bandwidth resources so that a portion of the bandwidth used by data transmission with low QoS requirements is used for data transmission with high QoS requirements. In step S610-14, it is determined that no resource scheduling adjustment is needed. Physical link determination and route update 620 includes steps S620-2 to S620-6. In step S620-2, the target physical network of the physical network is calculated and the next-hop VTEP address is determined. In step S620-4, a BGP EVPN route update is generated by modifying the next-hop address and QoS requirement extended community attribute. In step S620-6, the route update is sent to the source VTEP. Physical link switching 630 includes steps S630-2 and S630-4. In step S630-2, the source VTEP updates its forwarding table entries to switch data transmission to the new physical link. Afterward, scheduling is complete (S630-4).

[0093] It should be understood that, similar to Figure 4 and Figure 5 , Figure 6 The physical link resource scheduling adjustment process 600 shown is merely an example and not a limitation of this disclosure. For example, although Figure 6 The physical link resource scheduling adjustment process 600 shown is based on a VXLAN network, but the physical link resource scheduling adjustment process according to embodiments of this disclosure can be applied to networks similar to VXLAN networks. For example, although in Figure 6In step S610-6, it is determined whether there is a physical link between the source VTEP and the destination VTEP with a utilization rate lower than 30%. However, in other embodiments, it can be determined whether there is a physical link between the source VTEP and the destination VTEP with a utilization rate lower than other utilization thresholds (e.g., 20%). For example, although... Figure 6 Physical link resource scheduling and adjustment are based on physical link utilization. However, in other embodiments, physical link resource scheduling and adjustment can be based on the quality of service (QoS) requirements of the logical network. For example, if the link quality of the current physical link used for data transmission does not match the QoS requirements of the logical network, the current physical link can be adjusted to a physical link that matches the QoS requirements of the logical network.

[0094] In the foregoing, this disclosure combines Figures 2-6 A network method according to embodiments of the present disclosure is described. The network method according to embodiments of the present disclosure uses the link quality of each physical link connecting each logical network node to mark each physical link for link quality. When establishing a communication channel between any two logical network nodes, the quality of service (QoS) requirements of the data to be transmitted between these two logical network nodes are appended to the message used to establish the communication channel. This allows the establishment of the communication channel by selecting a physical link among multiple physical links between the two logical network nodes whose link quality matches the QoS requirements of the data to be transmitted, based on the QoS mark of the physical link and the QoS requirements in the message. In this way, the QoS requirements of logical network services can be matched with the actual link quality of the physical links of the physical network, achieving cross-network awareness between the logical and physical networks and improving network data transmission performance. For example, it ensures that critical services (such as financial transactions, real-time video, etc.) always operate on physical links that meet their QoS requirements. Furthermore, when changes in the service requirements of the logical network meet predetermined conditions, the logical network can declare its requirements to the physical network, transmitting the dynamic requirements of the logical network to the physical network. The physical network can adjust physical link resource scheduling based on this demand, achieving a good fit between logical and physical network resources and realizing a "demand-response" closed loop for cross-network resource scheduling without modifying the underlying forwarding device hardware. The network method of this disclosure forms a complete "perception-decision-execution" closed loop, solving the problem of disconnect between logical and physical network resources in existing VXLAN network architectures. Furthermore, the network method according to embodiments of this disclosure achieves intelligent scheduling effects similar to artificial intelligence (AI) (e.g., neural networks) through simple rules, avoiding the black-box problem of algorithms and improving operational transparency.

[0095] In the following text, this disclosure will combine Figure 7This describes network devices, network apparatuses, computer-readable storage media, and computer program products according to embodiments of the present disclosure.

[0096] Figure 7 An example configuration of a network device 700 according to an embodiment of the present disclosure is shown. The network device 700 may include a processor, i.e., a central processing unit (CPU) 720, and at least one memory 730. Additionally, the network device 700 may also include a transceiver 710. The memory 730 may store instructions. The transceiver 710 may be used to receive / transmit signals to implement data transmission. The processor 720 may be configured to execute the instructions stored in the memory 730 to perform the methods described in this disclosure. For example, when the network device 700 operates as a controller of a network comprising a plurality of logical network nodes, the processor 720 may execute the instructions stored in the memory 730 to: determine the link quality of each physical link between a first logical network node and a second logical network node among the plurality of logical network nodes; and to tag each physical link based on the determined link quality of each physical link. For example, when network device 700 operates as a first logical network node in a plurality of logical network nodes, processor 720 can execute instructions stored in memory 730 to: send a message for establishing a communication channel between the first and second logical network nodes, the message including information indicating the quality of service (QoS) requirements of the data to be transmitted; and use the established communication channel to transmit data. For example, when network device 700 operates as a second logical network node in a plurality of logical network nodes, processor 720 can execute instructions stored in memory 730 to: receive a message for establishing a communication channel between the first and second logical network nodes, the message including information indicating the QoS requirements of the data to be transmitted; and based on the QoS requirements, select at least one physical link between the first and second logical network nodes whose link quality matches the QoS requirements to establish a communication channel between the first and second logical network nodes. Furthermore, processor 720 can also be configured to execute the instructions referenced above. Figures 2-6 Other operations described are acceptable as long as there are no contradictions between them.

[0097] It should be understood that Figure 7 The network device configurations described herein are merely examples and not intended to limit the scope of this disclosure. The network device configurations in this disclosure may include, but are not limited to, those described above. Figure 7 The number of components in the component may be more or less.

[0098] Furthermore, this disclosure also provides a network apparatus including components for implementing steps of a network method according to any embodiment of this disclosure. For example, when the network apparatus operates as a controller of a network comprising a plurality of logical network nodes, the network apparatus may include: components for determining the link quality of each physical link between a first logical network node and a second logical network node among the plurality of logical network nodes; and components for tagging the physical links based on the determined link quality of each physical link. As another example, when the network apparatus operates as a first logical network node in a network comprising a plurality of logical network nodes, the network apparatus may include: components for sending a message for establishing a communication channel between the first logical network node and the second logical network node, the message including information indicating quality of service requirements for data to be transmitted; and components for using the established communication channel to perform data transmission. For example, when the network device operates as the second logical network node in a plurality of logical network nodes, the network device may include: a component for receiving a message for establishing a communication channel between the first logical network node and the second logical network node, the message including information indicating the quality of service (QoS) requirements of the data to be transmitted; and a component for selecting, based on the QoS requirements, at least one physical link between the first logical network node and the second logical network node whose link quality matches the QoS requirements to establish the communication channel between the first logical network node and the second logical network node.

[0099] Furthermore, this disclosure also provides a non-transitory computer-readable storage medium having instructions stored thereon that, when executed by a processor, can implement a network method according to any embodiment of this disclosure.

[0100] Furthermore, this disclosure also provides a computer program product containing stored instructions that, when executed by a processor, can implement a network method according to any embodiment of this disclosure.

[0101] This disclosure has now been combined with Figures 2-7 This disclosure describes network methods, network devices, network apparatuses, and computer-readable storage media and program products for network methods according to embodiments of the present disclosure. It should be noted that the above descriptions are merely some embodiments of the present disclosure and illustrations of the technical principles employed. For example, the formulas involved in this disclosure are merely examples and not limitations. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this disclosure.

[0102] Furthermore, while the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. In certain environments, multitasking and parallel processing may be advantageous. Similarly, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

[0103] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. A method for a network comprising a plurality of logical network nodes connected via a physical network, the method being executed by a network controller and comprising: Determine the link quality of each physical link between the first logical network node and the second logical network node among the plurality of logical network nodes; as well as Based on the determined link quality of each physical link, the link quality of each physical link is marked.

2. The method according to claim 1, wherein, Determining the link quality of each physical link includes determining at least one of the following for each physical link: Link availability, Data transmission latency of the link, Data transmission jitter in the link. Packet loss rate of data transmission on the link. Link bandwidth, Data throughput of the link Link utilization.

3. The method according to claim 1, wherein, Based on the link quality of each physical link, the link quality labeling of each physical link includes: Based on the link quality of each physical link, determine the quality of service (QoS) level of each physical link; and The corresponding physical links are labeled using the determined quality of service level.

4. The method according to claim 1, further comprising: Monitor the link quality of the current physical link used for data transmission between the first logical network node and the second logical network node; Determine whether the link quality of the current physical link is degraded relative to the marked link quality; as well as In response to a decrease in the link quality of the current physical link, a new physical link is determined for data transmission between the first logical network node and the second logical network node.

5. The method according to claim 4, wherein, Each physical link is marked by its quality of service (QoS) level, and the determination of the new physical link includes: Determine whether the current physical link is congested; In response to the current physical link congestion, an available physical link with the same quality of service (QoS) level as the current physical link is identified as the new physical link; and In response to the current physical link being congested and the absence of an available physical link, an available physical link with a lower quality of service (QoS) level than the current physical link is identified as the new physical link.

6. The method according to claim 4, further comprising: The link quality label of the current physical link is updated using the reduced link quality of the current physical link.

7. The method according to claim 1, wherein, The network controller includes a portion controlling the physical network and a portion controlling the logical network. The method further includes: in response to receiving the changed quality of service (QoS) requirements of the data to be transmitted between the first logical network node and the second logical network node, the portion controlling the logical network transmits the changed QoS requirements to the portion controlling the physical network.

8. The method according to claim 7, wherein, The physical link resource scheduling adjustment includes: Based on the mismatch between the changed quality of service (QoS) requirements and the link quality of the current physical link used for transmitting data with the changed QoS requirements, the current physical link is changed to an available physical link whose link quality matches the changed QoS requirements.

9. The method according to claim 7, wherein, The resource scheduling adjustment of the physical link includes: Determine the utilization rate of each physical link; In response to the existence of a physical link with a utilization rate lower than a first threshold among the physical links, the current physical link is changed to the physical link with the lowest utilization rate among the physical links.

10. The method of claim 9, further comprising: In response to the absence of physical links with utilization rates below the first threshold, it is determined whether the utilization rate of each physical link is greater than a second threshold, wherein the second threshold is greater than the first threshold; and In response to the fact that the utilization rate of each physical link is greater than the second threshold, a portion of the bandwidth used for data transmission with low quality of service (QoS) requirements is used for data transmission with high QoS requirements.

11. The method according to claim 10, wherein, The proportion of bandwidth available for data transmission with different low-quality-of-service (QoS) levels differs from that available for data transmission with the same high QoS level. The proportion of bandwidth available for data transmission with the same low quality of service (QoS) level requirement can differ depending on the QoS level requirement.

12. A method for a network comprising a plurality of logical network nodes, the plurality of logical network nodes being connected via a physical network, a first logical network node and a second logical network node of the plurality of logical network nodes being connected via at least one physical link, the method being performed by the first logical network node and comprising: Send a message for establishing a communication channel between the first logical network node and the second logical network node, the message including information indicating the quality of service requirements of the data to be transmitted; as well as Data transmission is performed using an established communication channel, which is established based on a physical link whose link quality matches the service quality requirement among the at least one physical link, and each physical link among the at least one physical link is marked by its link quality.

13. The method according to claim 12, wherein, The quality of service (QoS) requirements for the data are indicated by the QoS requirement extended community attribute in the path attribute of the Border Gateway Protocol (BGP) update message.

14. The method according to claim 13, wherein, The extended community attribute of the service quality requirement is indicated by information carried by at least one of the latency field, jitter field, packet loss rate field, and bandwidth field.

15. A method for a network comprising a plurality of logical network nodes, the plurality of logical network nodes being connected via a physical network, a first logical network node and a second logical network node of the plurality of logical network nodes being connected via at least one physical link, the method being performed by the second logical network node and comprising: Receive a message for establishing a communication channel between the first logical network node and the second logical network node, the message including information indicating the quality of service requirements of the data to be transmitted; Based on the service quality requirements, a communication channel between the first logical network node and the second logical network node is established by selecting a physical link whose link quality matches the service quality requirements from the at least one physical link. Each physical link in the at least one physical link is marked with its link quality.

16. The method according to claim 15, wherein, The quality of service (QoS) requirements for the data are indicated by the QoS requirement extended community attribute in the path attribute of the Border Gateway Protocol (BGP) update message.

17. The method according to claim 16, wherein, The extended community attribute of the service quality requirement is indicated by information carried by at least one of the latency field, jitter field, packet loss rate field, and bandwidth field.

18. A network device, comprising: Memory for storing instructions, and A processor that executes the instructions to implement the method according to any one of claims 1-17.

19. A network device, comprising: Components for performing the steps of the method according to any one of claims 1-17.

20. A computer-readable storage medium having instructions stored thereon that, when executed by a processor, implement the method according to any one of claims 1-17.

Images