Data sending method, network access system, device, medium, and program product

Abstract

A data sending method, a network access system, a device, a medium, and a program product. The system comprises a plurality of physical servers and a plurality of network access devices, wherein a unique global network address range is allocated to the physical servers, and global network addresses of network interface controllers and container instances deployed on the physical servers belong to the global network address range; the network interface controllers are connected to the plurality of network access devices; a plurality of routing table entries are configured for the network access devices, and indicate a mapping relationship among the global network address range, next-hop link local addresses, and interface numbers of the network access devices; and the link local addresses of the network access devices and the link local addresses of the network interface controllers belong to a same network address range.

Description

Data transmission method, network access system, device, medium and program product Cross-reference to related applications

[0001] The present disclosure claims priority to Chinese Patent Application No. 202411328529.6, filed on September 23, 2024, the entire contents of which are incorporated herein by reference. TECHNICAL FIELD

[0002] The present disclosure relates to the field of network technology, in particular to a data transmission method, a network access system, an electronic device, a computer readable storage medium and a computer program product, which can be applied to the field of network deployment of cloud infrastructure. BACKGROUND

[0003] Due to the problem of insufficient network address resources of Internet Protocol Version 4 (IPv4), the application and development of the Internet are restricted, so the online deployment of Internet Protocol Version 6 (IPv6) has developed rapidly.

[0004] In the cloud native scenario, a user will deploy a large number of container instances on a single server, and according to the Central Processing Unit (CPU) and memory capacity of the server configuration, the number of container instances is usually in the order of tens to hundreds. According to the design scheme of IPv6 container network, each container of the server publishes a neighbor discovery message to the switch, and the switch learns all IPv6 network container addresses. If the server connected to the switch deploys a sufficient number of container instances, the number of address entries learned on the switch will easily exceed the upper limit supported by the switch chip, resulting in the technical problem of insufficient capacity of the switch chip. SUMMARY

[0005] The embodiments of the present disclosure provide a data transmission method, a network access system, an electronic device, a computer readable storage medium and a computer program product to alleviate or solve one or more technical problems in the prior art.

[0006] In a first aspect, the embodiments of the present disclosure provide a data sending method, applied to a network access device, a plurality of device interfaces of the network access device are connected with network interface controllers of a plurality of physical servers one by one, the physical servers are assigned a unique global network address segment, and a link-local address of the network access device and a link-local address of the network interface controller belong to a same network address segment; the method comprises: determining a target control address and a target interface number corresponding to a destination address of target data from a plurality of routing table entries, a plurality of neighbor discovery table entries and a plurality of control address table entries according to the destination address, wherein the plurality of routing table entries represent a mapping relationship between the global network address segment of the physical server, a link-local address of the network interface controller corresponding to a next hop and an interface number of the device interface, the plurality of neighbor discovery table entries represent a mapping relationship between the link-local address of the network interface controller and a control address and the interface number of the device interface, and the plurality of control address table entries represent a mapping relationship between the control address of the network interface controller and the interface number of the device interface; and sending the target data to the network interface controller corresponding to the target control address through a device interface corresponding to the target interface number.

[0007] In a second aspect, the embodiments of the present disclosure provide a data sending method, applied to a physical server, the physical server is assigned a unique global network address segment, a network interface controller and a plurality of container instances are deployed on the physical server, and global network addresses of the network interface controller and the container instances belong to the global network address segment; the method comprises: in response to a sending request of target data of any container instance, obtaining a routing table entry pre-configured for the physical server, the routing table entry of the physical server indicating that a next hop address of the physical server is a link-local address of a network access device connected with the network interface controller, and an out-interface is the network interface controller, wherein the link-local address of the network access device and the link-local address of the network interface controller belong to a same network address segment; and sending the target data to the link-local address in the routing table entry through the network interface controller.

[0008] In a third aspect, the embodiments of the present disclosure provide a network access system, comprising a plurality of physical servers and a plurality of network access devices, the physical servers are assigned with a unique global network address segment, the global network addresses of the network interface controllers and the plurality of container instances deployed on the physical servers belong to the global network address segment; the network interface controllers of the physical servers are connected to the plurality of network access devices; the network access devices have a plurality of device interfaces, the plurality of device interfaces are connected to the network interface controllers of the plurality of physical servers one by one, the network access devices are preconfigured with a plurality of routing table entries, used to represent the mapping relationship between the global network address segment of the physical servers, the link local address of the next hop network interface controller and the interface number of the device interface, the link local address of the network access device and the link local address of the network interface controller belong to the same network address segment; and the physical servers are preconfigured with routing table entries, used to represent that the next hop address of the physical servers is the link local address of the network access device connected by the physical servers.

[0009] In a fourth aspect, the embodiments of the present disclosure provide an electronic device, comprising a memory, a processor and a computer program stored in the memory, the processor implements the method of any one of the embodiments of the present disclosure when executing the computer program.

[0010] In a fifth aspect, the embodiments of the present disclosure provide a computer readable storage medium, the computer readable storage medium stores a computer program, the computer program is executed by a processor to implement the method of any one of the embodiments of the present disclosure.

[0011] In a sixth aspect, the embodiments of the present disclosure provide a computer program product, comprising a computer program, the computer program is executed by a processor to implement the method of any one of the embodiments of the present disclosure.

[0012] According to the technical scheme of the embodiment of the present disclosure, a unique global network (Global-IP) address segment is allocated to a physical server, and a Global-IP address is allocated to a network interface controller and each container instance deployed on the physical server from the Global-IP address segment, and at the same time, the link local (Linklocal-IP) address of the network interface controller connected to the network access device is configured as the same address segment as the Linklocal-IP address, so that the static routing table items required for L3 forwarding can be configured in the physical server and the network access device. Further, the static route with the next hop as the Linklocal-IP address is configured on the physical server and the network access device, respectively, to realize L3 forwarding of data, so that the network access device needs to learn all IPv6 address corresponding NDP table items and MAC table items on the connected physical server, and the technical problem of insufficient capacity of the NDP / MAC table items in the network access device in the case of large deployment of container instances is solved.

[0013] Optionally, the physical server periodically sends an NS packet containing the Linklocal-IP address and the MAC address of the network interface controller. On the one hand, the dynamic learning capability of the NDP table items and the MAC table items of the network access device itself can be utilized to make the network access device always have the NDP table items and the MAC table items corresponding to the Linklocal-IP address of the network interface controller, so that even if the network access device temporarily does not have the corresponding NDP table items and MAC table items, the network access device can learn the NDP table items and the MAC table items after the physical server periodically sends the NS packet, and the traffic forwarding is restored. On the other hand, the NS packet is sent by the physical server side, and if the physical server abnormally, the sending event of the NS packet will naturally stop, so that there is no need for an additional scheme to detect the invalidation event of the NDP table items and the MAC table items.

[0014] The above description is only a summary of the technical scheme of the present disclosure, in order to more clearly understand the technical means of the present disclosure, the content of the specification can be implemented, and in order to make the above and other purposes, features and advantages of the present disclosure more obvious and easy to understand, the specific embodiments of the present disclosure are described below. BRIEF DESCRIPTION OF DRAWINGS

[0015] In the drawings, like reference numerals refer to like elements throughout the various drawings. The drawings are not necessarily to scale, emphasis instead being placed on illustrating the principles of the present disclosure. It should be understood that the drawings are only illustrations of some embodiments of the present disclosure and should not be considered limiting of the scope of the present disclosure.

[0016] FIG. 1 shows a schematic diagram of a leaf-spine network architecture used in a data center;

[0017] FIG. 2 shows a deployment diagram of a server and a switch in the related art;

[0018] FIG. 3 shows a schematic diagram of a network architecture provided by an embodiment of the present disclosure;

[0019] FIG. 4 shows a deployment configuration schematic diagram of a network access system provided by an embodiment of the present disclosure;

[0020] FIG. 5 shows a flowchart of a data sending method on the physical server side according to an embodiment of the present disclosure;

[0021] FIG. 6 shows a flowchart of a data sending method on the network access device side according to an embodiment of the present disclosure;

[0022] FIG. 7 shows a block diagram of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0023] In the following, only certain exemplary embodiments are simply described. As those skilled in the art can recognize, the described embodiments can be modified in various different ways without departing from the concept or scope of the present disclosure. Therefore, the drawings and the description are considered to be exemplary in nature, rather than limiting.

[0024] To facilitate understanding of the technical solutions of the embodiments of the present disclosure, the related art of the embodiments of the present disclosure is described below. The following related art can be combined with the technical solutions of the embodiments of the present disclosure in any way as an optional solution, which all belong to the protection scope of the embodiments of the present disclosure.

[0025] In the following, the following terms will be used.

[0026] IPv6: Internet Protocol version 6, designed and developed by the Internet Engineering Task Force (IETF), used to replace IPv4.

[0027] Media Access Control Address (MAC): MAC address, which is a unique identifier of a network device, is a hardware address burned into a network interface controller (NIC) device by the manufacturer of the NIC device during production of the NIC device. Network interface controller: NIC, which is a hardware device, is installed in a computer or other device, used to connect to a network and communicate data. Each NIC has a unique MAC address.

[0028] Open System Interconnect (OSI) model: The OSI model is a network interconnection model that defines a seven-layer framework for network interconnection: Layer 1, which transmits raw bit streams; Layer 2, which provides point-to-point data transmission and handles MAC addresses and frames; Layer 3, which is responsible for path selection and logical address (IP address) processing; Layer 4, which provides end-to-end communication and handles port and flow control; Layer 5, which manages sessions and data exchange; Layer 6, which handles data format and encryption; and Layer 7, which provides network services and application program interfaces.

[0029] L2 forwarding (Layer 2 Forwarding): Forwarding operation performed at the data link layer (Layer 2) of the OSI model. L2 forwarding is mainly used for communication within a local area network (LAN), where all devices are in the same broadcast domain and mainly rely on MAC addresses to determine the forwarding path of data frames. Therefore, the implementation of L2 forwarding needs to be based on MAC address tables.

[0030] L3 forwarding (Layer 3 Forwarding): Forwarding operation performed at the network layer (Layer 3) of the OSI model, mainly based on the IP address of the device to determine the forwarding path of the data packet, and through the lookup of the routing table to determine the forwarding path of the data packet. L3 forwarding is used for communication across different network segments (i.e., cross-segment communication) or Internet communication.

[0031] Neighbor Discovery Protocol (NDP): A part of the TCP / IP (Transmission Control Protocol / Internet Protocol) protocol stack, used in IPv6. NDP is used to replace the Address Resolution Protocol (ARP) in IPv4. It works at the data link layer and is responsible for discovering other nodes and corresponding IP addresses on the link, determining available routes, and maintaining information about available paths and other active nodes.

[0032] Neighbor Discovery entry: A record in the Neighbor Discovery table, describing the neighbor information maintained by the Neighbor Discovery Protocol in the IPv6 network.

[0033] Control Address Table Entry: Abbreviated as MAC entry, it is a record in the MAC address forwarding table (MAC address table) that records each MAC address and its corresponding outgoing interface.

[0034] Neighbor Solicitation (NS) message: The NS message is a type of message defined in NDP. In NDP, the NS message is primarily used to request the discovery of a neighbor's data link layer MAC address based on the neighbor's IPv6 address. In other words, when a neighbor knows the IPv6 address of another neighbor, it can send an NS message to request its data link layer address.

[0035] Global-IP address: In IPv6, the Global-IP address is a globally routable IP address used to transmit data between different networks. It is a unique IPv6 address for a device on the global Internet.

[0036] Link-local IP address: In IPv6, a link-local IP address is an IPv6 address used within a local area network or the same link, and its scope is limited to the link locality.

[0037] Spine-Leaf architecture: This is a data center network architecture consisting of a Core layer, a Spine layer, and a Leaf layer. Leaf switches are responsible for connecting physical servers, and Spine switches connect Leaf switches, enabling communication between physical servers connected to different Leaf switches. To improve the reliability of physical server access to switches, data centers can adopt a multi-server access (i.e., multiple accesses or active-active access) networking approach. As shown in Figure 1, taking dual-server access as an example, two Leaf switches form a group, and each physical server uses two physical network interfaces to connect to one group of two Leaf switches. Simultaneously, the two physical network interfaces use link aggregation (bonding) technology to form a single logical network interface for easier management.

[0038] In the related art, when two physical servers are connected to two Leaf switches, a network administrator allocates a same IP address segment to the two Leaf switches connected to the physical servers, and configures a gateway IP of the IP address segment on the two Leaf switches. The physical servers connected to the Leaf switches and the container instances deployed on the physical servers are allocated IP addresses from the IP address segment and belong to a same broadcast domain. The Leaf switches learn NDP table items and MAC table items of all the physical servers and all the container instances. Generally, the capacity of the NDP table items and the MAC table items of a commercial Ethernet switch chip is generally 10K-100K, and when the IPv6 address segment mask configured on the switch is less than 110, if a sufficient number of container instances are deployed on the physical servers connected to the switch, the number of the learned NDP table items and the MAC table items on the Leaf switch will easily exceed the upper limit supported by the switch chip.

[0039] Taking the physical server 1, the physical server 2, the Leaf switch 1 and the Leaf switch 2 in FIG. 1 as an example, the physical server 1 and the physical server 2 are connected to the Leaf switch 1 and the Leaf switch 2, and the deployment configuration is as follows.

[0040] (1) The Leaf switch 1 and the Leaf switch 2 are both configured with a same IPv6 address 2001::ffff / 64 as a gateway IP.

[0041] (2) The network card 1 of the physical server 1 is configured with an IPv6 address 2001::1 / 64, and the network card 2 of the physical server 2 is configured with an IPv6 address 2001::2:1 / 64.

[0042] (3) The physical server 1 is deployed with n container instances, and needs to be allocated n IPv6 addresses: 2001::11 / 64-2001::x / 64; the physical server 2 is deployed with m container instances, and needs to be allocated m IPv6 addresses: 2001::2:1 / 64-2001:2::y / 64, wherein n and m are integers greater than 1.

[0043] Leaf switch 1 and Leaf switch 2 perform L2 forwarding, that is, in the process of data forwarding, the NDP table item needs to be looked up to determine the mapping relationship between the IPv6 address and the MAC address, and the MAC table item needs to be further looked up to determine the mapping relationship between the MAC address and the device interface (including interface 1 and interface 2) of the switch, therefore, Leaf switch 1 and Leaf switch 2 need to learn the NDP table item and the MAC table item corresponding to network card 1, the NDP table item and the MAC table item corresponding to network card 2, the NDP table item and the MAC table item corresponding to the n container instances on physical server 1, and the NDP table item and the MAC table item corresponding to the m container instances on physical server 2. The specific process is as follows.

[0044] (1) Leaf switch 1 and Leaf switch 2 each need to learn 2+n+m NDP table items: 2001::1 is at mac_card1 from interface 1; 2001::2:1 is at mac_card2 from interface 2; 2001::11 is at mac_container11 from interface 1; 2001::2:11 is at mac_container21 from interface 2; ……; 2001::x is at mac_containern from interface 1; 2001::2:y is at mac_containerm from interface 2.

[0045] (2) Leaf switch 1 and Leaf switch 2 each need to learn 2+n+m MAC table items: mac_card1 out interface 1; mac_card2 out interface 2; mac_container11 out interface 1; mac_container21 out interface 2; ……; mac_containern out interface 1; mac_containerm out interface 2.

[0046] Since the IPv6 network segment mask length configured on the Leaf switch is 64, the maximum number of IPv6 addresses that can be allocated in this network segment is 264, which far exceeds the NDP / MAC table item capacity of the switch chip. Therefore, if the number of container instances deployed on physical server 1 and physical server 2 is large enough, it will exceed the NDP / MAC table item capacity of the switch chip, resulting in abnormal forwarding.

[0047] In a cloud-native (distributed cloud) scenario, a container instance is a basic running unit of a cloud-native application, which contains an application program and all its dependent environments. In order to meet the needs of cloud-native application in high concurrency, high reliability and elastic expansion, a large number of container instances need to be deployed and managed. In order to solve the technical problem of insufficient capacity of NDP / MAC table entries in the network access device in the case of a large number of deployed container instances, the embodiments of the present disclosure provide a data sending method, a network access system, an electronic device, a computer readable storage medium and a computer program product.

[0048] The technical solutions of the present disclosure and how the technical solutions of the present disclosure solve the foregoing technical problems will be described in detail below with specific embodiments. Several specific embodiments listed can be combined with each other, and the same or similar concepts or processes can not be described in detail in some embodiments. The embodiments of the present disclosure will be described in detail below with reference to the drawings.

[0049] FIG. 3 shows a schematic diagram of a network architecture according to an embodiment of the present disclosure. As shown in FIG. 3, the network architecture includes a plurality of network backbone devices and a plurality of network access systems, each of which includes a plurality of physical servers and a plurality of network access devices. The number of network backbone devices and network access systems is not specifically limited in the present disclosure, and the number of network access devices and physical servers in each network access system is not specifically limited in the present disclosure, and can be selected and deployed according to business needs. FIG. 3 takes four network backbone devices (network backbone device 1, network backbone device 2, network backbone device 3 and network backbone device 4), four physical servers (physical server 1, physical server 2, physical server 3 and physical server 4), four network access devices (network access device 1, network access device 2, network access device 3 and network access device 4) as an example, wherein network access system 1 includes physical server 1, physical server 2, network access device 1 and network access device 2; network access system 2 includes physical server 3, physical server 4, network access device 3 and network access device 4.

[0050] A network interface controller (NIC) is deployed on the physical server to convert the data of the physical server it is located in into the format required by the network protocol, and transmit and receive data through network media (such as Ethernet cable, optical fiber). A plurality of container instances are also deployed on the physical server to deploy, run and manage individual containerized application programs or microservices.

[0051] The network access device is used to connect the physical server to the network, realizing the physical and logical connection between the physical server and the network, so that data can be transmitted in the network. For example, it can include switches (such as Leaf switches), routers, etc., or other any applicable devices, which are not specifically limited in the embodiments of the present disclosure.

[0052] The network backbone device connects each network access device in each network access system, so that the physical servers connected under different network access devices can intercommunicate.

[0053] In one network access system, each physical server connects multiple network access devices in the network access system based on the network interface controllers thereon; each network access device has multiple device interfaces for connecting the multiple network interface controllers in the network access system one by one.

[0054] For example, as shown in FIG. 3, in the network access system 1: the physical server 1 connects the network access device 1 and the network access device 2 based on the network interface controller 1; the physical server 2 connects the network access device 1 and the network access device 2 based on the network interface controller 2. The network access device 1 has two device interfaces, i.e., interface 1 and interface 2, the interface 1 of the network access device 1 connects the network interface controller 1, and the interface 2 of the network access device 1 connects the network interface controller 2. The network access device 2 has two device interfaces, the interface 1 of the network access device 2 connects the network interface controller 1, and the interface 2 of the network access device 2 connects the network interface controller 2.

[0055] Similarly, in the network access system 2: the physical server 3 connects the network access device 3 and the network access device 4 based on the network interface controller 3; the physical server 4 connects the network access device 3 and the network access device 4 based on the network interface controller 4. The network access device 3 has two device interfaces, the interface 1 of the network access device 3 connects the network interface controller 3, and the interface 2 of the network access device 3 connects the network interface controller 4; the network access device 4 has two device interfaces, the interface 1 of the network access device 4 connects the network interface controller 3, and the interface 2 of the network access device 4 connects the network interface controller 4.

[0056] For ease of understanding, the network access system 1 is taken as an example below to introduce the deployment and configuration of the network access system in the embodiments of the present disclosure in combination with FIG. 3 and FIG. 4. In FIG. 4, the network access device is exemplified as a Leaf switch, and the network interface controller is exemplified as a network card.

[0057] As shown in FIG. 3 and FIG. 4, the physical servers are assigned with unique Global-IP address segments, and the Global-IP addresses of the network interface controllers and the multiple container instances deployed on the physical servers belong to the Global-IP address segments. Exemplarily, each physical server can be assigned with an independent Global-IP address segment according to the maximum number of container instances that need to be deployed on the physical server, and then the Global-IP addresses of the network interface controllers and the container instances deployed thereon are assigned from the Global-IP address segment.

[0058] For example, as shown in FIG. 4, the physical server 1 is allocated a Global-IP address segment of 2001 :: 1 / 64, the network interface controller of the physical server 1 is a network card 1, the Global-IP address allocated for the network card 1 is 2001 :: 1 / 64, and the physical server 1 is deployed with n container instances, which are exemplified as container 11, container 12, container 13, …, and container n, and the corresponding Global-IP addresses are 2001 :: 11 / 64, 2001 :: 12 / 64, 2001 :: 13 / 64, …, and 2001 :: x / 64, respectively. Here, n is an integer greater than 1.

[0059] The physical server 2 is allocated a Global-IP address segment of 2002 :: 1 / 64, the network interface controller of the physical server 2 is a network card 2, the Global-IP address allocated for the network card 2 is 2002 :: 1 / 64, and the physical server 2 is deployed with m container instances, which are exemplified as container 21, container 22, container 23, …, and container m, and the corresponding Global-IP addresses are 2002 :: 21 / 64, 2002 :: 22 / 64, 2002 :: 23 / 64, …, and 2002 :: y / 64, respectively. Here, m is an integer greater than 1.

[0060] Further, the Linklocal-IP address of the network access device belongs to the same network address segment as the Linklocal-IP address of the network interface controller. That is, the device interface of each network interface controller connected to the network access device and the Linklocal-IP address of the connected network interface controller belong to the same network address segment.

[0061] For example, as shown in FIG. 4, the network card 1 is allocated a Linklocal-IP address of fe80 :: 1 / 64, which is in the same network address segment as the Linklocal-IP address fe80 :: ffff / 64 of the Leaf switch 1. The network card 2 is allocated a Linklocal-IP address of fe80 :: 1 / 64, which is in the same network address segment as the Linklocal-IP address fe80 :: ffff / 64 of the Leaf switch 1. As can be seen, in the same network access system, the Linklocal-IP addresses of the two network access devices are the same, and the Linklocal-IP address is the gateway IP of the network access device.

[0062] The static route with the Linklocal-IP address as the next hop is configured on the physical server and the network access device respectively. Specifically, the network access device is pre-configured with a plurality of routing table entries for representing the mapping relationship between the global network address segment of the physical server, the link local address of the network interface controller corresponding to the next hop, and the interface number of the device interface, and the link local address of the network access device and the link local address of the network interface controller belong to the same network address segment; and the physical server is pre-configured with a routing table entry for representing that the next hop address of the physical server is the link local address of the network access device connected by the network interface controller of the physical server, and the out interface is the network interface controller of the physical server.

[0063] In the embodiments of the present disclosure, the pre-configuration can be understood as static configuration, for example, pre-configuring the routing table entry, that is, configuring the static routing table entry. The next hop address of the physical server can be understood as the next hop address of the traffic flowing out of the physical server, or the next hop address of the data transmission on the physical server.

[0064] For example, as shown in FIG. 4, the static default route is configured on the physical server 1, the next hop is the Linklocal-IP address fe80::ffff of the Leaf switch, the out interface is the network card 1, and the routing table entry example is: default next hop fe80::ffff out_if network card 1. Similarly, the static default route is configured on the physical server 2, the next hop is the Linklocal-IP address fe80::ffff of the Leaf switch, the out interface is the network card 2, and the routing table entry example is: default next hop fe80::ffff out_if network card 2. Therefore, the default next hop of the data sent by each container instance on the physical server is the Linklocal-IP address of the network access device.

[0065] The static route is configured on the Leaf switch 1 and the Leaf switch 2, including: (1) the routing table entry corresponding to the physical server 1, the static route points to the Global-IP address segment connected to the physical server 1, the next hop is the Linklocal-IP address of the network card 1, and the out interface is the interface 1 of the Leaf switch; (2) the routing table entry corresponding to the physical server 2, the static route points to the Global-IP address segment connected to the physical server 2, the next hop is the Linklocal-IP address of the network card 2, and the out interface is the interface 2 of the Leaf switch.

[0066] Based on this, no matter how many container instances are deployed on the physical server, when the network access device forwards data to the corresponding physical server, it is L3 forwarding according to the configured static route, therefore, the network access device only needs to learn the NDP table item and the MAC table item corresponding to the network interface controller connected to the physical server, and no longer needs to learn the NDP table item and the MAC table item corresponding to all container instances.

[0067] For example, as shown in FIG. 4, Leaf switch 1 is connected to two physical servers, and only needs to learn the NDP table item and the MAC table item corresponding to the two physical servers. For physical server 1, an NDP table item and a MAC table item are learned. An example of the NDP table item corresponding to physical server 1 is: fe80::1 is at mac_card1 from interface 1, indicating that when Leaf switch 1 sends data to fe80::1 through interface 1, the MAC address of card 1 should be used as the target MAC address; an example of the MAC table item corresponding to physical server 1 is: mac_card1 out interface 1, indicating that if the target MAC address of the data is card 1, the data should be sent out from interface 1. Similarly, for physical server 2, an NDP table item and a MAC table item corresponding to physical server 2 are also learned. An example of the NDP table item corresponding to physical server 2 is: fe80::1 is at mac_card2 from interface 2, indicating that when Leaf switch 1 sends data to fe80::1 through interface 2, the MAC address of card 2 should be used as the target MAC address; an example of the MAC table item corresponding to physical server 2 is: mac_card2 out interface 2, indicating that if the target MAC address of the data is card 2, the data should be sent out from interface 2.

[0068] Based on the network access system provided by the embodiments of the present disclosure, by allocating a unique Global-IP address segment to the physical server, and allocating Global-IP addresses to the network interface controllers and container instances deployed on the physical server from the Global-IP address segment, and configuring the Linklocal-IP address of the network interface controller connected to the network access device and its Linklocal-IP address as the same address segment, it is possible to configure the static route table items required for L3 forwarding in the physical server and the network access device. Further, the static route with the next hop being the Linklocal-IP address is configured on the physical server and the network access device, L3 forwarding of data is realized, so that the network access device no longer needs to learn the NDP table items and the MAC table items corresponding to all IPv6 addresses on the physical server connected, and the technical problem of insufficient capacity of the NDP / MAC table items in the network access device in the case of large deployment of container instances is solved.

[0069] The following describes an implementation of data sending based on the network access system, taking a physical server and a network access device as an execution subject respectively.

[0070] FIG. 5 shows a flowchart of a data sending method according to an embodiment of the present disclosure. The data sending method can be applied to and executed by any physical server in the network access system described above. As shown in FIG. 5, the data sending method includes step S501 and step S502.

[0071] Step S501: In response to a sending request of target data from any container instance, a routing table item pre-configured for the physical server is obtained, the routing table item of the physical server indicating that a next hop address of the physical server is a link local address of a network access device connected to a network interface controller, and an out interface is the network interface controller, wherein the link local address of the network access device and the link local address of the network interface controller belong to the same network address segment.

[0072] The physical server can be any physical server in the network access system described above, that is, the physical server is deployed with a network interface controller and a plurality of container instances, and is assigned with a unique Global-IP address segment. The Global-IP addresses of the network interface controller and the container instances all belong to the Global-IP address segment, that is, the Global-IP addresses of the network interface controller and the plurality of container instances deployed on the physical server are all allocated from the Global-IP address segment. For details, refer to the description of the network access system.

[0073] For example, the physical server receives a sending request of target data from a container instance, and obtains a static routing table item configured therefor, that is, a configured static route, which is a default route, indicating that a next hop address of default traffic output of the physical server is a Linklocal-IP address of a network access device, and an out interface is a network interface controller of the physical server. Meanwhile, as described above, the Linklocal-IP address of the network access device and the Linklocal-IP address of the network interface controller of the physical server are in the same network address segment.

[0074] Step S502: The target data is sent to the link local address in the routing table item through the network interface controller.

[0075] As described above, in the same network access system, the link local addresses of the network access devices are the same, and when the physical servers in the network access system send target data to the link local addresses in the routing table item, the target data can be broadcasted to the network access devices in the network access system.

[0076] According to the data sending method, a unique Global-IP address segment is allocated to the physical server, and Global-IP addresses are allocated to the network interface controller and each container instance deployed on the physical server from the Global-IP address segment. Meanwhile, the Linklocal-IP address of the network interface controller connected to the network access device is configured as the same address segment as the Linklocal-IP address, so that the static route table items required for L3 forwarding can be configured in the physical server and the network access device. Further, the physical server is configured with a static route with a next hop of the Linklocal-IP address, so as to prepare for L3 forwarding of data on the physical server side, and the network access device can avoid learning all IPv6 address corresponding NDP table items and MAC table items on the physical server, thereby solving the technical problem of insufficient capacity of the NDP / MAC table items in the network access device in the case of large deployment of container instances.

[0077] In an embodiment, the network access device comprises a plurality of device interfaces, each device interface corresponding to an interface number, and the plurality of device interfaces comprises a target device interface connected to the network interface controller. The data sending method can comprise: sending a neighbor solicitation message to the network access device, the neighbor solicitation message carrying a link-local address and a control address of the network interface controller, so that the network access device generates a neighbor discovery table item and a control address table item corresponding to the network interface controller based on the neighbor solicitation message, wherein the neighbor discovery table item represents a matching relationship between the link-local address of the network interface controller, the control address of the network interface controller and the interface number of the target device interface, and the control address table item represents a matching relationship between the control address of the network interface controller and the interface number of the target device interface.

[0078] Exemplarily, the neighbor solicitation message is an NS message, the control address is a MAC address, the neighbor discovery table item is an NDP table item, and the control address table item is a MAC table item. The operating system kernel of the physical server is modified to enable the physical server to periodically send an NS message to the network access device connected thereto. The NS message carries the Linklocal-IP address and the MAC address of the network interface controller of the physical server,

[0079] If a network access device receives an NS message from the physical server, it will generate an NDP entry and a MAC entry accordingly. For example, as shown in Figure 4, physical server 1 periodically sends NS messages to Leaf switches (Leaf switch 1 and Leaf switch 2). These NS messages carry the Linklocal-IP address and MAC address of network interface card 1 (NIC 1). The Leaf switch receives the NS messages and learns an NDP entry and a MAC entry. The NDP entry for physical server 1, `fe80::1is at mac_NIC1frominterface1`, represents the matching relationship between NIC 1's Linklocal-IP address (fe80::1), NIC 1's MAC address (mac_NIC1), and Leaf switch 1's interface number (interface1). Specifically, when Leaf switch 1 sends target data to `fe80::1` through interface1, it should use NIC 1's MAC address as the target MAC address. The MAC entry for physical server 1, mac_network1outinterface1, indicates the matching relationship between the MAC address of network card 1 (mac_network1) and the interface number of Leaf switch 1 (interface1). Specifically, it can be understood that if the target MAC address of the target data is network card 1, then the target data should be forwarded out from interface1.

[0080] The physical server sends NS messages containing the Linklocal-IP address and MAC address of the network interface controller from multiple physical links. On one hand, it leverages the network access devices' existing ability to dynamically learn NDP and MAC entries, ensuring that the network access devices always have NDP and MAC entries corresponding to the Linklocal-IP address of the network interface controller. Therefore, even if there is a brief period where the network access devices lack the corresponding NDP and MAC entries, the physical server's periodic sending of NS messages will trigger the network access devices to learn the NDP and MAC entries, resuming traffic forwarding. On the other hand, since the NS messages are sent periodically by the physical server, if the physical server malfunctions, the NS message sending will naturally stop. Therefore, no additional scheme is needed to detect the failure of NDP and MAC entries.

[0081] In one implementation, the neighbor request message is sent at a frequency shorter than the aging period of the neighbor discovery table entries configured on the network access device.

[0082] The aging period of the NDP entry can be understood as a time limit for the NDP entry to remain valid. Each NDP entry has a specific aging period after being created. If no NS packet is received within the aging period, the NDP entry is considered invalid. Therefore, setting the sending period of the NS packet to be less than the aging period of the NDP entry can keep the NDP entry valid within the aging period. If an abnormality occurs in the physical server and no NS packet is sent, the NDP entry can be automatically invalidated, avoiding forwarding errors, so that no additional mechanism or component is needed to detect the invalidation of the NDP entry.

[0083] In an embodiment, the network access system can include a central control component capable of communicating with the physical servers and network access devices in the system. If a physical server is online, the central control component configures corresponding static NDP entries and MAC entries in the network access device connected to the physical server. If an abnormality such as offline or failure occurs in a physical server, the central control component deletes the corresponding static NDP entries and MAC entries in the network access device connected to the physical server. This way of statically configuring and deleting NDP entries and MAC entries can configure corresponding NDP entries and MAC entries for the physical server in the network access device when the physical server is online, implement normal L3 forwarding of data, and delete the corresponding NDP entries and MAC entries when the physical server is abnormal, avoiding the occurrence of forwarding errors.

[0084] FIG. 6 shows a flowchart of a data sending method according to an embodiment of the present disclosure. The data sending method can be applied to any network access device in the network access system described above and executed by the network access device. As shown in FIG. 6, the data sending method includes step S601 and step S602.

[0085] Step S601: determining a target control address and a target interface number corresponding to a destination address from a plurality of routing table entries, a plurality of neighbor discovery table entries, and a plurality of control address table entries according to the destination address of target data, wherein the plurality of routing table entries represent a mapping relationship between a global network address segment of a physical server, a link-local address of a next-hop network interface controller, and an interface number of a device interface, the plurality of neighbor discovery table entries represent a mapping relationship between the link-local address of the network interface controller and a control address and the interface number of the device interface, and the plurality of control address table entries represent a mapping relationship between the control address of the network interface controller and the interface number of the device interface.

[0086] The network access device can be any network access device in the network access system, that is, the multiple device interfaces of the network access device are connected to the network interface controllers of the multiple physical servers one by one, the physical servers are assigned a unique Global-IP address segment, and the link-local address of the network access device and the Linklocal-IP address of the network interface controller belong to the same network address segment. The multiple routing table entries, the multiple NDP table entries, and the multiple MAC table entries on the network access device can be referred to the description above, and will not be described here again.

[0087] After receiving the target data to be forwarded, the network access device sequentially searches the routing table entry, the NDP table entry, and the MAC table entry based on the destination address of the target data, so as to find the target MAC address of the network interface controller and the target interface number of the network access device.

[0088] Step S602: sending the target data to the network interface controller corresponding to the target control address through the device interface corresponding to the target interface number.

[0089] After determining the target MAC address and the target interface number, the network access device can encapsulate the target data into an Ethernet frame with the target MAC address as the destination MAC address, and send the target data to the network interface controller of the corresponding physical server through the device interface corresponding to the target interface number.

[0090] According to the data sending method, the physical server is assigned a unique Global-IP address segment, the Linklocal-IP address of the network interface controller connected to the network access device is configured as the same address segment as the Linklocal-IP address, the network access device is configured with a static route with the next hop being the Linklocal-IP address, and when the network access device forwards data to the corresponding physical server, the data is forwarded according to the configured static route, so the network access device only needs to learn one NDP table entry and one MAC table entry corresponding to the network interface controller of the connected physical server, and does not need to learn the NDP table entries and the MAC table entries corresponding to all container instances, thereby solving the technical problem of insufficient table capacity in the network access device in the case of large deployment of container instances.

[0091] In an embodiment, in step S601, determining the target control address and the target interface number corresponding to the destination address from the plurality of routing table entries, the plurality of neighbor discovery table entries and the plurality of control address table entries according to the destination address of the target data can comprise: selecting a target routing table entry from the plurality of routing table entries, wherein a global network address segment in the target routing table entry matches the destination address; matching a target neighbor discovery table entry from the plurality of neighbor discovery table entries based on a next hop link local address and an interface number in the target routing table entry; matching a target control address table entry from the plurality of control address table entries based on a control address in the target neighbor discovery table entry; and parsing the target control address table entry to obtain the target control address and the target interface number.

[0092] With reference to FIG. 4, the execution subject is taken as an example of the Leaf switch 1. The Leaf switch 1 receives a target data, which is usually an IPv6 packet, and the destination address of the target data is 2001::11 / 64. Two routing table entries are configured on the Leaf switch 1, and the Global-IP address segment to which the destination address 2001::11 / 64 belongs is 2001:: / 64. Therefore, the routing table entry “2001:: / 64 next hop fe80::1 out_if interface 1” is matched as a target routing table entry. The Leaf switch 1 parses the target routing table entry “2001:: / 64 next hop fe80::1 out_if interface 1” to obtain the next hop address fe80::1 and the out interface interface 1, and then matches a corresponding target NDP table entry “fe80::1 is at mac_card1 from interface 1”. The Leaf switch 1 parses the target NDP table entry “fe80::1 is at mac_card1 from interface 1” to obtain the MAC address of the card 1 corresponding to the next hop address fe80::1, and then matches a corresponding target MAC table entry “mac_card1 out interface 1”. Therefore, the target MAC address is the MAC address of the card 1, and the target interface number is interface 1.

[0093] Further, in step S602, the Leaf switch 1 encapsulates the target data into an Ethernet frame with the MAC address of the card 1 as a destination MAC address, and sends the target data to the card 1 of the physical server 1 through the interface 1 of the Leaf switch 1.

[0094] Based on this, the L3 forwarding can be implemented on the network access device side according to the provisions of the NDP protocol, without changing the logical connection relationship between the network access device and the physical server. That is, the network access device only needs to learn one NDP table entry and one MAC table entry corresponding to the network interface controller connected to the physical server, and does not need to learn the NDP table entries and the MAC table entries corresponding to all IPv6 addresses of the physical server.

[0095] In an embodiment, the target physical server is included in the plurality of physical servers, and the data sending method of the embodiment of the present disclosure can further include: receiving a neighbor request message sent by the target physical server, the neighbor request message carrying a link local address and a control address of a target network interface controller deployed on the target physical server; and generating a neighbor discovery table item and a control address table item corresponding to the target network interface controller based on the neighbor request message.

[0096] If the network access device receives the NS message sent by the target physical server, the network access device generates an NDP table item and a MAC table item for the target physical server. Details can be referred to the description above. Based on this, the network access device can always have the NDP table item and the MAC table item corresponding to the Linklocal-IP address of the network interface controller by using the dynamic learning capability of the NDP table item and the MAC table item of the network access device itself. Therefore, even if the network access device temporarily does not have the corresponding NDP table item and MAC table item, the network access device can learn the NDP table item and the MAC table item after the target physical server periodically sends the NS message, and restore the traffic forwarding.

[0097] In an embodiment, the method of the embodiment of the present disclosure can further include: pre-configuring a route validation policy, wherein the route validation policy includes: in the case that the neighbor discovery table item and the control address table item corresponding to the target network interface controller are generated, validating the route table item corresponding to the target network interface controller; and in the case that the neighbor discovery table item and the control address table item corresponding to the target network interface controller are not generated, invalidating the route table item corresponding to the target network interface controller.

[0098] For example, for the NDP table item and the MAC table item configured on the Leaf switch, a static route validation policy is configured, including: if the Leaf switch learns the NDP table item and the MAC table item corresponding to the next hop Linklocal-IP address in the static route, the route table item is validated; and if the Leaf switch does not learn the NDP table item and the MAC table item corresponding to the next hop Linklocal-IP address in the static route, the route table item is invalidated.

[0099] Based on this, if the physical server is abnormal, the network access device cannot learn the NDP table item and the MAC table item, and at this time, the static route table item in the network access device is invalidated, so that the event of forwarding error can be avoided.

[0100] In an embodiment, in the case that the target physical server is online, the network access device is configured with the neighbor discovery table item and the control address table item corresponding to the target network interface controller deployed on the target physical server; in the case that the physical server is abnormal, the neighbor discovery table item and the control address table item corresponding to the target network interface controller are deleted.

[0101] Exemplarily, the network access system can include a central control component capable of communicating with the physical servers and the network access devices in the system, and if a certain physical server is online, the central control component configures the corresponding static NDP table item and MAC table item in the network access device connected by the physical server; if a certain physical server is abnormal (such as offline or failure), the central control component deletes the corresponding static NDP table item and MAC table item in the network access device connected by the physical server.

[0102] This way of statically configuring and deleting the NDP table item and the MAC table item can configure the corresponding NDP table item and MAC table item in the network access device for the physical server when the physical server is online, realize normal L3 forwarding of data, and delete the corresponding NDP table item and MAC table item when the physical server is abnormal, so as to avoid the occurrence of forwarding error events.

[0103] It should be noted that the above application scenarios or application examples provided in the embodiments of the present disclosure are for the purpose of understanding, and the application of the technical solutions of the embodiments of the present disclosure is not specifically limited. In addition, the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present disclosure are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation portal for user to choose authorization or refusal.

[0104] Corresponding to the method provided by the embodiments of the present disclosure, the embodiments of the present disclosure also provide a data sending device applied to a physical server, wherein the physical server is assigned a unique global network address segment, the physical server is deployed with a network interface controller and a plurality of container instances, and the global network addresses of the network interface controller and the container instances belong to the global network address segment; the data sending device comprises: a routing table item acquisition module, configured to acquire a routing table item pre-configured for the physical server in response to a sending request of a target data by any of the container instances, wherein the routing table item of the physical server indicates that the next hop address of the physical server is a link local address of a network access device connected with the network interface controller, and the out interface is the network interface controller, wherein the link local address of the network access device and the link local address of the network interface controller belong to the same network address segment; and a first target data sending module, configured to send the target data to the link local address in the routing table item through the network interface controller.

[0105] In an implementation, the network access device comprises a plurality of device interfaces, each of which corresponds to an interface number, and the plurality of device interfaces comprises a target device interface connected with the network interface controller, and the data sending device further comprises a neighbor request packet sending module, configured to send a neighbor request packet to the network access device, wherein the neighbor request packet carries the link local address of the network interface controller and a control address, so that the network access device generates a neighbor discovery table item and a control address table item corresponding to the network interface controller based on the neighbor request packet, wherein the neighbor discovery table item indicates the matching relationship between the link local address of the network interface controller, the control address of the network interface controller and the interface number of the target device interface, and the control address table item indicates the matching relationship between the control address of the network interface controller and the interface number of the target device interface.

[0106] In an implementation, the sending period of the neighbor request packet is less than the aging period of the neighbor discovery table item configured on the network access device.

[0107] In an implementation, the network access device pre-stores a routing table item corresponding to the network interface controller, configured to indicate that the next hop address of the data belonging to the global network address segment is the link local address of the network interface controller, and the out interface is the interface number of the target device interface.

[0108] Corresponding to the method provided by the embodiments of the present disclosure, the embodiments of the present disclosure further provide a data sending device applied to a network access device, a plurality of device interfaces of the network access device are connected with network interface controllers of a plurality of physical servers one by one, the physical servers are allocated with a unique global network address segment, and a link-local address of the network access device and a link-local address of the network interface controller belong to a same network address segment; the data sending device comprises: a target control address and target interface number determination module configured to determine a target control address and a target interface number corresponding to a destination address of target data from a plurality of routing table entries, a plurality of neighbor discovery table entries and a plurality of control address table entries according to the destination address, wherein the plurality of routing table entries represent a mapping relationship between a global network address segment of the physical server, a link-local address of a network interface controller corresponding to a next hop and an interface number of the device interface, the plurality of neighbor discovery table entries represent a mapping relationship between the link-local address of the network interface controller and a control address and the interface number of the device interface, and the plurality of control address table entries represent a mapping relationship between the control address of the network interface controller and the interface number of the device interface; and a second target data sending module configured to send the target data to the network interface controller corresponding to the target control address through a device interface corresponding to the target interface number.

[0109] In an implementation manner, the target control address and target interface number determination module comprises: a target routing table entry matching sub-module configured to select a target routing table entry with a global network address segment matching the destination address from the plurality of routing table entries; a target neighbor discovery table entry matching sub-module configured to match a target neighbor discovery table entry from the plurality of neighbor discovery table entries based on a link-local address of a next hop and an interface number in the target routing table entry; a target control address table entry matching sub-module configured to match a target control address table entry from the plurality of control address table entries based on a control address in the target neighbor discovery table entry; and a parsing sub-module configured to parse the target control address table entry to obtain the target control address and the target interface number.

[0110] In an implementation manner, the plurality of physical servers comprise a target physical server, and the data sending device further comprises: a neighbor request packet receiving module configured to receive a neighbor request packet sent by the target physical server, the neighbor request packet carrying a link-local address and a control address of a target network interface controller deployed on the target physical server; and a table entry generating module configured to generate a neighbor discovery table entry and a control address table entry corresponding to the target network interface controller based on the neighbor request packet.

[0111] In an implementation, the data sending apparatus further comprises a route validation policy configuration module configured to pre-configure a route validation policy, wherein the route validation policy comprises: validating a route table item corresponding to the target network interface controller if a neighbor discovery table item and a control address table item corresponding to the target network interface controller are generated; invalidating the route table item corresponding to the target network interface controller if the neighbor discovery table item and the control address table item corresponding to the target network interface controller are not generated.

[0112] In an implementation, in a case that the target physical server is online, the network access device is configured with a neighbor discovery table item and a control address table item corresponding to a target network interface controller deployed on the target physical server; in a case that the physical server is abnormal, the neighbor discovery table item and the control address table item corresponding to the target network interface controller are deleted.

[0113] The functions of each module in each device of the embodiments of the present disclosure can refer to the corresponding description in the above method, and have the corresponding beneficial effects, which will not be described here again.

[0114] FIG. 7 is a block diagram of an electronic device for implementing the embodiments of the present disclosure. As shown in FIG. 7, the electronic device includes a memory 701 and a processor 702, and the memory 701 stores a computer program capable of running on the processor 702. The processor 702 implements the method in the above embodiments when executing the computer program. The number of the memory 701 and the processor 702 can be one or more. In a specific implementation, the electronic device can further include a communication interface 703 for communicating with external devices and performing data transmission.

[0115] In a specific implementation, if the memory 701, the processor 702 and the communication interface 703 are independently implemented, the memory 701, the processor 702 and the communication interface 703 can be connected with each other through a bus and complete communication between each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, only one thick line is used in FIG. 7, but it does not mean that there is only one bus or only one type of bus.

[0116] Optionally, in a specific implementation, if the memory 701, the processor 702 and the communication interface 703 are integrated on a chip, the memory 701, the processor 702 and the communication interface 703 can complete the communication among each other through an internal interface.

[0117] The computer readable storage medium stores a computer program, and the program is executed by a processor to implement the method provided in the embodiments of the present disclosure.

[0118] The computer program product comprises a computer program, and the program is executed by a processor to implement the method provided in the embodiments of the present disclosure.

[0119] The chip comprises a processor, and the processor is configured to call and run instructions stored in a memory, so that a communication device installed with the chip executes the method provided in the embodiments of the present disclosure.

[0120] The chip comprises an input interface, an output interface, a processor and a memory, and the input interface, the output interface, the processor and the memory are connected through an internal connection path. The processor is configured to execute code in the memory, and when the code is executed, the processor is configured to execute the method provided in the embodiments of the present disclosure.

[0121] It should be understood that the processor described above can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc. It should be noted that the processor can be a processor supporting an advanced RISC machine (ARM) architecture.

[0122] Further, the memory can include a read-only memory and a random access memory, optionally. The memory can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memory. The non-volatile memory can include a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically EPROM (EEPROM), or a flash memory, among others. The volatile memory can include a random access memory (RAM), which is used as an external cache. By way of example, and not limitation, many forms of RAM are available. The RAM can include a static random access memory (SRAM), a dynamic random access memory (DRAM), a synchronous dynamic random access memory (SDRAM), a double data rate SDRAM (DDR SDRAM), an enhanced SDRAM (ESDRAM), a Sync link DRAM (SLDRAM), and a direct Rambus RAM (DR RAM), among others.

[0123] In the above-described embodiments, all or part of the embodiments can be implemented by software, hardware, firmware, or any combination thereof. When implemented by software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions according to the present disclosure are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable apparatus. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another computer-readable storage medium.

[0124] In the description of the present disclosure, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present disclosure. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, different embodiments or examples described in the present disclosure and the features of different embodiments or examples can be combined and combined by those skilled in the art without contradiction, if necessary.

[0125] In addition, the terms "first", "second" are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include at least one of the features. In the description of the present disclosure, the meaning of "a plurality of" is two or more, unless otherwise specifically limited.

[0126] Any process or method described in the flowchart or otherwise described herein can be understood as a representation of code, including one or more executable instructions for implementing the specific logical functions or steps, modules, segments or portions. And the scope of the preferred embodiments of the present disclosure includes additional implementations, in which the functions can be performed in the order shown or discussed, including in a substantially simultaneous manner or in reverse order according to the functions involved.

[0127] The logic and / or steps described in the flowchart or otherwise described herein, for example, can be considered as a list of executable instructions for implementing the logic function, which can be embodied in any computer-readable medium for use by or in conjunction with an instruction execution system, device or apparatus, such as a computer-based system, a system including a processor or other system that can fetch and execute instructions from the instruction execution system, device or apparatus.

[0128] It should be understood that parts of the present disclosure can be implemented in hardware, software, firmware or a combination thereof. In the above-described embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. All or part of the steps of the above-described embodiment method can be completed by a program instructing the relevant hardware, which can be stored in a computer-readable storage medium, and the program includes one or a combination of the steps of the method embodiment when executed.

[0129] In addition, each function unit in each embodiment of the present disclosure can be integrated in one processing module, or each unit can be physically present separately, or two or more units can be integrated in one module. The integrated module can be realized in the form of hardware or in the form of a software function module. When the integrated module is realized in the form of a software function module and sold or used as an independent product, it can also be stored in a computer readable storage medium. The storage medium can be a read-only memory, a magnetic disk or an optical disk, etc.

[0130] The above is only an exemplary embodiment of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any person skilled in the art can easily think of various changes or replacements within the technical range disclosed in the present disclosure, and these should be covered within the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be subject to the protection scope of the claims.

Claims

1. A data sending method applied to a network access device, a plurality of device interfaces of the network access device being connected to network interface controllers of a plurality of physical servers one-to-one, the physical servers being assigned a unique global network address segment, a link-local address of the network access device and a link-local address of the network interface controllers belonging to a same network address segment; the method comprising: determining a target control address and a target interface number corresponding to a destination address of target data from a plurality of routing table entries, a plurality of neighbor discovery table entries and a plurality of control address table entries, wherein the plurality of routing table entries represent a mapping relationship between a global network address segment of the physical servers, a link-local address of a network interface controller corresponding to a next hop and an interface number of the device interface, the plurality of neighbor discovery table entries represent a mapping relationship between a link-local address of the network interface controller and a control address and an interface number of the device interface, and the plurality of control address table entries represent a mapping relationship between a control address of the network interface controller and an interface number of the device interface; sending the target data to a network interface controller corresponding to the target control address through a device interface corresponding to the target interface number.

2. The method of claim 1, wherein, The determining of the target control address and the target interface number corresponding to the destination address of the target data from the plurality of routing table entries, the plurality of neighbor discovery table entries and the plurality of control address table entries comprises: selecting a target routing table entry with a global network address segment matching the destination address from the plurality of routing table entries; matching a target neighbor discovery table entry from the plurality of neighbor discovery table entries based on a link-local address of a next hop and an interface number in the target routing table entry; matching a target control address table entry from the plurality of control address table entries based on a control address in the target neighbor discovery table entry; parsing the target control address table entry to obtain the target control address and the target interface number.

3. The method of claim 1 or 2, wherein, The plurality of physical servers include a target physical server, and the method further comprises: receiving a neighbor solicitation packet sent by the target physical server, the neighbor solicitation packet carrying a link-local address and a control address of a target network interface controller deployed on the target physical server; generating a neighbor discovery table entry and a control address table entry corresponding to the target network interface controller based on the neighbor solicitation packet. 4.The method of claim 3, further comprising: configuring a routing validation policy, wherein the routing validation policy comprises: validating a routing table entry corresponding to the target network interface controller in a case where the neighbor discovery table entry and the control address table entry corresponding to the target network interface controller are generated; and invalidating the routing table entry corresponding to the target network interface controller in a case where the neighbor discovery table entry and the control address table entry corresponding to the target network interface controller are not generated.

5. The method of claim 1 or 2, wherein, In the case that the target physical server is online, the network access device is configured with neighbor discovery table items and control address table items corresponding to the target network interface controller deployed on the target physical server; in the case that the physical server is abnormal, the neighbor discovery table items and the control address table items corresponding to the target network interface controller are deleted.

6. A data sending method applied to a physical server, wherein the physical server is assigned a unique global network address segment, a network interface controller and a plurality of container instances are deployed on the physical server, and global network addresses of the network interface controller and the container instances belong to the global network address segment; the method comprises the following steps of: in response to a sending request of any container instance for target data, obtaining a routing table item pre-configured for the physical server, wherein the routing table item of the physical server indicates that a next hop address of the physical server is a link local address of a network access device connected with the network interface controller, and an out interface is the network interface controller, wherein the link local address of the network access device and the link local address of the network interface controller belong to a same network address segment; sending the target data to the link local address in the routing table item through the network interface controller.

7. The method of claim 6, wherein, The network access device comprises a plurality of device interfaces, each device interface corresponding to an interface number, and the plurality of device interfaces comprise a target device interface connected with the network interface controller, and the method further comprises the following steps of: sending a neighbor request packet to the network access device, wherein the neighbor request packet carries the link local address of the network interface controller and a control address, so that the network access device generates a neighbor discovery table item and a control address table item corresponding to the network interface controller based on the neighbor request packet, wherein the neighbor discovery table item indicates a matching relationship among the link local address of the network interface controller, the control address of the network interface controller and the interface number of the target device interface, and the control address table item indicates a matching relationship between the control address of the network interface controller and the interface number of the target device interface.

8. The method of claim 7, wherein, The sending period of the neighbor request packet is less than the aging period of the neighbor discovery table item configured on the network access device.

9. The method of claim 7 or 8, wherein, The network access device pre-stores a routing table item corresponding to the network interface controller, which is used to indicate that a next hop address of data belonging to the global network address segment is the link local address of the network interface controller, and an out interface is the interface number of the target device interface.

10. A network access system comprising a plurality of physical servers and a plurality of network access devices, the physical servers are assigned a unique global network address segment, global network addresses of network interface controllers and a plurality of container instances deployed on the physical servers belong to the global network address segment, and network interface controllers of the physical servers are connected with the plurality of network access devices; The network access device has a plurality of device interfaces connected one-to-one with network interface controllers of the plurality of physical servers, is preconfigured with a plurality of routing table entries for representing a mapping relationship between a global network address segment of the physical servers, a link-local address of a next-hop network interface controller and an interface number of the device interface, a link-local address of the network access device and the link-local address of the network interface controller belong to a same network address segment; and the physical server is preconfigured with a routing table entry for representing a next-hop address of the physical server as a link-local address of a network access device connected by the physical server. 11.An electronic device comprising a memory, a processor, and a computer program stored on the memory, wherein the processor implements the method of any one of claims 1 to 9 when executing the computer program. 12.A computer-readable storage medium having stored therein a computer program, wherein the computer program, when executed by a processor, implements the method of any one of claims 1 to 9. 13.A computer program product comprising a computer program, wherein the computer program, when executed by a processor, implements the method of any one of claims 1 to 9.

Images