A network element load balancing method, device, equipment and storage medium

Abstract

The embodiment of the present application relates to the technical field of data processing, in particular to a network element load balancing method, device, equipment and storage medium. The embodiment aims to realize network element load balancing while ensuring accurate data forwarding. The method comprises: a switch receiving traffic data sent by an upstream component; the switch sending the traffic data to at least one network element node in a network element cluster in a load balanced manner according to a first route published by the at least one network element node; the network element node sending processed traffic data to a downstream component; the switch sending response data sent by the downstream component to the network element node; in the case of abnormal operation of the network element node, the switch sends the response data sent by the downstream component to the remaining network element nodes in the network element cluster; and the network element cluster sends the response data to the upstream component through the switch.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and more specifically, to a network element load balancing method, apparatus, device, and storage medium. Background Technology

[0002] Cloud computing networks typically deploy various network elements with different functions. These network elements receive data packets from upstream components, process them, and then send them to downstream components. In existing technologies, to ensure stable data transmission, multiple network elements are usually configured for data transmission, allowing upstream components to distribute data evenly across these elements.

[0003] In existing technologies, when multiple network elements are set up, downstream components cannot ensure that the message is accurately sent to the network element that previously sent the data when replying to the network element. This leads to session desynchronization among multiple network elements, which in turn prevents data forwarding and affects the data transmission efficiency of cloud computing. Summary of the Invention

[0004] This application provides a network element load balancing method, apparatus, device, and storage medium, aiming to achieve network element load balancing while ensuring accurate data forwarding.

[0005] The first aspect of this application provides a network element load balancing method, the method comprising:

[0006] The switch receives traffic data sent by upstream components;

[0007] The switch load balances the traffic data to at least one of the network element nodes based on the first route published by at least one network element node in the network element cluster.

[0008] The network element node sends the processed traffic data to downstream components through the switch;

[0009] When the network element node is operating normally, the switch sends the response data sent by the downstream component to the network element node according to the second route published by the network element node;

[0010] In the event of an abnormal operation of a network element node, the switch, based on the third route published by the network element node, sends the response data sent by the downstream component to the remaining network element nodes in the network element cluster;

[0011] The network element cluster sends the response data to the upstream component through the switch.

[0012] Optionally, when the switch load-balances the traffic data to at least one of the network element nodes based on a first route published by at least one network element node in the network element cluster, the method further includes:

[0013] The controller unit of the network element node obtains network element configuration data from the management server. The network element configuration data includes the network element host name, network element cluster identifier, virtual address, interconnection address, network element node address, and network element forwarding rules.

[0014] The controller unit sends the network element forwarding rules, the virtual address, the interconnection address, and the network element node address to the data forwarding unit of the network element node;

[0015] The controller unit sends the virtual address and the network element address to the BGP unit in the network element node.

[0016] Optionally, the method further includes:

[0017] The network element controller unit sends a routing entry sending instruction to the BGP unit;

[0018] The BGP unit sends an instruction based on the routing entry to obtain the corresponding route, which includes the destination address, next-hop address, and AS PATH information.

[0019] The BGP unit sends the route to the switch through the data forwarding unit.

[0020] Optionally, the BGP unit sends the route to the switch through the data forwarding unit, including:

[0021] The BGP unit sends the first route to the switch through the data forwarding unit. The destination address corresponding to the first route is the virtual address, the next-hop address corresponding to the first route is the interconnect address, and the AS PATH information is empty.

[0022] The BGP unit sends the second route to the switch through the data forwarding unit. The destination address corresponding to the second route is the network element node address of the network element node, the next hop address corresponding to the second route is the interconnection address, and the AS PATH information is empty.

[0023] The BGP unit sends the third route to the switch through the data forwarding unit. The destination address corresponding to the third route is the network element node address of all network element nodes in the network element cluster except the network element node where the BGP unit is located. The next-hop address corresponding to the third route is the interconnection address. The AS PATH information corresponding to the third route is the AS number of the network element node.

[0024] Optionally, the method further includes:

[0025] The network element cluster establishes a synchronization session between each network element node based on the interconnection address corresponding to each network element node;

[0026] The synchronization session is used to synchronize the session information of data traffic in each network element node.

[0027] Optionally, when the network element node is operating normally, before the switch sends the response data sent by the downstream component to the network element node according to the second route published by the network element node, the method further includes:

[0028] The switch receives heartbeat information sent by the BGP unit on the network element node;

[0029] If the switch receives the heartbeat information normally, it is determined that the network element node is operating normally;

[0030] If the switch is unable to receive the heartbeat information normally, it is determined that the network element node is malfunctioning.

[0031] Optionally, the method further includes:

[0032] If the network element node is found to be malfunctioning, the switch disconnects from the network element node.

[0033] Delete the routing information sent by the network element node.

[0034] A second aspect of this application provides a network element load balancing device, the device comprising:

[0035] The data receiving module is used by the switch to receive traffic data sent by upstream components;

[0036] The first data sending module is used by the switch to send the traffic data to at least one of the network element nodes in a load-balanced manner according to the first route published by at least one network element node in the network element cluster.

[0037] The second data transmission module is used by the network element node to send the processed traffic data to downstream components through the switch;

[0038] The third data transmission module is used to, when the network element node is operating normally, send the response data sent by the downstream component to the network element node according to the second route published by the network element node;

[0039] The fourth data sending module is used to, in the event of an abnormal operation of the network element node, send the response data sent by the downstream component to the remaining network element nodes in the network element cluster according to the third route published by the network element node.

[0040] The fifth data transmission module is used by the network element cluster to send the response data to the upstream component through the switch.

[0041] Optionally, the device further includes:

[0042] The configuration data acquisition module is used by the controller unit of the network element node to acquire network element configuration data from the management server. The network element configuration data includes the network element host name, network element cluster identifier, virtual address, interconnection address, network element node address, and network element forwarding rules.

[0043] The first configuration data sending module is used by the controller unit to send the network element forwarding rules, the virtual address, the interconnection address and the network element node address to the data forwarding unit of the network element node;

[0044] The second configuration data sending module is used by the controller unit to send the virtual address and the network element address to the BGP unit in the network element node.

[0045] Optionally, the device further includes:

[0046] The instruction sending module is used for the network element controller unit to send a routing entry sending instruction to the BGP unit;

[0047] The route acquisition module is used by the BGP unit to obtain the corresponding route by sending an instruction according to the route entry. The route includes the destination address, the next-hop address, and AS PATH information.

[0048] A route sending module is used by the BGP unit to send the route to the switch through the data forwarding unit.

[0049] Optionally, the routing sending module includes:

[0050] The first route sending submodule is used by the BGP unit to send the first route to the switch through the data forwarding unit. The destination address corresponding to the first route is the virtual address, the next-hop address corresponding to the first route is the interconnect address, and the AS PATH information is empty.

[0051] The second route sending submodule is used by the BGP unit to send the second route to the switch through the data forwarding unit. The destination address corresponding to the second route is the network element node address of the network element node, the next hop address corresponding to the second route is the interconnection address, and the AS PATH information is empty.

[0052] The third route sending submodule is used by the BGP unit to send the third route to the switch through the data forwarding unit. The destination address corresponding to the third route is the network element node address of all network element nodes in the network element cluster except the network element node where the BGP unit is located. The next hop address corresponding to the third route is the interconnection address. The AS PATH information corresponding to the third route is the AS number of the network element node.

[0053] Optionally, the device further includes:

[0054] The session synchronization module is used by the network element cluster to establish a synchronization session between each network element node based on the interconnection address corresponding to each network element node;

[0055] The information synchronization module is used to synchronize the session information of data traffic in each of the network element nodes through the synchronization session.

[0056] Optionally, the device further includes:

[0057] A heartbeat information receiving module is used by the switch to receive heartbeat information sent by the BGP unit on the network element node;

[0058] The first state determination module is used to determine that the network element node is operating normally when the switch normally receives the heartbeat information;

[0059] The second state determination module is used to determine that the network element node is malfunctioning when the switch cannot receive the heartbeat information normally.

[0060] Optionally, the device further includes:

[0061] A connection disconnection module is used to disconnect the switch from the network element node when it is determined that the network element node is malfunctioning.

[0062] The information deletion module is used to delete the routing information sent by the network element node.

[0063] A third aspect of this application provides a readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps described in the first aspect of this application.

[0064] A fourth aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the method described in the first aspect of this application.

[0065] Using the network element load balancing method provided in this application, a switch receives traffic data sent by an upstream component; the switch, according to a first route published by at least one network element node in the network element cluster, load-balances the traffic data to at least one of the network element nodes; the network element node, through the switch, sends the processed traffic data to a downstream component; when the network element node is operating normally, the switch, according to a second route published by the network element node, sends response data sent by the downstream component to the network element node; when the network element node is operating abnormally, the switch, according to a third route published by the network element node, sends the response data sent by the downstream component to the remaining network element nodes in the network element cluster; the network element cluster, through the switch, sends the response data to the upstream component.

[0066] In this method, the switch evenly distributes traffic data to multiple nodes in the network element cluster, ensuring load balancing among the network element nodes. The network element nodes pre-publish multiple routes to the switch. When a network element node is operating normally, downstream components send response data to the network element node according to the route corresponding to the network element node, ensuring normal data transmission. When a network element node is running abnormally, the response data is sent to the remaining network elements in the network element cluster according to the route sent by the network element node, ensuring high availability of the network elements and preventing interruption of data transmission. This method achieves network element load balancing while accurately transmitting data. Attached Figure Description

[0067] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0068] Figure 1 This is a schematic diagram of data transmission according to an embodiment of this application;

[0069] Figure 2 This is a schematic diagram of a network element load balancing architecture proposed in an embodiment of this application;

[0070] Figure 3 This is a flowchart of a network element load balancing method proposed in an embodiment of this application;

[0071] Figure 4 This is a schematic diagram of routing entries proposed in an embodiment of this application;

[0072] Figure 5 This is a schematic diagram of normal traffic forwarding according to an embodiment of this application;

[0073] Figure 6 This is a schematic diagram of abnormal traffic forwarding proposed in an embodiment of this application;

[0074] Figure 7 This is a schematic diagram of a network element load balancing device proposed in an embodiment of this application;

[0075] Figure 8 This is a schematic diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0076] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0077] To facilitate explanation, the terms used in this application will first be explained:

[0078] BGP (Border Gateway Protocol) is a routing protocol used to exchange network layer reachability information between routing domains. It is primarily used to exchange reachability routing information between Autonomous Systems (AS), construct propagation paths between AS domains, and prevent routing loops.

[0079] MAC address (Media Access Control Address): Also known as a local area network address, Ethernet address, or physical address.

[0080] LVS (Linux Virtual Server): Linux Virtual Server, an open-source load balancing project.

[0081] SNAT (Source Network Address Translation): Source network address translation.

[0082] DNAT (Destination Network Address Translation): Destination network address translation.

[0083] ECMP (Equal-cost multi-path): When there are multiple routes to the same destination IP with equal cost in an IP routing network, traffic can be forwarded to multiple next hops according to specific rules to achieve network link-level load balancing and enable fast switching when a link fails.

[0084] VIP (Virtual IP address): Virtual IP address.

[0085] In related technologies, computer networks typically deploy multiple network elements with different functions. These network elements receive VXLAN data packets from upstream components, parse specific layer fields (such as IP addresses) and data in the data packets, modify specific fields of the data packets according to pre-set rules, and finally encapsulate and forward the data packets to downstream components in the network.

[0086] To achieve reliable packet forwarding and avoid network outages caused by the failure of a single network element, it is usually necessary to deploy multiple network elements with the same function. When the primary network element fails, a specific method and mechanism is used to switch the backup network element to become the primary network element and continue forwarding packets. This requires additional investment in network element equipment, increasing network construction and maintenance costs. However, the forwarding capacity of a network element is not improved compared to a single network element, resulting in low resource utilization. This approach is not suitable for large-scale networks with high bandwidth requirements. Currently, some technologies propose using multiple network elements to achieve a multi-master (multi-active) deployment. Multiple network elements with the same function forward data together. This requires that the upstream device sending packets to the network elements can identify multiple network elements as next hops and send multiple packets to multiple available network elements according to a specific load balancing strategy. When one network element fails, the upstream device needs to detect the failure and remove the failed device from the list of available next hops to avoid sending packets to the failed network element and causing packet loss.

[0087] Network elements are divided into two types: stateful network elements and stateless network elements. Stateless network elements forward data packets based solely on forwarding rules and do not depend on the session. In IP networks, a session typically includes information such as source IP, destination IP, source port number, destination port number, and protocol, i.e., information related to historical data packets. Stateful network elements forward data based on forwarding rules and the session. Network elements that implement SNAT and DNAT are typically stateful network elements. If a stateful network element is deployed in a multi-active mode, it is required that all incoming and outgoing data packets of the same session be forwarded to the same network element; otherwise, packets may be dropped and unable to be forwarded due to missing session information.

[0088] Currently, data forwarding for network element load balancing is typically implemented using software such as BGP, LVS, Nginx, and Keepalived. These software combinations, acting as network elements or part of a network element, enable load balancing of data forwarding from multiple network elements to the next hop. BGP, for example, can achieve load balancing of ECMP data traffic across multiple network elements by sending equal-cost routing entries to upstream network devices (such as switches). However, when downstream devices reply with packets, it cannot guarantee that the packets will be sent to the network element that previously sent the request packet, leading to session asynchrony and packet forwarding failures. LVS achieves load balancing by modifying the MAC address of the next hop, requiring the client, LVS network element, and backend network element or service to all be in a Layer 2 network. Layer 2 networks are typically very limited in scale due to loop and broadcast storm issues and cannot be used in large-scale networks. Furthermore, LVS does not support stateful packet processing such as SNAT / DNAT. The combination of Nginx and high-availability failover software such as Keepalived allows Nginx to record session information internally, while Keepalived enables automatic primary / backup failover when Nginx fails. However, due to the lack of features such as session synchronization, it cannot achieve multi-active mode, cannot scale horizontally, and cannot meet the needs of large-scale networks.

[0089] To address the problem that existing technologies cannot achieve high availability and load balancing deployment of stateful network elements and multi-active data forwarding, this invention provides a method and apparatus for achieving high availability and load balancing of stateful network elements suitable for cloud computing networks. The configuration database stores network element cluster configuration information. The network element controller component of each network element node uses this database to publish the front-end VIP address for connecting to upstream network devices and the BIP (Backend IP address) address for connecting to downstream network devices, based on BGP. The switches connected to the network element use ECMP to load balance upstream traffic to the network element. A stateful data forwarding component handles data forwarding and session synchronization within the network element cluster. The forwarding component uses the network element's BIP address as the source IP to send packets to downstream components. When downstream components reply, the destination IP is the network element's BIP address. The switch calculates the priority based on the AS PATH length of the cluster network element's BIP address and delivers the reply packet to the network element that originally sent the session request packet. If that network element fails, the packet is sent to other network elements within the cluster. Network element session synchronization ensures normal data forwarding in fault scenarios, enabling multi-active load balancing deployment of stateful network elements. It fully utilizes all network elements for data forwarding to improve throughput. When a network element fails, traffic can be automatically switched without session loss, supporting various data forwarding needs of large-scale cloud computing networks.

[0090] This application proposes a high-availability load balancing architecture for multi-active stateful network elements, referencing... Figure 1 , Figure 1 This is a schematic diagram of data transmission according to an embodiment of this application, as shown below. Figure 1 As shown, the request message from the upstream component is forwarded by the switch to one of the network elements in the multi-active stateful network element cluster. During transmission, multiple different requests can be sent to multiple nodes in the network element cluster using hash values. The network element nodes process the message and generate a downstream request message, which is then sent to the downstream component through the switch. The switch then sends the response message from the downstream component to the original network element that forwarded the request according to the route published by the network element. The original network element processes the response message and generates a response message for the upstream component, which is then forwarded to the upstream component by the switch.

[0091] refer to Figure 2 , Figure 2 This is a schematic diagram of a network element load balancing architecture proposed in an embodiment of this application, as shown below. Figure 2 As shown, the architecture includes:

[0092] The network element configuration database unit is used to store the configuration information of the network element cluster.

[0093] The network element controller unit is used to receive configuration information from the network element configuration database unit, control the data forwarding unit to forward data, and control the BGP unit to publish routes to the switch.

[0094] The BGP unit is used to implement the BGP protocol and sends the corresponding routes to the switch according to the instructions of the network element controller unit.

[0095] The data forwarding unit is used to forward data according to the forwarding rules issued by the network element controller.

[0096] The switching unit is used to evenly distribute data sent by upstream components to various network element nodes, and to return downstream response data to the corresponding network element nodes according to the routes published by the BGP unit.

[0097] refer to Figure 3 , Figure 3 This is a flowchart of a network element load balancing method proposed in an embodiment of this application. For example... Figure 3 As shown, the method includes the following steps:

[0098] S11: The switch receives traffic data sent by upstream components.

[0099] In this embodiment, when an upstream component (i.e., a client) in the cloud network needs to request data from a downstream component (i.e., a server), it first sends the corresponding traffic data in the form of a message, and the switch receives the traffic data sent by the upstream component.

[0100] S12: The switch load balances the traffic data to at least one of the network element nodes according to the first route published by at least one network element node in the network element cluster.

[0101] In this embodiment, the network element cluster includes multiple network element nodes, and the first route contains the virtual address of the network element cluster. The switch sends the traffic data to at least one network element node in the network element cluster corresponding to the virtual address in a load-balanced manner according to the first route published by the network element nodes in the network element cluster.

[0102] In this embodiment, the switch unit establishes a BGP neighbor with the network element BGP module, receives routing entries from the network element through the BGP protocol, and summarizes the routing entries published by multiple network elements to form a routing table. Among them, the route with the VIP address as the destination address generates an ECMP equivalent route because the AS PATH published by each network element in the cluster is consistent. ECMP is configured to perform traffic hashing according to the five-tuple of the packet (source IP, destination IP, source port, destination port, protocol). The traffic sent from the upstream component to the VIP will be distributed evenly to the network elements in the cluster. That is, the received traffic data is evenly distributed to each network element node according to the load of each network element node, so that the load of each network element node is the same.

[0103] S13: The network element node sends the processed traffic data to the downstream component through the switch.

[0104] In this embodiment, after receiving traffic data, the network element node processes the traffic data and then sends the data to the switch. The source address carried is the network element node address. The switch uses the address of the network element node as the source address and sends the traffic data to the downstream components.

[0105] In this embodiment, the data forwarding module in the network element node is connected to the physical network card, receives VXLAN data packets with a destination IP address of VIP, determines that the data packets are sent from the upstream component, and performs packet forwarding operations such as SNAT / DNAT on the inner VXLAN packets, and sends the packets to the downstream component using the BIP address of this network element node as the source IP.

[0106] S14: When the network element node is operating normally, the switch sends the response data sent by the downstream component to the network element node according to the second route published by the network element node.

[0107] In this embodiment, when the network element node is operating normally, the switch sends the response data sent by the downstream component to the network element node according to the second route published by the BGP unit on the network element node.

[0108] S15: In the event of an abnormal operation of the network element node, the switch sends the response data sent by the downstream component to the remaining network element nodes in the network element cluster according to the third route published by the network element node.

[0109] In this embodiment, when a network element node malfunctions, the switch sends the response data sent by the downstream component to the remaining network element nodes in the network element cluster according to the third route sent by the BGP unit of the network element node.

[0110] S16: The network element cluster sends the response data to the upstream component through the switch.

[0111] In this embodiment, after receiving the response data, the network element node in the network element cluster sends the response data to the switch, and the switch then sends it to the upstream component.

[0112] In this embodiment, the switch module establishes a BGP neighbor relationship with the network element BGP module, receives routing entries from the network element via the BGP protocol, and summarizes these entries into a routing table. Routes with the VIP address as the destination address generate ECMP equivalent routes because the AS PATHs published by all network elements within the cluster are consistent. ECMP is configured to perform traffic hashing based on the packet's five-tuple (source IP, destination IP, source port, destination port, protocol), converting traffic data into hash values. Traffic from upstream components to the VIP is then load-balanced across the network elements within the cluster. Routes to the BIP address, due to inconsistent AS PATH lengths, form multiple routes with different priorities and unequal costs. Packets destined for the BIP are preferentially sent to the BIP's original home node. The switch's BGP is configured to enable a heartbeat mechanism. The heartbeat interval can be set manually, for example, 3 seconds. Three consecutive heartbeat timeouts are considered abnormal, and the number of times can also be set manually. When a node fails, the BGP neighbor will automatically disconnect abnormally upon heartbeat timeout, and the BGP routes published by that neighbor (network element) will be automatically withdrawn. The switch will then forward packets destined for the VIP and BIP to other nodes in the cluster according to the remaining routes. The data forwarding module receives VXLAN packets with a destination IP address of BIP address through the physical network card, determines that it is a return packet from a downstream component, and performs packet forwarding operations such as SNAT / DNAT for the inner VXLAN packets according to the session. Using the VIP address of this network element node as the source IP, the packet is sent back to the upstream component, completing a complete data transmission and reception cycle.

[0113] refer to Figure 4 , Figure 4 This is a schematic diagram of routing entries proposed in an embodiment of this application, as shown below. Figure 4 As shown, each network element node publishes one VIP address route without AS PATH and three BIP address routes. The BIP route of this node does not have AS PATH, while the BIP routes of other nodes in the cluster carry AS PATH.

[0114] refer to Figure 5 , Figure 5 This is a schematic diagram of normal traffic forwarding proposed in an embodiment of this application, as shown below. Figure 5 As shown, Figure 4There are three network element nodes. When all network elements in the cluster are normal, the switch will load balance the traffic from the upstream component with the destination IP as the network element cluster VIP to each network element. Each network element carries about 1 / 3 of the total traffic. After the network element processes the packet, it uses BIP to send the packet to the downstream component. The downstream component replies with the packet with the destination IP as the BIP of each network element. The switch sends the reply packet back to the corresponding network element node according to the routing priority.

[0115] refer to Figure 6 , Figure 6 This is a schematic diagram of abnormal traffic forwarding proposed in an embodiment of this application, as shown below. Figure 6 As shown, assuming network element node 1 fails, the high-priority BIP1 BGP route published by network element node 1 is automatically withdrawn. Network elements 2 and 3 still publish the BIP address route of network element 1. The switch switches the traffic previously sent to network element 1 by downstream components to network element 2 and network element 3 nodes in the cluster. Because the data forwarding module of network element 1 has synchronized the sessions to network elements 2 and 3, network elements 2 and 3 can forward based on the sessions synchronized before the failure of network element 1, achieving near-lossless high-availability switching in the failure scenario. Only a very small number of sessions that have been established in network element 1 but not synchronized (depending on the number of sessions synchronized and the time configuration) will be lost due to the lack of timely synchronization and need to be re-established.

[0116] In another embodiment of this application, before the switch load-balances the traffic data to at least one of the network element nodes according to a first route published by at least one network element node in the network element cluster, the method further includes:

[0117] S21: The controller unit of the network element node obtains network element configuration data from the management server. The network element configuration data includes the network element host name, network element cluster identifier, virtual address, interconnection address, network element node address, and network element forwarding rules.

[0118] In this embodiment, the controller unit of the network element node obtains network element configuration data from the network element configuration database module of the management server. The network element configuration data includes the network element host name, network element cluster identifier (used to identify the cluster where the network element node is located), virtual address of the network element cluster, interconnection address, network element node address (BIP), and network element forwarding rules.

[0119] In this embodiment, the network element configuration database module is deployed on the management server node. It stores the network element hostname, the cluster to which the network element belongs, the VIP address of the cluster, the interconnect IP address of the network element host, the BIP address of the network element host, and the forwarding configuration of the network element. The database module is deployed using a three-node cluster to ensure reliable storage of the network element configuration. The network element configuration data is planned and written during network element cluster deployment. The IP allocation principle is that one VIP address is assigned to each network element cluster for all network elements to share. The VIP address serves as the destination IP used by upstream components when sending traffic to this network element cluster. Each network element is assigned an interconnect IP address for Layer 2 interconnection with its uplink switch. Each network element is also assigned a BIP address for its host, which serves as the destination IP used by downstream components when responding to traffic.

[0120] S22: The controller unit sends the network element forwarding rule, the virtual address, the interconnection address, and the network element node address to the data forwarding unit of the network element node.

[0121] In this embodiment, the network element controller unit sends the network element forwarding rules, virtual address, interconnection address, and network element node address to the data forwarding unit of the network element node.

[0122] S23: The controller unit sends the virtual address and the network element address to the BGP unit in the network element node.

[0123] In this embodiment, the network element controller unit sends the virtual address and the network element address to the BGP unit in the network element node.

[0124] In this embodiment, the network element controller module is deployed on network element nodes, using a general-purpose server as the network element node. The network element controller module connects to the network element configuration database to retrieve configuration information, and distributes forwarding rules, local cluster interconnection IP, VIP address, and BIP address to the data forwarding module. It also distributes the VIP and BIP addresses to the BGP module and periodically checks whether the configurations of the data forwarding module and BGP module are consistent with the configuration in the network element configuration database. When the data forwarding module configuration is inconsistent, it automatically updates the data forwarding module configuration according to the database configuration. When the data forwarding module configuration is abnormal and the configuration cannot be distributed, the network element controller records an alarm and calls the BGP module to cancel all routes published by this network element, preventing the switch from sending traffic to the abnormal network element and avoiding abnormal packet loss caused by incorrect forwarding configuration.

[0125] In this embodiment, the method further includes:

[0126] S24: The network element controller unit sends a routing entry sending instruction to the BGP unit.

[0127] In this embodiment, the routing entry sending instruction is used to notify the BGP unit which routing entries need to be sent.

[0128] In this embodiment, the network element controller unit sends a routing entry sending instruction to the BGP unit.

[0129] S25: The BGP unit sends an instruction according to the routing entry to obtain the corresponding route, the route including the destination address, the next-hop address, and AS PATH information.

[0130] In this embodiment, AS path is an important attribute in the BGP (Border Gateway Protocol) routing protocol. It is mainly used to record the transmission path of routes between ASs (Autonomous Systems), ensuring that no loops are formed during the transmission of routes, and is used for route optimization.

[0131] In this embodiment, the BGP unit sends instructions based on the routing entry to obtain the corresponding route, which includes the destination address, next-hop address, and AS PATH information.

[0132] In this embodiment, the BGP unit sends the route to the switch through the data forwarding unit, including:

[0133] S25-1: The BGP unit sends the first route to the switch through the data forwarding unit. The destination address corresponding to the first route is the virtual address, the next-hop address corresponding to the first route is the interconnect address, and the AS PATH information is empty.

[0134] S25-2: The BGP unit sends the second route to the switch through the data forwarding unit. The destination address corresponding to the second route is the network element node address of the network element node, the next-hop address corresponding to the second route is the interconnection address, and the AS PATH information is empty.

[0135] S25-3: The BGP unit sends the third route to the switch through the data forwarding unit. The destination address corresponding to the third route is the network element node address of all network element nodes in the network element cluster except the network element node where the BGP unit is located. The next-hop address corresponding to the third route is the interconnection address. The AS PATH information corresponding to the third route is the AS number of the network element node.

[0136] In this embodiment, the BGP module implements the BGP protocol and provides the controller with an interface to operate BGP routing entries. The controller sends the routing entries to be published to the BGP module. Each routing entry includes the destination address, the next-hop address, and AS PATH information. The BGP module, based on instructions from the network element controller module, publishes three types of routes: 1. A VIP route is published, with the destination address being the network element cluster VIP address and the next-hop address being the interconnecting IP address of this network element, and the AS PATH being empty; 2. A BIP route is published, with the destination address being the BIP address of this network element node and the next-hop address being the interconnecting IP address of this network element; 3. (Number of network elements in this cluster - 1) BIP routes are published, with the destination address being the BIP address of other network elements in this cluster besides the BIP of this node, the next-hop address being the interconnecting IP address of this network element, and the AS PATH being 15 AS numbers of this node (which can be adjusted according to the physical network topology). The AS PATH represents the path to this destination address. Here, the AS PATH is longer than the AS PATH of the original BIP node. Therefore, when selecting the next hop of the route, the switch will prioritize the path with the shorter AS PATH. That is, traffic packets sent through the BIP of this node will return to this node first. Only when this node fails will the traffic of the BIP be load-balanced and sent to other network element nodes in this cluster. The BGP module uses the virtual interface network card provided by the data forwarding module to establish BGP neighbor relationships with the switch module using the interconnection IP address of the network element according to the pre-configured BGP peer information, and uses the BGP protocol to pass routing information to the switch module.

[0137] S26: The BGP unit sends the route to the switch through the data forwarding unit.

[0138] In this embodiment, the BGP unit sends the corresponding route to the switch through the data forwarding unit.

[0139] In another embodiment of this application, the method further includes:

[0140] S31: The network element cluster establishes a synchronization session between each network element node based on the interconnection address corresponding to each network element node.

[0141] In this embodiment, each network element node in the network element cluster establishes a synchronization session based on the interconnection address. This synchronization session enables data transmission between each network element node, maintaining data synchronization.

[0142] S32: Synchronize the session information of data traffic in each of the network element nodes through the synchronization session.

[0143] In this embodiment, each network element node synchronizes its data traffic session information to the other nodes in the cluster at regular intervals through a synchronization session, thereby ensuring that when a network element node fails, the other network element nodes can continue to execute the data processing tasks of that network element node.

[0144] In this embodiment, multiple data forwarding modules within the cluster use the interconnection IP address of this node to construct session synchronization messages to perform session synchronization between multiple network element nodes within the cluster. The synchronization includes the source IP, destination IP, source port, destination port, protocol, etc. in the VXLAN message. The session synchronization is configured to send a synchronization message every 3 seconds or every 100 newly established sessions. This configuration can be adjusted according to the actual traffic scenario, so that the session will be synchronized to other network elements in the cluster within a second-level period. This session synchronization is only required when the network element fails after the session is established. When a failure occurs, the session has usually been established and synchronized to each network element. Therefore, the session does not require strong consistency within the cluster, and the performance of newly established sessions will not be affected by waiting for session synchronization.

[0145] In another embodiment of this application, when the network element node is operating normally, before the switch sends the response data sent by the downstream component to the network element node according to the second route published by the network element node, the method further includes:

[0146] S41: The switch receives heartbeat information sent by the BGP unit on the network element node.

[0147] In this embodiment, the BGP unit on the switch receives heartbeat information sent by the BGP unit on the network element node.

[0148] S42: If the switch receives the heartbeat information normally, determine that the network element node is operating normally.

[0149] S43: If the switch cannot receive the heartbeat information normally, determine that the network element node is malfunctioning.

[0150] In this embodiment, if the switch receives heartbeat information normally, it is determined that the network element node is operating normally; if the switch cannot receive heartbeat information normally, it is determined that the network element node is operating abnormally.

[0151] In this embodiment, the method further includes:

[0152] S44: If it is determined that the network element node is malfunctioning, the switch disconnects from the network element node.

[0153] S45: Delete the routing information sent by the network element node.

[0154] In this embodiment, if it is determined that the network element node is malfunctioning, the switch disconnects from the network element node and deletes the routing information sent by the network element node, including the virtual address and the network element node address.

[0155] In this embodiment, the data forwarding module creates a virtual interface network card for the BGP module and establishes a virtual link between the BGP module and the switch module. This link forwards packets sent by the BGP module to the switch module through the physical network card, and delivers BGP packets sent by the switch module to the physical network card to the virtual interface for reception by the BGP module. The virtual interface is implemented by the data forwarding module; therefore, when the data forwarding module malfunctions and cannot forward traffic, BGP packets will also fail to forward. The switch module will be able to detect BGP heartbeat packet timeouts, determine that the network element node is abnormal, and automatically delete the VIP and BIP address routes published by the faulty network element, achieving automatic fault switching without relying on external components to detect and check the network element status.

[0156] In another embodiment of this application, a highly available load-balanced NAT network element cluster is proposed, comprising:

[0157] A three-node MySQL Gelara database cluster is set up as the network element configuration database. The database includes three tables: the host table stores the hostname, interconnection IP address, BIP address, and the network element cluster ID to which it belongs; the cluster table stores the ID, name, VIP address, and type of the network element cluster; and the nat_rule table stores the NAT forwarding rules of the network element, including NAT type, source IP, destination IP, port number, etc.

[0158] The network element controller is implemented using the Go programming language. Deployed on each network element node, the controller connects to a database cluster to read network element configuration data. It uses a local socket connection to the data forwarding module to distribute forwarding rules and IP address information. The open-source gobgpd program, used for the BGP module, distributes VIP and BIP routing entries. In the gobgpd configuration, the uplink switch is configured as the BGP peer according to the network plan, serving as the object for route advertising. The data forwarding module is developed using C language and the DPDK (Data Plane Development Kit) framework. It receives data packets with destination IPs of VIP and BIP, parses VXLAN data packets, and performs SNAT and DNAT on VXLAN inner-layer data packets according to NAT forwarding rules. The switch module uses a box-type switch supporting the standard EBGP protocol. VIP and BIP address ranges are planned for upstream and downstream communication. Interconnect IP addresses are planned to enable communication between network element interconnection IPs and switch virtual interfaces. BGP is configured to receive routes from the BGP module, and ECMP is configured to enable load balancing packet forwarding based on five-tuple equal-cost routing. Three network element nodes are deployed to achieve high availability and load balancing. Subsequently, the horizontal scaling of the nodes will be adjusted in a timely manner based on the network interface card traffic monitoring of the network element nodes to support different throughput requirements in the future.

[0159] In the above embodiments of this application, interconnect addresses, virtual addresses, and network element addresses are planned and stored through a network element configuration database unit. The controller controls the BGP publishing of VIPs and BIPs. The data forwarding unit receives VIP requests, processes the data, and sends data to the backend using BIPs. The backend then replies with data to the original network element node, achieving load balancing deployment of stateful network elements and supporting horizontal scaling to meet high-throughput network requirements. Furthermore, by using BIP route priority publishing with ASPATH and network element forwarding unit session synchronization, when one network element in the cluster fails, the traffic and existing sessions carried on that network element can be automatically switched to other network element nodes in the cluster, ensuring uninterrupted service and achieving high availability of network elements. This improves network service quality without requiring strong consistency synchronization, and session synchronization does not affect the performance of new network element sessions. A virtual interface network card is generated based on the data forwarding unit, serving as the link between the BGP unit and the switch BGP unit. When the data forwarding unit fails, the BGP link automatically disconnects, achieving automatic switching of network element forwarding failures without relying on external components and with high switching accuracy. The above methods achieve network element load balancing while ensuring high availability of network elements and improving the accuracy of data forwarding.

[0160] Based on the same inventive concept, one embodiment of this application provides a network element load balancing device. (Reference) Figure 7 , Figure 7 This is a schematic diagram of a network element load balancing device 700 according to an embodiment of this application. Figure 7 As shown, the device includes:

[0161] The data receiving module 701 is used for the switch to receive traffic data sent by upstream components;

[0162] The first data sending module 702 is used by the switch to send the traffic data to at least one of the network element nodes in a load-balanced manner according to the first route published by at least one network element node in the network element cluster.

[0163] The second data sending module 703 is used for the network element node to send the processed traffic data to downstream components through the switch;

[0164] The third data sending module 704 is used to send the response data sent by the downstream component to the network element node according to the second route published by the network element node when the network element node is operating normally.

[0165] The fourth data sending module 705 is used to, in the event of an abnormal operation of the network element node, send the response data sent by the downstream component to the remaining network element nodes in the network element cluster according to the third route published by the network element node.

[0166] The fifth data sending module 706 is used by the network element cluster to send the response data to the upstream component through the switch.

[0167] Optionally, the device further includes:

[0168] The configuration data acquisition module is used by the controller unit of the network element node to acquire network element configuration data from the management server. The network element configuration data includes the network element host name, network element cluster identifier, virtual address, interconnection address, network element node address, and network element forwarding rules.

[0169] The first configuration data sending module is used by the controller unit to send the network element forwarding rules, the virtual address, the interconnection address and the network element node address to the data forwarding unit of the network element node;

[0170] The second configuration data sending module is used by the controller unit to send the virtual address and the network element address to the BGP unit in the network element node.

[0171] Optionally, the device further includes:

[0172] The instruction sending module is used for the network element controller unit to send a routing entry sending instruction to the BGP unit;

[0173] The route acquisition module is used by the BGP unit to obtain the corresponding route by sending an instruction according to the route entry. The route includes the destination address, the next-hop address, and AS PATH information.

[0174] A route sending module is used by the BGP unit to send the route to the switch through the data forwarding unit.

[0175] Optionally, the routing sending module includes:

[0176] The first route sending submodule is used by the BGP unit to send the first route to the switch through the data forwarding unit. The destination address corresponding to the first route is the virtual address, the next-hop address corresponding to the first route is the interconnect address, and the AS PATH information is empty.

[0177] The second route sending submodule is used by the BGP unit to send the second route to the switch through the data forwarding unit. The destination address corresponding to the second route is the network element node address of the network element node, the next hop address corresponding to the second route is the interconnection address, and the AS PATH information is empty.

[0178] The third route sending submodule is used by the BGP unit to send the third route to the switch through the data forwarding unit. The destination address corresponding to the third route is the network element node address of all network element nodes in the network element cluster except the network element node where the BGP unit is located. The next hop address corresponding to the third route is the interconnection address. The AS PATH information corresponding to the third route is the AS number of the network element node.

[0179] Optionally, the device further includes:

[0180] The session synchronization module is used by the network element cluster to establish a synchronization session between each network element node based on the interconnection address corresponding to each network element node;

[0181] The information synchronization module is used to synchronize the session information of data traffic in each of the network element nodes through the synchronization session.

[0182] Optionally, the device further includes:

[0183] A heartbeat information receiving module is used by the switch to receive heartbeat information sent by the BGP unit on the network element node;

[0184] The first state determination module is used to determine that the network element node is operating normally when the switch normally receives the heartbeat information;

[0185] The second state determination module is used to determine that the network element node is malfunctioning when the switch cannot receive the heartbeat information normally.

[0186] Optionally, the device further includes:

[0187] A connection disconnection module is used to disconnect the switch from the network element node when it is determined that the network element node is malfunctioning.

[0188] The information deletion module is used to delete the routing information sent by the network element node.

[0189] Based on the same inventive concept, another embodiment of this application provides a readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the network element load balancing method as described in any of the above embodiments of this application.

[0190] Based on the same inventive concept, another embodiment of this application provides an electronic device. Figure 8 This is a schematic diagram of an electronic device 800 according to an embodiment of this application, including a memory 802, a processor 801, and a computer program stored in the memory and executable on the processor. When executed by the processor, the program implements the steps in the network element load balancing method described in any of the above embodiments of this application.

[0191] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0192] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0193] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, or computer program products. Therefore, embodiments of this application can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of this application can take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0194] This application describes embodiments with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0195] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0196] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, causing a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0197] Although preferred embodiments of the present application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present application.

[0198] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0199] The network element load balancing method, apparatus, device, and storage medium provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for load balancing of network elements, the method comprising: The method comprises: The switch receives traffic data sent by an upstream component; The switch sends the traffic data to at least one of the network element nodes in the network element cluster in a load-balanced manner according to a first route published by at least one of the network element nodes in the network element cluster; The network element node sends processed traffic data to a downstream component through the switch; In the case where the network element node is operating normally, the switch sends response data sent by the downstream component to the network element node according to a second route published by the network element node; In the case where the network element node is operating abnormally, the switch sends the response data sent by the downstream component to the remaining network element nodes in the network element cluster according to a third route published by the network element node; The network element cluster sends the response data to the upstream component through the switch; Before the switch sends the traffic data to at least one of the network element nodes in the network element cluster in a load-balanced manner according to a first route published by at least one of the network element nodes in the network element cluster, the method further comprises: A controller unit of the network element node acquires network element configuration data from a management server, the network element configuration data comprising a network element host name, a network element cluster identifier, a virtual address, an interconnection address, a network element node address, and a network element forwarding rule; The controller unit sends the network element forwarding rule, the virtual address, the interconnection address, and the network element node address to a data forwarding unit of the network element node; The controller unit sends the virtual address and the network element address to a BGP unit in the network element node; The method further comprises: The network element controller unit sends a route entry sending instruction to the BGP unit; The BGP unit acquires a corresponding route according to the route entry sending instruction, the route comprising a destination address, a next-hop address, and AS PATH information; The BGP unit sends the route to the switch through the data forwarding unit; The BGP unit sends the route to the switch through the data forwarding unit, comprising: The BGP unit sends the first route to the switch through the data forwarding unit, the destination address corresponding to the first route being the virtual address, the next-hop address corresponding to the first route being the interconnection address, and the AS PATH information being empty; The BGP unit sends the second route to the switch through the data forwarding unit, the destination address corresponding to the second route being the network element node address of the network element node, the next-hop address corresponding to the second route being the interconnection address, and the AS PATH information being empty; The BGP unit sends the third route to the switch through the data forwarding unit, the third route corresponds to the network element node addresses of all network element nodes in the network element cluster except the network element node where the BGP unit is located, the third route corresponds to the interconnection address as the next hop address, and the third route corresponds to the AS number of the network element node as the AS PATH information.

2. The network element load balancing method of claim 1, wherein, The method further comprises: The network element cluster establishes a synchronization session between each network element node according to the interconnection address corresponding to each network element node. Through the synchronization session, the session information of the data flow in each network element node is synchronized.

3. The method of claim 1, wherein, In the case that the network element node is in normal operation, the method further comprises: The switch receives heartbeat information sent by the BGP unit on the network element node; In the case that the switch normally receives the heartbeat information, it is determined that the network element node is in normal operation; In the case that the switch cannot normally receive the heartbeat information, it is determined that the network element node is in abnormal operation.

4. The method of claim 3, wherein, The method further comprises: In the case that it is determined that the network element node is in abnormal operation, the switch disconnects the connection with the network element node; The routing information sent by the network element node is deleted.

5. A network element load balancing apparatus, characterized by, The device comprises: A data receiving module for the switch to receive traffic data sent by an upstream component; A first data sending module for the switch to send the traffic data to at least one network element node in a network element cluster according to a first route published by the at least one network element node; A second data sending module for the network element node to send processed traffic data to a downstream component through the switch; A third data sending module for the switch to send response data sent by the downstream component to the network element node according to a second route published by the network element node in the case that the network element node is in normal operation; A fourth data sending module for the switch to send the response data sent by the downstream component to the remaining network element nodes in the network element cluster according to a third route published by the network element node in the case that the network element node is in abnormal operation; A fifth data sending module for the network element cluster to send the response data to the upstream component through the switch; The device further comprises: A configuration data obtaining module for a controller unit of the network element node to obtain network element configuration data from a management server, the network element configuration data comprising a network element host name, a network element cluster identifier, a virtual address, an interconnection address, a network element node address, and a network element forwarding rule; A first configuration data sending module for the controller unit to send the network element forwarding rule, the virtual address, the interconnection address, and the network element node address to a data forwarding unit of the network element node; a second configuration data sending module, configured to send, by the controller unit, the virtual address and the network element address to a BGP unit in the network element node; an instruction sending module, configured to send, by the network element controller unit, a routing entry sending instruction to the BGP unit; a route obtaining module, configured to obtain, by the BGP unit, a corresponding route according to the routing entry sending instruction, the route including a destination address, a next-hop address, and AS PATH information; a route sending module, configured to send, by the BGP unit, the route to the switch through the data forwarding unit; the route sending module includes: a first route sending submodule, configured to send, by the BGP unit, a first route to the switch through the data forwarding unit, the first route corresponding to a virtual address as the destination address, the first route corresponding to the interconnection address as the next-hop address, and the AS PATH information being empty; a second route sending submodule, configured to send, by the BGP unit, a second route to the switch through the data forwarding unit, the second route corresponding to a network element node address of the network element node as the destination address, the second route corresponding to the interconnection address as the next-hop address, and the AS PATH information being empty; a third route sending submodule, configured to send, by the BGP unit, a third route to the switch through the data forwarding unit, the third route corresponding to network element node addresses of all network element nodes in the network element cluster except the network element node where the BGP unit is located as the destination addresses, the third route corresponding to the interconnection address as the next-hop address, and the third route corresponding to the AS number of the network element node as the AS PATH information.

6. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by a processor to implement the steps in the method of any one of claims 1 to 4.

7. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the steps in the method of any one of claims 1 to 4.

Images