Dynamic routing optimization method for hybrid forwarding plane network

Abstract

The invention discloses a dynamic routing optimization method for a hybrid forwarding plane network, which is characterized in that a centralized controller broadcasts a whole network flow table to routing equipment in a network domain in a flooding mode, and the routing equipment in the network domain introduces the whole network flow table in calculating a shortest path in a dynamic routing calculation process so as to optimize the dynamic routing. And optimizing and generating a routing table and conflict information, reporting the conflict information to the centralized controller by the routing equipment, and optimizing a service guarantee strategy by the centralized controller. The method has the beneficial effects that in a communication network based on a hybrid forwarding plane architecture, high-quality generation of a dynamic routing path is realized, a routing next hop is preferably formed, and the problems of network congestion and transmission delay increase generated after a centralized controller manages and controls the communication network are avoided.

Description

Technical Field

[0001] The present invention relates to a dynamic routing optimization technology, and in particular to a dynamic routing optimization method for a hybrid forwarding plane network. Background Art

[0002] Dynamic routing is a mechanism by which network devices automatically learn routing information and dynamically adjust routes through protocols. Unlike static routing, dynamic routing protocols periodically exchange topology information such as link status and distance vectors, updating routing tables in real time based on network changes such as link failures and cost adjustments. Dynamic routing protocols offer advantages such as adaptive network path adjustments, eliminating the need for manual routing configuration. Typical dynamic routing protocols include OSPF and BGP.

[0003] OSPF (Open Shortest Path First) is a link-state-based interior gateway protocol and a dynamic routing protocol suitable for enterprise and carrier networks. It calculates loop-free shortest paths by building a network-wide topology database and running the SPF algorithm.

[0004] SPF (Shortest Path First), also known as the Dijkstra algorithm, is a classic algorithm used to calculate the shortest path from one node to all other nodes in a network. It is the core calculation logic of OSPF.

[0005] Link State Advertisement (LSA) is the core data unit of the OSPF protocol, used to describe the router's interface status, neighbor relationships, and network topology information.

[0006] Software-defined networking, or SDN for short, is a new network architecture model. Its core idea is to separate the network's control plane from the data forwarding plane, thereby giving the network greater flexibility, adaptability, and centralized management capabilities. SDN is a key technology for future network automation.

[0007] Software-defined networking and dynamic routing are both network management and scheduling technologies. Dynamic routing, while a relatively early adopter, is a foundational network technology with mature applications and ongoing optimizations to accommodate IPv6 and SDN. Software-defined networking is a new network technology, widely used in data centers and well-proven, and is gradually being adopted and implemented. Technically, SDN offers high flexibility, supporting network virtualization and automated management, while dynamic routing adapts to changing network topologies and offers strong stability. The two can complement each other in specific scenarios, with SDN leveraging dynamic routing protocols to enhance network functionality.

[0008] With the development of technology, the integration and application of software-defined networking technology and dynamic routing technology have formed a new "one network" network architecture with multiple forwarding planes. This network will simultaneously have multiple forwarding plane control technologies. In the path decision-making of the dynamic routing forwarding plane, there may be unreasonable network path scheduling, such as circuitous routing. Summary of the Invention

[0009] In response to the problems raised in the background technology, the present invention discloses a dynamic routing optimization method for a hybrid forwarding plane network, comprising a centralized controller and routing devices, wherein the centralized controller transmits a network-wide network flow table to each routing device via a first flooding method. The routing devices transmit the network-wide network flow table and control the flooding of the network-wide network flow table via identifiers to avoid flooding storms. The routing devices store the network-wide network flow table separately from the controlled forwarding flow table, and the network-wide network flow table does not affect the data forwarding process of the routing devices. In dynamic routing, the routing devices use the network-wide network flow table to influence the calculation results of the dynamic routing.

[0010] This method optimizes the routing generation process of dynamic routing, mainly optimizing the network topology generation and SPF calculation process. The optimized dynamic routing information generation steps are as follows: S101: Build a link state database. Each router builds its own link state database based on the received LSAs to form a connection topology diagram of the network. S102: Using this node as the starting node, select other nodes as target nodes for calculating the path; S103, extracting network segment information from the target node, and using the network segment as the target network segment for path calculation; S104. Calculate all paths from the starting node to the target network segment. In the calculated paths, match the network flow table of the entire network that includes the target network segment. If the first node in the path is the same as the starting node in the network flow table of the entire network, use the next node of the first node in the network flow table of the entire network as the next node of the first node in the path calculation. S105, obtaining the shortest path from the starting node to the target network segment; S106: Determine the next node of the starting node in the shortest path as the next hop, and generate target network segment routing information; S107. Repeat steps S103 to S106, and calculate all network segments of the target node as the target network segments. S108. Repeat steps S102 to S107, calculate all nodes except the current node as target nodes, and merge the routing information of all target nodes to form a complete routing table of the current node.

[0011] The current node, target node, and starting node are all routing devices.

[0012] In the step of generating optimized dynamic routing information, if the first node included in the path calculation is the same as the starting node in the network flow table for the entire network, the next node of the first node in the network flow table for the entire network is used as the next node of the first node in the path calculation to participate in the calculation. After the calculation is completed, the shortest path from the starting node to the target network segment is obtained according to the path, and the next node of the starting node in the shortest path is determined as the next hop. This method designs a conflict reporting mechanism. After the next hop is determined, if the next hop calculated according to the network connection topology SPF is inconsistent with the next hop after dynamic routing optimization, the conflict information is reported through a second flooding method. The conflict information includes the unique identifier of the conflict information, the conflicting node, the conflicting network segment, the SPF next node, and the next node after dynamic routing optimization.

[0013] This method designs a first flooding mode to send the network flow table of the entire network to each routing device. Specifically, the centralized controller encapsulates the network flow table of the entire network into a flow table IP multicast message and sends it to the routing device. The flow table IP multicast message includes a specific multicast address, port number and a unique identifier of the network flow table of the entire network. The routing device will forward the flow table IP multicast message to all interfaces except the one that receives the flow table IP multicast message, and will not repeatedly forward the flow table IP multicast message with the same unique identifier information of the network flow table of the entire network.

[0014] This method designs a transmission scheme for reporting conflict information in a second flooding mode. Specifically, the routing device encapsulates the conflict information into a conflict IP multicast message. The conflict IP multicast message is identified and transmitted through a specific multicast address, port number and conflict information unique identifier. The routing device forwards the conflict IP multicast message to all interfaces except the interface that receives the conflict IP multicast message, and does not repeatedly forward the conflict IP multicast message with the same conflict information unique identifier. The conflict information unique identifier is a data bit designed for the data field in the conflict IP multicast message.

[0015] Routing devices store network-wide conflict information. The centralized controller queries routing devices for network-wide conflict information through unicast messages. It also supports subscribing to conflict information from routing devices. After subscription, the routing devices send the received conflict information to the centralized controller in real time. The centralized controller then optimizes the service assurance strategy based on the conflict information.

[0016] The beneficial effect of the present invention is that, in a communication network based on a hybrid forwarding plane architecture, the network flow table of the entire network is integrated into the dynamic routing table generation process, thereby achieving high-quality generation of dynamic routing paths, preferably forming the next hop of routing, and avoiding the network congestion and increased transmission delay problems caused by the centralized controller's control over the communication network. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide further understanding of the present invention and constitute a part of this application. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0018] Attachment Figure 1 、 one A typical communication network topology diagram; Attachment Figure 2 , schematic diagram of forwarding flow table delivery; Attachment Figure 3 , schematic diagram of path information generated by OSPF; Attachment Figure 4 ,Path diagram of typical data services before dynamic routing optimization; Attachment Figure 5 , Schematic diagram of path information generated after OSPF optimization; Attachment Figure 6 ,Path diagram of typical data services based on dynamic routing optimization. DETAILED DESCRIPTION

[0019] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0020] See attached Figure 1 As shown, a typical communication network topology diagram includes six network routers A, B, C, D, E, and F. The present invention is described as a typical case based on the communication network topology diagram. Unless otherwise specified, the communication network mentioned in the invention will be described with this topology diagram.

[0021] See attached Figure 2 As shown in the schematic diagram of forwarding flow table delivery, the centralized controller delivers the network flow table to B, where the content of the flow table is that the next hop forwarding to the network segment address contained in F is C.

[0022] See attached Figure 3 Figure 2 shows a schematic diagram of the path information generated by OSPF. Based on the dynamic routing protocol OSPF, it dynamically generates information about the network segment addresses of each network router to F. OSPF calculates the shortest path using the SPF algorithm. The figure shows the next hop information from each network routing device to F: Routing information in A: to F, the next hop address is B; Routing information in B: to F, the next hop address is D; Routing information in C: to F, the next hop address is E; Routing information in D: to F, the next hop address is F; Routing information in E: to F, the next hop address is D; Routing information in F: None.

[0023] See attached Figure 4 As shown in the figure, a typical data service is based on a path diagram before dynamic routing optimization. Data flowing from A to F is transmitted in the communication network. The path process starting from A and the selection basis are as follows: When the data is at A, it will be handed over to B according to the dynamic routing protocol; When the data is at B, it will be delivered to C according to the service path issued by the centralized controller; When the data is at C, it will be handed over to E according to the dynamic routing protocol; When the data is at E, it will be handed over to D according to the dynamic routing protocol; When the data is at D, it will be handed over to F according to the dynamic routing protocol.

[0024] See attached Figure 5 Figure 2 shows the path information generated after OSPF optimization. Based on the dynamic routing protocol OSPF and the network flow table of the centralized controller, the information about the network segment addresses from each network router to F is dynamically generated. OSPF calculates the shortest path based on the SPF algorithm. The figure shows the next hop information from each network routing device to F: Routing information in A: to F, the next hop address is C; Routing information in B: to F, the next hop address is D; Routing information in C: to F, the next hop address is E; Routing information in D: to F, the next hop address is F; Routing information in E: to F, the next hop address is D; Routing information in F: None.

[0025] See attached Figure 6 As shown in the figure, a typical data service is based on a path diagram after dynamic routing optimization. Data flowing from A to F is transmitted in the communication network. The path process starting from A and the selection basis are as follows: When the data is at A, it will be handed over to C according to the dynamic routing protocol; When the data is at C, it will be handed over to E according to the dynamic routing protocol; When the data is at E, it will be handed over to D according to the dynamic routing protocol; When the data is at D, it will be handed over to F according to the dynamic routing protocol.

Claims

1. A dynamic routing optimization method for a hybrid forwarding plane network, characterized in that: The invention comprises a centralized controller and a routing device, wherein the centralized controller sends a network flow table of the entire network to each routing device through a first flooding mode; the routing device transmits the network flow table of the entire network and controls the flooding of the network flow table of the entire network by means of an identifier to avoid a flooding storm; the routing device stores the network flow table of the entire network separately from the controlled forwarding flow table, and the network flow table of the entire network does not affect the data forwarding process of the routing device; In dynamic routing, the routing device uses the network flow table of the entire network to influence the calculation results of the dynamic routing. The steps for generating dynamic routing information are as follows: S101: Build a link state database. Each router builds its own link state database based on the received LSAs to form a network connection topology. S102: Using this node as the starting node, select other nodes as target nodes for calculating the path; S103, extracting network segment information from the target node, and using the network segment as the target network segment for path calculation; S104. Calculate all paths from the starting node to the target network segment. In the calculated paths, match the network flow table of the entire network that includes the target network segment. If the first node in the path is the same as the starting node in the network flow table of the entire network, use the next node of the first node in the network flow table of the entire network as the next node of the first node in the path calculation. S105, obtaining the shortest path from the starting node to the target network segment; S106: Determine the next node of the starting node in the shortest path as the next hop, and generate target network segment routing information; S107. Repeat steps S103 to S106, and calculate all network segments of the target node as the target network segments. S108. Repeat steps S102 to S107, calculate all nodes except the current node as target nodes, and merge the routing information of all target nodes to form a complete routing table of the current node.

2. A dynamic routing optimization method for a hybrid forwarding plane network according to claim 1, characterized in that: A conflict reporting mechanism is designed. After the next hop is determined, if the next hop calculated according to the SPF of the network connection topology is inconsistent with the next hop after dynamic routing optimization, the conflict information is reported through the second flooding method. The conflict information contains the unique identifier of the conflict information, the conflicting node, the conflicting network segment, the SPF next node, and the next node after dynamic routing optimization.

3. The method for dynamic routing optimization for a hybrid forwarding plane network according to claim 1, wherein: The centralized controller encapsulates the network flow table of the entire network into a flow table IP multicast message and sends it to the routing device. The flow table IP multicast message includes a specific multicast address, port number and a unique identifier of the network flow table of the entire network. The routing device will forward the flow table IP multicast message to all interfaces except the one that receives the flow table IP multicast message, and will not repeatedly forward the flow table IP multicast message with the same unique identifier information of the network flow table of the entire network.

4. A dynamic routing optimization method for a hybrid forwarding plane network according to claims 1 and 2, characterized in that: The routing device encapsulates the conflict information into a conflict IP multicast message. The conflict IP multicast message is identified and transmitted through a specific multicast address, port number and conflict information unique identifier. The routing device forwards the conflict IP multicast message to all interfaces except the one that receives the conflict IP multicast message, and does not repeatedly forward the conflict IP multicast message with the same conflict information unique identifier. The conflict information unique identifier is a data bit designed for the data field in the conflict IP multicast message.

Images