74 results about "Convergence (routing)" patented technology

A system and method for network routing is provided where significant (those that impact optimal network routes) state changes of network components are considered. A set of optimal communication paths are generated for a number of actual and potential component failure scenarios. An optimal communication path is generated for each failure scenario. In addition, a method that enables continued communication using intermediate routing points and routing update propagation during periods of network non-convergence or congestion is disclosed.

In a network comprising a provider edge router coupled to each of a mated pair of core routers in a core network, a system and methodology for rerouting upstream traffic destined for the provider edge router in the event of a link failure between one of the core routers and the provider edge router. By detecting a link failure between a first of the mated pair of core routers and the provider edge router, and directing the upstream traffic destined for the provider edge router to a second of the mated pair of core routers, internal gateway protocol (IGP) reconvergence events are not triggered.

The invention belongs to the technical field of artificial intelligence, and particularly relates to a network autonomous intelligent management and control method based on deep reinforcement learning. The method comprises the steps of firstly constructing a network topology, then introducing a CNN, an LSTM layer and a delay updating strategy to construct a routing decision-making model based on a DDPG reinforcement learning algorithm, and finally performing iterative training on the routing decision-making model based on deep reinforcement learning. In each iterative training, an intelligent agent obtains an output action, namely a group of link weights, according to a measured network state and a neural network, and calculates a service route by using a shortest path algorithm according to the link weights. According to a routing calculation result, the intelligent agent issues a flow table, and acquires end-to-end time delay and a packet loss probability of the service to calculate a reward value of the iteration. The algorithm has good convergence, and can effectively reduce the end-to-end delay and packet loss rate of the service.

The invention discloses a link switching method and a device thereof, which is applied to an MPLS(Multi-Protocol Label Switching) L2VPN system including a UPE (User-facing Provider Edge) and an NPE (Network Provider Edge). The method comprises the following steps: the NPE detects the connectivity of master and slave U-PWs (Pseudo Wires) and the connectivity of an N-PW; when the NPE detects U-PW faults, an MAC (Media Access Control) address clear command is sent through the N-PW in the VSI (Visual Switch Instance) of the U-PWs, so that the NPE which receives the MAC address clear command clears the corresponding MAC address in the VSI and associated with the N-PW which transmits the MAC address clear command; when the NPE detects N-PW faults, a flow rerouting command is sent through the U-PWs in the VSI of the N-PW, so that the UPE which receives the flow rerouting command carries out U-PW switch. By using the invention, the convergence speed of the UPE of the MPLS L2VPN can be accelerated.

The invention discloses a load balancing method and device for multiple transmission lines. The method comprises the following steps: receiving an initial application message sent to a second networkconnected with an access station by a first network connected with a convergence station; if it is determined that the first routing device has the main device attribute, obtaining path scheduling information of the initial application message according to a preset load balancing scheduling strategy, the first line information and at least one part of second line information of at least one secondtransmission line, thereby determining a forwarding path for forwarding the initial application message; If it is determined that the first routing device has the standby device attribute, packagingthe initial application message in a first preset packaging mode to obtain a to-be-forwarded application message, and forwarding the to-be-forwarded application message to a selected second routing device having the main device attribute, so that the selected second routing device determines a forwarding path for forwarding the initial application message. According to the method, the load balancing problem of a plurality of transmission lines of routing equipment of convergence sites deployed at different places is solved.

The embodiment of the invention discloses a rapid routing convergence method and device. The method comprises the steps of S1, obtaining a class to which a change situation of a network topology belongs according to the shortest path tree (SPT) of a network, a link status database (LSDB) and neighbor node information contained in an LSP when any routing node in the network receives the link statuspacket (LSP); and S2, performing a corresponding updating policy according to the type of the change situation. According to the embodiment of the invention, the change situation of the network topology is classified, so for each class of topology change, a suitable algorithm is selectively used to update a route, and the unnecessary route computation quantity is reduced. Compared with the priorart, the embodiment of the invention has the advantages that the route computation time is reduced and the CPU load is reduced.

The invention discloses a method for generating a forwarding table item, a method for sending a message, network equipment and a system, which can realize that specified network equipment does not use a standby forwarding path to forward the message, thereby reducing the technical problems of network resource waste or network congestion caused by a loop problem to a certain extent. The method comprises the following steps that: a first network device obtains routing information published by a second network device; the first network device confirms that the routing information published by the second network device is matched with a convergence route stored by the first network device, and generates a forwarding table item corresponding to the convergence route according to the segment identifier of the second network device, the forwarding table item comprising the segment identifier of the second network device, a forwarding path corresponding to the forwarding table item passes through the second network equipment.

The invention belongs to the technical field of cross-domain routing, and discloses a seamless credible cross-domain routing system of a heterogeneous convergence network and a control method thereof. The control method of the seamless credible cross-domain routing system of the heterogeneous convergence network comprises the following steps: evaluating the trust value of a network node through a trust calculation method and an incentive mechanism, and exciting the honesty behavior of the network node; issuing a intelligent contract for cross-domain routing, and determining a network state update process; designing an AR-DRL algorithm, wherein the algorithm is based on a block chain and a DDPG algorithm to obtain an optimal routing strategy, and seamless cross-domain routing is realized. According to the invention, automatic cross-domain routing is realized through an intelligent contract and a machine learning algorithm; the credibility of routing nodes is improved through an incentive mechanism and a consensus mechanism; according to the network state information provided by the operator in the registration process, the authenticity of the network state information in the routing process is evaluated remotely, counterfeiting and serial attack of the network operator can be resisted, and the authenticity and reliability of the routing information are ensured.

The invention provides a dynamic self-adaptive route path planning method based on a two-layer topology routing algorithm. The method comprises: S1, building a traffic network; S2, grouping the vehicle node and the lamp pole node; S3, calculating a routing path between the vehicle node and the lamp pole node; S4, evaluating the performance of each route path between the vehicle node and the lamp pole node; S5, calculating the effective link and the availability weight of the routing path of the ground intelligent lamp pole network; S6, calculating the total cost of data transmission of the terrestrial intelligent lamp pole network, and establishing data transmission by the route path with the minimum total cost and the best availability weight value; S7, screening the vehicle node and thelight pole node the meet the service requirements to participate in data transmission, and obtaining the topological routing path of the traffic network. The invention discloses a dynamic self-adaptive route path planning method based on a two-layer topology route algorithm, which can improve the work utilization efficiency of a traffic network, solve the information congestion, enhance the convergence, reduce the route overhead and improve the service satisfaction.

The invention discloses an L3VPN routing iteration outlet method. The method is characterized in that an FIB of a forwarding plane can be layered and separated into a VC layer and an LSP layer respectively for iteration processing, and each layer is used to determine whether a next hop IP of current unicast routing and a direct-connected routing are in a same network segment; when the answer is yes, the forwarding information package processing can be carried out directly; otherwise, the next hop IP is used to execute a PEER command for the iteration, till the next hop IP and the direct-connected routing are in the same network segment, and then the forwarding information package processing can be carried out for the later searching of the routing tables and the searching of the PW or LSP labels. According to the invention, by layering and separating the FIB table of the forwarding plane, the routing, the PW, and the TUNNEL can be separated, and then when the main control data is changed, the independent data updating of the forwarding plane can be carried out, and the updating of a plurality of table item contents of the forwarding plane is not required during the updating of one data, and then the control of the business convergence time can be facilitated, when the network faults of the forwarding plane occurs.

The invention provides a method for realizing the shortest path of double layers based on a cluster computing unit. The method comprises the following steps: adding node identification information responsible for each cluster computing unit layer in a flooding information packet of the cluster computing unit layer according to a routing protocol expanding the cluster computing unit layer so as to improve the permeability of layer topology of the cluster computing unit layer, calculating the shortest path reaching a target cluster computing unit from a source cluster computing unit, and directly transmitting a path selection request to the target cluster computing unit along the shortest path to avoid blooding of the path selection requests, thereby greatly reducing the problem of network overloading caused by information blooding. Meanwhile, the adopted method of double layer shortest path can greatly save path selecting resources, reduce convergence time of path selection in large scale, and reduce the complexity of path selection. According to the method, the problem of excessive blooding information caused by mass path selection requests, and problems of time convergence difficulty and computing resource waste in path selection in multiple domains can be solved.

The invention provides a QG-OLSR routing method based on a quantum genetic strategy. A mobile ad hoc network has the characteristics of mobility, self-organization, multi-hop, distributed control and the like, so that a routing algorithm thereof must quickly adapt to frequent changes of a network topology structure and must continuously keep routing information with high instantaneity, high accuracy and little redundant information for each node, and the network resources can be saved to the uttermost. Compared with the existing quantum genetic algorithm, the method improves the quantum genetic algorithm, and optimizes the selection of an MPR node at first in combination with the characteristics of an OLSR routing protocol, thereby overcoming the shortcomings of the traditional method during the selection of an MPR set and proving the convergence and the global optimal solution of the algorithm. As been checked and proved, by adoption of the method, the redundant information in the network can be effectively reduced, and the data transmission efficiency in network topology is improved.

The invention discloses a high-availability load balancing method and system and a computer readable storage medium, and belongs to the technical field of communication control. The method comprises the following steps that: a core switch searches corresponding routing information according to an IP data packet of an access user and a preset routing rule; the core switch determines a routing pathpointed by a route under a corresponding convergence group according to priority selection conditions among different convergence groups and the routing information, and sends the message of the userto a routing path, and the convergence groups comprise at least one convergence switch; the convergence groups send the message to a corresponding load balancing cluster; and the load balancing cluster sends the message to a corresponding server and returns an access result according to the user IP, the load balancing cluster VIP and a preset forwarding rule, wherein the preset forwarding rule comprises forwarding according to preset source-in and source-out data structure information. According to the invention, the reliability of network service and the safety and reliability of equipment are improved.

The invention provides a distribution type flow table matching method based on software definition in a multi-identification network system, and the method comprises the following steps: S1, separating control and forwarding of a multi-identification router to form a three-layer multi-identification routing architecture comprising an application layer, a control layer and a forwarding layer; S2, realizing unification of network packet formats through a multi-identifier network packet coding mode; S3, setting a distribution type flow table structure, and controlling a flow table jumping direction through distribution of table-id and a next table-id field of a table item; S4, carrying out matching on the distributed flow tables in the multi-identifier router; S5, when the forwarding layer receives the network packet which cannot be processed, sending the content of the network packet through the packet-in information. According to the method, the requirement of cross-identification-domain routing addressing in the multi-identification network can be met, automatic deployment and maintenance of complex topology can be supported conveniently, and the method has higher routing convergence speed and lower router load.

The invention discloses a multi-domain optical network security multicast routing method based on distributed PCE. The method is used for planning a multicast tree from a source node to a destinationnode in a multi-domain optical network. Trust management and an improved ant colony method are combined on the basis of constructing a distributed PCE multicast routing model, and damage behaviors ofmalicious nodes are prevented while routing calculation is carried out. Analysis and experiment results show that compared with a typical multicast routing method, the method achieves comprehensive and excellent performance in the aspects of convergence, multicast tree cost and blocking rate in a malicious node environment.

The invention discloses a QoS routing method based on a quantum maximum and minimum ant colony algorithm. Quantum calculation and a maximum and minimum ant colony algorithm are combined; quantum coding is carried out on a link in the network topology by adopting a state vector in quantum calculation; pheromones are updated through quantum rotation entanglement characteristics in combination with amaximum and minimum inspection mechanism; a quantum selection strategy is added in roulette link selection, under the condition of small fitness difference, the evolution speed and direction are effectively controlled, the optimization capability and convergence performance are superior to those of a traditional ant colony algorithm, and the QoS routing selection problem can be effectively and accurately solved under the condition that constraint conditions are met.

The embodiment of the invention relates to the field of communication, and discloses a routing convergence method and device, communication equipment and a storage medium. In the invention, the method comprises the steps of: adding a special processor in a node for processing flooding of a message, when the node receives a link state message containing link fault information, judging whether the link fault information exists in a storage area of the special processor or not, if so, not flooding the link state message, and if not, flooding the link state message to an adjacent node, and updating the link state database according to the link fault information and carrying out routing calculation and modification under the condition that data matched with the link fault information does not exist in the link state database. According to the invention, the special processor is used for performing flooding processing on the link state message, so that the time for performing single-time flooding on the node is shortened, the flooding time consumption of the whole network is reduced, the routing convergence speed is improved, and the influence of the routing convergence process on services is reduced.

The embodiment of the invention provides a dual-routing networking topology and an application method thereof. The topology comprises a client, a client router, a first GPON line, a second GPON line,a first convergence router and a second convergence router, a LAN port of the client router is connected with the client, a first WAN port of the client router is connected with the first convergencerouter through the first GPON line, a second WAN port of the client router is connected with the second convergence router through the second GPON line, the first convergence router and the second convergence router are connected with the Internet, and the client router, the first convergence router and the second convergence router are respectively configured with BFD sessions, so that fault switching is realized through the BFD sessions. According to the system and the method provided by the embodiment of the invention, whole-course physical double routing is realized, single-point hidden dangers do not exist, meanwhile, the configuration operation of fault switching is effectively simplified, and quick construction of double-routing networking is facilitated.

The disclosed embodiments provide a system for managing a broken link in a network with a Clos topology. During operation, the system detects, at a first node in the network, a broken link between the first node and a second node in the network. Next, the system identifies one or more upstream nodes in the network that can make routing decisions to avoid the broken link. The system then transmits a first indication of the broken link to the upstream node(s) without propagating the first indication to remaining nodes in the network. Finally, the system updates, based on the first indication, routing information at the upstream node(s) to avoid the broken link.

The invention discloses a dynamic self-adaptive compressed sensing data collection method for a network topology structure. The method comprises the following steps: (1) nodes are deployed and initialized; (2) double random walk is realized through the combination of same-layer random walk and cross-layer random walk; (3) based on dual random walk, a compressed sensing measurement process and a measurement result synchronous uploading process are completed; and (4) the SINK node performs compressed sensing data recovery. The method gets rid of dependence on global position coordinates; directed random walk and routing node spatial distribution non-uniformity compensation are realized at the same time through dual random walk; the spatial distribution uniformity of random walk path nodes isgood, the bottleneck problem caused by the convergence effect can be avoided, and meanwhile, the data recovery precision is higher; unification of compressed sensing measurement and measurement result transmission is realized, dependence on a static transmission path is eliminated, and network topology dynamic adaptive performance is good.

The invention provides a routing address storage method and device, and the method comprises the steps: a plurality of routing addresses and the common prefixes of the plurality of routing addresses are stored through a multi-stage static random access memory (SRAM), wherein the last stage of SRAM in the multi-stage SRAM is used for storing the plurality of routing addresses, other stages of SRAMs except the last stage of SRAM in the multi-stage SRAMs are used for storing common prefixes of the plurality of routing addresses; a common prefix of a first-stage common prefix in the multi-stage SRAM is stored in a ternary content addressable memory (TCAM), so that the problem of overlarge resource consumption of TCAM resources by an ALPM routing tree algorithm in related technologies can be solved, a routing address is stored through a multi-stage SRAM structure, namely, the common prefix of the routing address is stored by using one or more stages of SRAMs to form cascade connection; the prefix of the topmost stage is stored in the TCAM, and due to multi-stage prefix convergence, resource consumption of the TCAM is low.

The invention provides a network routing convergence processing and message processing method, device and equipment, relates to the technical field of communication, and is used for solving the problem of high processor resource consumption caused by high updating frequency of a VPN FTN. According to the method, a VPN routing table entry adopts a two-stage separation structure, namely, a structureof separating next-hop node information in the VPN from next-hop node information in a public network is adopted, so when a public network forwarding path is changed and the VPN routing table entry is issued, due to the fact that a public network updating table entry block comprising the next-hop node information in the public network is issued, only the corresponding public network updating table entry block needs to be searched according to the next hop node address in the first-level table entry block corresponding to the VPN routing table entry, and the block identifier of the second-level table entry block in the first-level table entry block is updated without updating the next hop node information of the public network for each VPN routing table entry, so that the resource consumption of the processor is reduced.

The invention relates to an RIP protocol optimization method suitable for on-satellite IP routing exchange, and belongs to the technical field of satellite communication. According to the invention, the satellite network RIP routing dynamic rapid learning based on the on-satellite IP routing exchange is realized through the design of interface network configuration information automatic sensing, on-satellite RIP message in-beam forwarding, RIP message format and routing algorithm optimization and the like. The method can be applied to satellite network scenes with on-satellite IP routing exchange and inter-satellite interconnection, and the problems that when a standard RIP routing protocol is applied to a satellite network, cross-satellite and cross-beam routing convergence is slow, and dynamic networking RIP routing learning in a single beam is difficult are solved.

The invention discloses a network security routing method and system, and the method comprises the steps: carrying out the security value initialization processing of each node in each node in a functional domain, and determining the security value of each intra-domain interface corresponding to each node; according to a conventional routing rule, traversing possible paths to the maximum extent in a preset period; carrying out convergence processing on paths participating in security routing on each node according to a preset in-band signaling rule based on the security value to obtain a path security value; and based on the path security value, performing secure routing forwarding on the screened key service traffic according to a routing forwarding strategy. According to the scheme, the path participating in the security routing can be converged to the optimal security degree according to the in-band signaling rule, so that global security routing forwarding optimization is carried out on the screened key service flow, and the scheme can be integrated into most existing networks of users, and is wide in applicability.

The invention discloses a route processing method, which is used for enabling network equipment in a network system to carry out route convergence in parallel and reducing the route convergence time. The first network device obtains first information, the first information indicates a fault of a first link, and the first link is a link between any two network devices in a network system to which the first network device belongs; the first network device sends first information to a second network device through a BGP, for example, sends a link fault message to the second network device through the BGP, wherein the link fault message carries the first information, and the second network device is a neighbor device of the first network device; the first network device determines an unreachable route in a routing table of the first network device according to the first information and the topology of the network system, wherein a path corresponding to the unreachable route comprises a first link; and the first network device marks the inaccessible route as invalid.

The invention discloses a warehouse sorting path optimization method based on a routing information protocol, a storage medium and equipment. On the basis of a group intelligent topological network formed by connecting all intelligent nodes in a warehouse, the nodes communicate with one another and can only communicate with directly connected neighbor nodes, and the routing information table of each node is continuously updated until convergence is achieved and each node obtains the shortest path reaching all the nodes. According to the method, each node cannot obtain global information, all tasks are sent out by the nodes, and all the nodes are cooperatively completed at the same level. The optimal path can be quickly constructed, so the warehouse can obtain the optimal path more quickly,more accurately and more stably.

The invention discloses a space-based networking routing method and equipment based on centralized and distributed combination. The method comprises the following steps that: a ground controller generates whole network topology prediction information in a period according to network information of a space-based network; the topology prediction information is analyzed by each space-based network node, and a link state notification containing full-period topology prediction information is generated; each space-based network node divides time periods according to the topology prediction information, so that the network topology in each time period is kept unchanged; each space-based network node selects a link state announcement of a time period to form a prediction topology, and a routing table is calculated; and each space-based network node detects the own link state in real time, compares the link state with the predicted topology, and triggers routing update under the condition that information is different. According to the space-based networking routing method disclosed by the invention, the prediction topology in each time period is kept unchanged, the network topology information can be quickly acquired at the starting moment of each time period, the routing convergence is effectively accelerated, the abnormal condition of the network can be sensed in real time, and the routing table item can be updated in time.

The invention provides a low-delay network content transmission method, device and equipment and a medium, and the method comprises the steps: constructing a content popularity model through employing the Zipf's law, and constructing a network topology model comprising a micro base station, a macro base station and cloud; constructing a network content total time delay model, wherein the network content total time delay model comprises a model for calculating network content transmission time delay on a link and a model for calculating average stay time delay in the micro base station, the macro base station and the cloud; performing a routing decision according to a routing strategy, regarding requests for the same content as the same request, solving the average delay in the network content total delay model based on a caching strategy, and optimizing caching and the routing decision based on a solving result. The convergence effect of the same content is considered to reduce the utilization rate of the server, so the network content transmission delay is optimized, the network content distribution efficiency is improved, and the network transmission pressure is relieved.

The invention provides a routing dynamic configuration method and device, electronic equipment and a storage medium. The method can comprise the steps of obtaining a current operation state of first network equipment; determining a target routing strategy corresponding to the current running state according to a pre-established corresponding relationship between the routing strategy and each running state of the first network equipment; and updating the route on the routing equipment according to the target routing strategy. According to the technical scheme of the invention, the method can achieve the automatic updating of the route, avoids the frequent convergence of the route while reducing the workload of manual configuration, and improves the efficiency and flexibility of a network system.