WO2021212823A1 - 一种重新路由方法及装置、电子设备、计算机可读介质 - Google Patents

一种重新路由方法及装置、电子设备、计算机可读介质 Download PDF

Info

Publication number
WO2021212823A1
WO2021212823A1 PCT/CN2020/130650 CN2020130650W WO2021212823A1 WO 2021212823 A1 WO2021212823 A1 WO 2021212823A1 CN 2020130650 W CN2020130650 W CN 2020130650W WO 2021212823 A1 WO2021212823 A1 WO 2021212823A1
Authority
WO
WIPO (PCT)
Prior art keywords
network
information
service
affected
fragmented
Prior art date
Application number
PCT/CN2020/130650
Other languages
English (en)
French (fr)
Inventor
张天祥
陆钱春
齐进
李锋
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Priority to US17/920,500 priority Critical patent/US20230155920A1/en
Publication of WO2021212823A1 publication Critical patent/WO2021212823A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/28—Routing or path finding of packets in data switching networks using route fault recovery
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/16—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/02—Topology update or discovery
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/02—Topology update or discovery
    • H04L45/08—Learning-based routing, e.g. using neural networks or artificial intelligence
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/12—Shortest path evaluation
    • H04L45/123—Evaluation of link metrics
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/12—Shortest path evaluation
    • H04L45/125—Shortest path evaluation based on throughput or bandwidth
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/22—Alternate routing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/12—Discovery or management of network topologies
    • H04L41/122—Discovery or management of network topologies of virtualised topologies, e.g. software-defined networks [SDN] or network function virtualisation [NFV]
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/50—Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate

Definitions

  • the embodiments of the present disclosure relate to the field of Internet technology, and in particular to a rerouting method and device, electronic equipment, and computer-readable media.
  • Rerouting is a service recovery method.
  • the controller or the first node
  • the controller needs to re-plan the route, that is, calculate an optimal path to resume the service.
  • a new LSP is established through signaling, and the new LSP is used to deliver services.
  • the controller can collect global topology information.
  • the network is turbulent, such as a link or node failure, in order not to affect the user experience, it is necessary to quickly restore the affected services. Because the affected services have bandwidth constraints and are limited by network bandwidth resources, the controller takes a long time to reroute batch services, has a low success rate, and reduces user experience.
  • the embodiments of the present disclosure provide a rerouting method and device, electronic equipment, and computer readable medium, which are used to reduce the rerouting time, increase the rerouting success rate, and improve the user experience.
  • embodiments of the present disclosure provide a rerouting method, which includes: updating network fragmentation information based on the service information of the affected service; wherein the service information includes service bandwidth resources, source nodes, and sink nodes,
  • the network fragmentation information is information of the fragmented network obtained after dividing the original network topology structure; the fragments of the affected service are obtained based on the business information of the affected service and the updated network fragmentation information Network index; wherein the fragmented network index is the corresponding relationship between the affected service and the fragmented network; a new route of the affected service is determined based on the affected service and the fragmented network index.
  • embodiments of the present disclosure provide a rerouting device, which includes: an update module configured to update network fragmentation information based on the service information of the affected service; wherein, the service information includes service bandwidth resources, so The network fragmentation information is information of the fragmented network obtained after dividing the original network topology; the fragmented network index module is set to obtain information based on the service information of the affected service and the updated network fragmentation information The fragmented network index of the affected service; wherein the fragmented network index is the corresponding relationship between the affected service and the fragmented network to which the service belongs; the route generation module is set to be based on the affected service and The fragmented network index obtains the new route of the affected service.
  • embodiments of the present disclosure provide an electronic device, which includes: one or more processors; a memory, on which one or more programs are stored, when the one or more programs are One or more processors execute any of the above rerouting methods; one or more I/O interfaces are connected between the processor and the memory, and are configured to implement the processor and the memory Information exchange of memory.
  • embodiments of the present disclosure provide a computer-readable medium on which a computer program is stored, and when the program is executed by a processor, any one of the rerouting methods described above is implemented.
  • FIG. 1 is a flowchart of a rerouting method provided by an embodiment of the disclosure
  • FIG. 2 is a topological diagram of an original network and a topological diagram of a divided fragmented network in an embodiment of the disclosure
  • FIG. 3 is a topological diagram of the original network and the topological diagram of the fragmented network after the split occurs in the embodiment of the disclosure;
  • FIG. 4 is a flowchart of updating network fragmentation information based on service information of affected services in the rerouting method provided by an embodiment of the present disclosure
  • FIG. 5 is a flowchart of obtaining an AI model in the rerouting method provided by an embodiment of the disclosure
  • FIG. 6 is a flowchart of obtaining a new route of the affected service based on the affected service and the fragmented network index in the rerouting method provided by an embodiment of the present disclosure
  • FIG. 7 is a flowchart of a rerouting method provided by an embodiment of the disclosure.
  • FIG. 8 is a flowchart of obtaining a new route of the affected service based on the affected service and the fragmented network index in the rerouting method provided by the embodiment of the present disclosure
  • FIG. 9 is a functional block diagram of a rerouting device provided by an embodiment of the disclosure.
  • FIG. 10 is a block diagram of an electronic device provided by an embodiment of the disclosure.
  • the controller can obtain global topology information through the forwarding device.
  • a software-defined network is used to collect global topology information to realize rapid service recovery in a fiber-cut scenario.
  • FIG. 1 is a flowchart of a rerouting method provided by an embodiment of the disclosure. Referring to Figure 1, the rerouting method includes:
  • Step 101 Update network fragmentation information based on the service information of the affected service.
  • the business information of the affected service can be determined according to the point of failure.
  • the service information of the affected service includes at least bandwidth resources, and may also include information such as source node, sink node, service quality, current deployment path, and constraint conditions.
  • the network fragmentation information is the fragmented network information obtained after dividing the original network topology structure.
  • the original network topology refers to the topology information reported to the controller by the forwarding device protocol when the controller starts or restarts for the first time.
  • the topology information includes node, link bandwidth, link delay, and link Road weight and other attributes. It is not difficult to understand that topology information is the topology information of the entire software-defined network, including both the network information being used and the unused network information.
  • the original network is divided according to the bandwidth resources of the original network.
  • the original network is divided by multiple methods such as bandwidth equalization or bandwidth randomization to obtain multiple fragmented networks.
  • FIG. 2 is an original network topology diagram and a fragmented network topology diagram after division in an embodiment of the disclosure.
  • the original network topology T0 includes four nodes A, B, C, D and four links AB, BD, DC, and CA.
  • the remaining bandwidth resource on the AB link is 30.
  • the weight is 1; the remaining bandwidth resource on the BD link is 26, and the weight of the link is 3; the remaining bandwidth resource on the DC link is 30, and the weight of the link is 1; the remaining bandwidth resource on the CA link If the value is 30, the weight of the link is 1.
  • the original network topology T0 is divided into fragmented networks T1, T2, T3, and the topological structure of the fragmented networks T1, T2, and T3 is consistent with the original network topology T0, that is, the fragmented networks T1, T2, and T3 each include four Nodes A, B, C, D and four links AB, BD, DC, CA.
  • the remaining bandwidth resource on the AB link is 10, and the weight of the link is 1; the remaining bandwidth resource on the BD link is 5, and the weight of the link is 3; DC link The remaining bandwidth resource on the upper link is 10, and the weight of the link is 1. The remaining bandwidth resource on the CA link is 10, and the weight of the link is 1.
  • the remaining bandwidth resource on the AB link is 10, and the weight of the link is 1; the remaining bandwidth resource on the BD link is 8, and the weight of the link is 3; the remaining bandwidth on the DC link
  • the bandwidth resource is 10, and the weight of the link is 1.
  • the remaining bandwidth resource on the CA link is 10, and the weight of the link is 1.
  • the remaining bandwidth resource on the AB link is 10, and the weight of the link is 1; the remaining bandwidth resource on the BD link is 13, and the weight of the link is 3; the remaining bandwidth on the DC link
  • the bandwidth resource is 10, and the weight of the link is 1.
  • the remaining bandwidth resource on the CA link is 10, and the weight of the link is 1.
  • the original network topology T0 is divided into segmented networks T1, T2, T3, and the topology of the segmented networks T1, T2, T3 is inconsistent with the original network topology T0. For example, if the bandwidth resource of link C-A in the fragmented network T1 is 0, then the topology of the fragmented network T1 includes four nodes A, B, C, and D and three links A-B, B-D, and D-C. The topological structure of the fragmented networks T2 and T3 is consistent with the original network topological structure T0.
  • the sum of the bandwidth resources corresponding to each segmented network is equal to the bandwidth resource of the link in the original network topology structure T0.
  • the bandwidth resources of each fragmented network are isolated from each other.
  • the affected service is traced back, the path of the affected service is determined, and the bandwidth resource of the affected service is released.
  • the network fragmentation information is updated, and the links and nodes corresponding to the affected services are about to be deleted, and link resources are restored.
  • FIG. 3 is an original network topology diagram and a divided network fragment topology diagram when a fiber cut occurs in an embodiment of the disclosure.
  • the links A->C in the three fragmented networks T1, T2, and T3 all have fiber cuts, and the topology changes accordingly.
  • the bandwidth resources of link A->C are allocated to link B->A, link C->D, and link D->B. For example, when the bandwidth resource of link A->C is 30, link B->A, link C->D, and link D->B will obtain a total of 30 bandwidth resources.
  • the remaining bandwidth of link B->A and link C->D is increased by 10 and 7 respectively in fragmented network T1 and fragmented network T2, that is, link B->A and chain in fragmented network T1
  • the bandwidth resources before the fiber break are all 10, and the bandwidth resources after the fiber break are all 20
  • the link B->A and link C->D in the fragmented network T2 are in The bandwidth resources before the fiber break are all 10, and the bandwidth resources after the fiber break are all 17.
  • the remaining bandwidth resource on the AB link is 20, and the weight of the link is 1; the remaining bandwidth resource on the BD link is 5, and the weight of the link is 3; The remaining bandwidth resource on the road is 20, and the weight of the link is 1.
  • the remaining bandwidth resource on the AB link is 17, and the weight of the link is 1; the remaining bandwidth resource on the BD link is 8, and the weight of the link is 3; the remaining bandwidth on the DC link The bandwidth resource is 17, and the weight of the link is 1.
  • the remaining bandwidth resource on the AB link is 10, and the weight of the link is 1; the remaining bandwidth resource on the BD link is 13, and the weight of the link is 3; the remaining bandwidth on the DC link The bandwidth resource is 10, and the weight of the link is 1.
  • the network fragment information is updated based on the service information of the affected service, for example, the link A->C is deleted, and the bandwidth resources of the link B->A and the link C->D are changed.
  • the original network topology T0 is divided into three fragmented networks T1, T2, and T3 in this embodiment, this does not mean that the number of fragmented networks is limited. In fact, the number of fragmented networks can be arbitrarily set according to the complexity of the original network topology T0 or other reasons, and this embodiment does not limit the number of fragmented networks.
  • Step 102 Obtain a segmented network index of the affected service based on the service information of the affected service and the updated network segmentation information.
  • step 102 after determining the affected service, re-plan the route of the affected service.
  • this embodiment obtains the fragmented network index of the affected services based on the service information of the affected services and the updated network fragmentation information, and then on this basis , Re-plan the route.
  • the fragmented network index is the corresponding relationship between the affected service and the fragmented network to which the service belongs again.
  • the fragmented network index refers to the index of the affected service obtained from the updated network fragment information, that is, the index of the fragmented network to which the affected service is redistributed.
  • Service 1 is from node B to node D, the bandwidth resource occupied is 10, currently belongs to the fragmented network T1, and the path is B-> D;
  • Service 2 is from node B to node D, and the occupied bandwidth resource is 10, currently belongs to the fragmented network T1, and the path is B->A->C->D;
  • service 3 is from node B to node D, occupied The bandwidth resource is 7, currently belongs to the fragmented network T2, and the path is B->D;
  • the service 4 is from node B to node D, and the occupied bandwidth resource is 7, and the current belongs to the fragmented network T2, and the path is B->A- >C->D;
  • Service 5 is from node B to node D, and the occupied bandwidth resource is 1, currently belongs to the fragmented network T3, and the path is B->D;
  • service 6 is from node B to node D, and the occupied bandwidth resource It is 1, currently belongs to the fragmented network T3,
  • Table 1 is the business information of the six services in the embodiments of this disclosure.
  • FIG. 3 is a diagram of the network fragmentation topology after updating the network fragmentation information. Based on the updated network fragmentation information, the fragmented network index of the affected service is obtained.
  • service 2 requires a bandwidth resource of 10
  • service 4 requires a bandwidth resource of 7
  • the bandwidth resource of B->D in fragmented network T1 is 5, which cannot meet the requirements of service 2 and service 4. Therefore, fragmented network T1 cannot Allocate new business.
  • the bandwidth resource of B->D in the segmented network T2 is 8 and the service resource of B->D in the segmented network T3 is 13.
  • the bandwidth resource of the segmented network T2 can only meet the service 4, and the bandwidth resource of the segmented network T3 It can meet service 4 and service 2, but only service 2 is allocated to the fragmented network T3, and service 4 is allocated to the fragmented network T2 to restore both business 2 and business 4 at the same time. Therefore, the service 2 is allocated to the fragmented network T3.
  • Service 4 is allocated to the fragmented network T2.
  • Step 103 Obtain a new route of the affected service based on the affected service and the fragmented network index.
  • step 103 the newly allocated services are calculated in the segmented network T1, segmented network T2, and segmented network T3 at the same time, and the new route of service 2 is obtained as B->D, and the new route of service 4 is B ->D; So far, the controller has completed the rerouting of service 2 and service 4. It is not difficult to understand that when calculating a new route, the fragmented network T1 does not actually participate, because the fragmented network T1 is not allocated to services to be restored.
  • FIG. 4 is a flowchart of updating network fragmentation information based on the service information of the affected service in the rerouting method provided by an embodiment of the present disclosure. As shown in Figure 4, the network fragmentation information is updated based on the affected service information, including:
  • Step 401 Trace back the affected business based on the business information of the affected business.
  • step 401 after obtaining the service information of the affected service, the source node, sink node, and bandwidth resource of the affected service are obtained, and the affected service is traced back to determine the service information of the affected service.
  • Step 402 Release the bandwidth resources of the affected services.
  • step 402 the bandwidth resource of service 2 is released.
  • Step 403 Update the network fragment information based on the service information of the affected service and the released bandwidth resource of the affected service.
  • the network fragmentation information is updated based on the service information of the affected service and the released bandwidth resources. For example, update the source node B and source node D of service 2, and update the bandwidth resources of link A-B, link B-D, link D-C, and link A-C.
  • the bandwidth resource released by link B-D is allocated to link A-B, link D-C, and link A-C as 10.
  • the fragmented network index of the affected business is obtained through the AI model, and then based on the affected business information and the updated network fragmentation information, the AI model is used to obtain the affected business's index. Fragmented network index.
  • FIG. 5 is a flowchart of obtaining an AI model in the rerouting method provided by an embodiment of the disclosure. As shown in Figure 5, the training steps of the AI model include:
  • Step 501 Obtain original network topology information.
  • the original network topology information includes nodes, link bandwidth resources, link delays, and link weights.
  • the original network topology information comes from a forwarding device connected to the controller.
  • the forwarding device reports to the controller through the forwarding protocol.
  • Step 502 Divide the original network topology information according to network resources to obtain network fragmentation information.
  • the network resources can be bandwidth resources or other resources. Divide the original network topology information according to network resources to obtain network fragmentation information.
  • the original network topology information is divided according to a bandwidth resource equalization method or a bandwidth resource random method to obtain network fragmentation information.
  • the network fragmentation information includes information of multiple fragmented networks.
  • the information of the fragmented network includes topological structure, link attributes, and so on.
  • the topology of the fragmented network is consistent with the topology of the original network.
  • the sum of the corresponding bandwidth resources of each fragmented network T1, T2, T3 is equal to the bandwidth resource of the chain in the original network topology T0.
  • the sum of the bandwidth resources corresponding to the link BD in the fragmented networks T1, T2, and T3 are 5, 8, and 13, respectively.
  • the sum of these bandwidth resources is 26, which is equal to the link in the original network topology T0.
  • Bandwidth resources of BD This is the characteristic of slicing.
  • Step 503 Use the service information of the affected service, the updated network fragment information, and the home fragment network of the affected service as training samples, and use the deep learning algorithm to train the model to obtain the AI model.
  • step 503 greedy and heuristic methods are used to construct training samples, and a machine learning model is established.
  • the business information and network fragmentation information of the affected business are used as input, and the home fragmented network of the affected business is output. After the AI model is trained, save it for subsequent recall.
  • the deep learning training model is used to simulate the fiber disconnection phenomenon of the physical network T0, and the service information of the affected service after the fiber disconnection and the network fragment information updated after the fiber disconnection are used as input, and the attribution of the affected service is The sharded network is the output.
  • the AI model can be trained offline. Using the AI model, the fragmented network index of all affected services can be obtained.
  • service 4 is an affected service
  • the occupied bandwidth resource is 7. Therefore, in the updated network fragmentation information, the remaining bandwidth in the fragmented network T1 and the fragmented network T2 increases by 7, and the service 4 is re-allocated to the fragmented network.
  • the slice network T2, that is, the affected service 4 belongs to the slice network T2 again.
  • service 1 can be re-attributed to the segmented network T1
  • service 2 can be re-attributed to the segmented network T3.
  • FIG. 6 is a flowchart of obtaining a new route of the affected service based on the affected service and the fragmented network index in the rerouting method provided by the embodiment of the present disclosure.
  • the new routes of the affected services are obtained based on the affected services and the corresponding fragmented network index, including:
  • Step 601 Determine the fragmented network of the affected service according to the fragmented network index of the affected service.
  • the AI model is used to obtain the fragmented network index of the affected business, and the fragmented network of the affected business is determined according to the fragmented network index.
  • Step 602 Calculate the path of the affected service in the fragmented network corresponding to the affected service, and obtain a new route of the affected service.
  • step 602 the path of the affected service is calculated in a parallel manner in each fragmented network to obtain a new route of the affected service.
  • the new route of the affected business is returned to the forwarding device, and the forwarding device issues the new route in a distributed manner to restore the affected business.
  • the rerouting method obtaineds the service information of the affected service, updates the network fragment information based on the service information of the affected service, and obtains the affected service based on the service information of the affected service and the updated network fragment information Based on the affected business and the fragmented network index to determine the new route of the affected business, this method updates the network fragment information based on the business information of the affected business, and then uses the business information of the affected business and the updated The network fragmentation information obtains the fragmented network index of the affected services. Therefore, new routes can be determined for batches of affected services in parallel, which reduces the rerouting time, improves the rerouting efficiency, and improves the rerouting success. Rate and improve user experience.
  • the rerouting method provided by the embodiment of the present disclosure takes into account the service type and network fragmentation resources, and updates the network fragmentation information based on the service information of the affected service, so that the limited network resources can ensure the transmission success rate of the service as much as possible. Transmission quality.
  • the rerouting method provided by the embodiment of the present disclosure divides the original network according to network resources, and then obtains the fragmented network index of the affected service based on the service information of the affected service and the updated network fragmentation information, This makes it possible to recalculate the routes of the affected services in a parallel manner, which greatly reduces the time-consuming rerouting.
  • the time-consuming rerouting is proportional to the number of fragmented networks.
  • FIG. 7 is a flowchart of another rerouting method provided by an embodiment of the disclosure.
  • the failure of a node causes network turbulence as an example. Services related to the node are affected, and these affected services need to be rerouted.
  • the rerouting method includes:
  • Step 701 Determine the business information of the affected business.
  • the topological structure of the original network and the fragmented network is shown in Figure 2.
  • the topological structure of the original network and the fragmented network includes four nodes A, B, C, D and four links AB, BD, DC , CA, where the remaining bandwidth resource on the AB link is 30, and the weight of the link is 1; the remaining bandwidth resource on the BD link is 26, and the weight of the link is 3; the remaining bandwidth resources on the DC link If it is 30, the weight of the link is 1; the remaining bandwidth resource on the CA link is 30, and the weight of the link is 1.
  • the link of service 7 is A->C->D.
  • the service 7 is traced back to obtain the service information of service 7, that is, the source node of service 7 is A and the sink node is D.
  • the occupied bandwidth resource is 15.
  • Step 702 Release the bandwidth resources of the affected services.
  • the bandwidth resource of the affected service determined in step 701 is released. If the bandwidth resources of link A->C and link C->D are released, the released bandwidth resource is 15.
  • Step 703 Update network fragmentation information based on the released bandwidth resources of the affected services.
  • the released bandwidth resources of link A->C and link C->D are allocated to link AB and link BD.
  • the bandwidth resource obtained by link AB is 15, and the bandwidth resource obtained by link BD The bandwidth resource is 15.
  • the bandwidth resource obtained by the link AB is 8, and the bandwidth resource obtained by the link BD is 10, and in the fragmented network T2, the bandwidth resource obtained by the link AB is 7, and the bandwidth resource obtained by the link BD The bandwidth resource is 5.
  • the link A-B and the link B-D are not allocated bandwidth resources.
  • the updated network fragmentation information is: the topology of the fragmentation network includes three nodes A, B, and D and two links AB and BD, where the AB link The remaining bandwidth resource on the upper limit is 45, and the weight of the link is 1. The remaining bandwidth resource on the BD link is 41, and the weight of the link is 3.
  • the remaining bandwidth resource on the A-B link is 18, and the weight of the link is 1; the remaining bandwidth resource on the B-D link is 15, and the weight of the link is 3.
  • the remaining bandwidth resource on the A-B link is 17, and the weight of the link is 1, and the remaining bandwidth resource on the B-D link is 13, and the weight of the link is 3.
  • the remaining bandwidth resource on the A-B link is 10, and the weight of the link is 1; the remaining bandwidth resource on the B-D link is 13, and the weight of the link is 3.
  • Step 704 Obtain a segmented network index of the affected service based on the service information of the affected service and the updated network segmentation information.
  • the AI model is used to obtain the fragmented network index of the affected service 7 according to the service information of the affected service 7 and the updated network fragmentation information.
  • the fragmented network index of service 7 is fragmented network T1, that is, the affected service 7 belongs to fragmented network T1 again. Because in the fragmented network T2, the remaining bandwidth resource of the link A-B is 17, but the remaining bandwidth resource of the link B-D is only 13, and the bandwidth resource required by the service 7 is 15, therefore, the fragmented network T2 cannot satisfy the service 7. In the fragmented network T3, the remaining bandwidth resource of the link A-B is 10, and the remaining bandwidth resource of the link B-D is only 13, which cannot satisfy the bandwidth resource required by the service 7. In the fragmented network T1, the remaining bandwidth resource of the link A-B is 18, but the remaining bandwidth resource of the link B-D is 15, which can meet the bandwidth resource required by the service 7.
  • a link and node failure is used as an example to introduce a rerouting method, but the present disclosure is not limited to this.
  • the rerouting method provided in the embodiments of the present disclosure is also used to reroute the affected services.
  • Step 705 Calculate the path of the affected service in the fragmented network corresponding to the affected service, and obtain a new route of the affected service.
  • the new route of the affected business is returned to the forwarding device, and the forwarding device issues the new route in a distributed manner to restore the affected business.
  • the new route to obtain the affected service in the fragmented network T1 is B->D.
  • FIG. 9 is a functional block diagram of a rerouting device provided by an embodiment of the disclosure.
  • the rerouting device includes:
  • the update module 901 is configured to update the network fragment information based on the service information of the affected service.
  • the network fragmentation information is the fragmented network information obtained after dividing the original network topology structure
  • the service information includes information such as service bandwidth resources, source nodes, sink nodes, service quality, current deployment path, and constraint conditions.
  • Service information can be reported to the controller through the forwarding device, and the controller manages the service path and network resources.
  • the update module uses the topology loading unit and the fragment resource management unit, wherein the topology loading unit is configured to obtain node information, link information, etc. of the original network, and report to the controller through a protocol.
  • the fragmented resource management unit is configured to perform fragmented management of the original network topology, divide the original network topology into multiple fragmented networks, and the structure of each fragmented network is consistent with the original network topology.
  • the fragmented resource management unit is also configured to maintain service information corresponding to each fragmented network.
  • the update module 901 is configured to update the network fragment information. For example, trace the affected services based on the service information of the affected services, determine the deleted nodes and links, as well as the recovery of link resources, and the released bandwidth resources, and then update the network fragmentation information.
  • the segmented network index module 902 is configured to obtain the segmented network index of the affected service based on the service information of the affected service and the updated network segmentation information.
  • the fragmented network index is the corresponding relationship between the affected service and the fragmented network to which the service belongs again.
  • the segmented network index module 902 may be an AI model, and input the service information of the affected service and the updated network segmentation information into the AI model to obtain the segmented network index of the affected service.
  • the route generation module 903 is configured to obtain a new route of the affected service based on the affected service and the fragmented network index.
  • the route generation module 903 calculates the path of the affected service in each segmented network in a parallel manner to obtain a new route of the affected service.
  • the new route of the affected business is returned to the forwarding device, and the forwarding device issues the new route in a distributed manner to restore the affected business.
  • the rerouting device obtained in this embodiment obtains the business information of the affected service through the acquisition module, uses the update module to update the network fragmentation information based on the business information of the affected business, and the fragmented network index module is based on the business information and updates of the affected business. After the network fragmentation information obtains the fragmented network index of the affected service, the route generation module determines the new route of the affected service based on the affected service and the fragmented network index. The device reduces the rerouting time and improves the rerouting Efficiency, moreover, improve the success rate of rerouting and improve user experience.
  • an electronic device which includes:
  • the memory 1002 has one or more programs stored thereon, and when the one or more programs are executed by one or more processors, the one or more processors implement any one of the above-mentioned rerouting methods;
  • One or more I/O interfaces 1003 are connected between the processor and the memory, and are configured to implement information interaction between the processor and the memory.
  • the processor 1001 is a device with data processing capabilities, including but not limited to a central processing unit (CPU), etc.
  • the memory 1002 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically Such as SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory (FLASH); I/O interface (read and write interface) 1003 is connected between processor 1001 and memory 1002 , Can realize the information interaction between the processor 1001 and the memory 1002, which includes, but is not limited to, a data bus (Bus) and the like.
  • RAM random access memory
  • ROM read-only memory
  • EEPROM electrically erasable programmable read-only memory
  • FLASH flash memory
  • I/O interface (read and write interface) 1003 is connected between processor 1001 and memory 1002 , Can realize the information interaction between the processor 1001 and the memory 1002, which includes, but is not limited to, a data bus (Bus)
  • the processor 1001, the memory 1002, and the I/O interface 1003 are connected to each other through a bus, and further connected to other components of the computing device.
  • the embodiments of the present disclosure provide a computer-readable medium on which a computer program is stored, and when the program is executed by a processor, any one of the above-mentioned rerouting methods is implemented.
  • Such software may be distributed on a computer-readable medium, and the computer-readable medium may include a computer storage medium (or a non-transitory medium) and a communication medium (or a transitory medium).
  • the term computer storage medium includes volatile and non-volatile memory implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Sexual, removable and non-removable media.
  • Computer storage media include but are not limited to RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cassette, tape, magnetic disk storage or other magnetic storage device, or Any other medium used to store desired information and that can be accessed by a computer.
  • a communication medium usually contains computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and may include any information delivery medium. .

Abstract

本公开提供了一种重新路由方法，其包括：基于所述受影响业务的业务信息更新网络分片信息；其中，所述业务信息包括业务带宽资源，所述网络分片信息是对原始网络拓扑结构划分后获得的分片网络的信息；基于所述受影响业务的业务信息和更新后的所述网络分片信息获得所述受影响业务的分片网络索引；其中，所述分片网络索引是所述受影响业务与分片网络的对应关系；基于所述受影响业务和所述分片网络索引确定所述受影响业务的新路由。该重新路由方法获得重新路由的耗时短、效率高，而且成功率高。本公开还提供了一种重新路由装置、电子设备、计算机可读介质。

Description

一种重新路由方法及装置、电子设备、计算机可读介质 技术领域
本公开实施例涉及互联网技术领域，特别涉及一种重新路由方法及装置、电子设备、计算机可读介质。
背景技术
重路由是一种业务恢复方式，当业务路径上的标记转发路由(Label Switch Path，简称LSP)中断时，控制器(或首节点)需要重新规划路由，即计算一条恢复业务的最佳路径，通过信令建立起一条新的LSP，由新的LSP来传送业务。
在软件定义网络(Software Defined Network，简称SDN)场景下，控制器可以收集全局拓扑信息。当网络发生动荡时，如链路或者节点发生故障时，为了不影响用户体验，需要快速恢复受影响的业务。由于受影响业务具有带宽约束且受限于网络带宽资源，控制器对批量业务重新路由的耗时长，成功率低，降低了用户体验。
公开内容
本公开实施例提供一种重新路由方法及装置、电子设备、计算机可读介质，用以降低重新路由的时间、提高重新路由的成功率，提升用户体验。
第一方面，本公开实施例提供一种重新路由方法，其包括：基于所述受影响业务的业务信息更新网络分片信息；其中，所述业务信息包括业务带宽资源、源节点和宿节点，所述网络分片信息是对原始网络拓扑结构划分后获得的分片网络的信息；基于所述受影响业务的业务信息和更新后的所述网络分片信息获得所述受影响业务的分片网络索引；其中，所述分片网络索引是所述受影响业务与分片网络的对应关系；基于所述受影响业务和所述分片网络索引确定所述受影响业务的新路由。
第二方面，本公开实施例提供一种重新路由装置，其包括：更新模块， 设置为基于所述受影响业务的业务信息更新网络分片信息；其中，所述业务信息包括业务带宽资源，所述网络分片信息是对原始网络拓扑结构划分后获得的分片网络的信息；分片网络索引模块，设置为基于所述受影响业务的业务信息和更新后的所述网络分片信息获得所述受影响业务的分片网络索引；其中，所述分片网络索引是所述受影响业务与该业务重新归属的分片网络的对应关系；路由生成模块，设置为基于所述受影响业务和所述分片网络索引获得所述受影响业务的新路由。
第三方面，本公开实施例提供一种电子设备，其包括：一个或多个处理器；存储器，其上存储有一个或多个程序，当所述一个或多个程序被所述一个或多个处理器执行，使得所述一个或多个处理器上述任意一种重新路由方法；一个或多个I/O接口，连接在所述处理器与存储器之间，配置为实现所述处理器与存储器的信息交互。
第四方面，本公开实施例提供一种计算机可读介质，其上存储有计算机程序，所述程序被处理器执行时实现上述任意一种重新路由方法。
附图说明
附图用来提供对本公开实施例的进一步理解，并且构成说明书的一部分，与本公开的实施例一起用于解释本公开，并不构成对本公开的限制。通过参考附图对详细示例实施例进行描述，以上和其它特征和优点对本领域技术人员将变得更加显而易见，在附图中：
图1为本公开实施例提供的一种重新路由的方法的流程图；
图2为本公开实施例中原始网络拓扑图以及划分后的分片网络的拓扑图；
图3为本公开实施例中发生断纤情况时原始网络拓扑图以及划分后分片网络的拓扑图；
图4为本公开实施例提供的重新路由的方法中基于受影响业务的业务信息更新网络分片信息的流程图；
图5为本公开实施例提供的重新路由的方法中获得AI模型的流程图；
图6为本公开实施例提供的重新路由的方法中基于受影响业务和分片网络索引获得受影响业务的新路由的流程图；
图7为本公开实施例提供的一种重新路由的方法的流程图；
图8为本公开实施例提供的重新路由的方法中基于受影响业务和分片网络索引获得受影响业务的新路由的流程图；
图9为本公开实施例提供的一种重新路由装置的原理框图；
图10为本公开实施例提供的一种电子设备的组成框图。
具体实施方式
为使本领域的技术人员更好地理解本公开的技术方案，下面结合附图对本公开提供的重新路由的方法及装置、电子设备、计算机可读介质进行详细描述。
在下文中将参考附图更充分地描述示例实施例，但是所述示例实施例可以以不同形式来体现且不应当被解释为限于本文阐述的实施例。反之，提供这些实施例的目的在于使本公开透彻和完整，并将使本领域技术人员充分理解本公开的范围。
在不冲突的情况下，本公开各实施例及实施例中的各特征可相互组合。
如本文所使用的，术语“和/或”包括一个或多个相关列举条目的任何和所有组合。
本文所使用的术语仅用于描述特定实施例，且不意欲限制本公开。如本文所使用的，单数形式“一个”和“该”也意欲包括复数形式，除非上下文另外清楚指出。还将理解的是，当本说明书中使用术语“包括”和/或“由……制成”时，指定存在所述特征、整体、步骤、操作、元件和/或组件，但不排除存在或添加一个或多个其它特征、整体、步骤、操作、元件、组件和/或其群组。
除非另外限定，否则本文所用的所有术语(包括技术和科学术语)的含 义与本领域普通技术人员通常理解的含义相同。还将理解，诸如那些在常用字典中限定的那些术语应当被解释为具有与其在相关技术以及本公开的背景下的含义一致的含义，且将不解释为具有理想化或过度形式上的含义，除非本文明确如此限定。
在软件定义网络中，控制器能够通过转发设备获得全局拓扑信息。本公开实施例是利用软件定义网络收集全局拓扑信息的情况下实现断纤场景下的业务快速恢复。
第一方面，本公开实施例提供了一种重新路由的方法。图1为本公开实施例提供的一种重新路由的方法的流程图。参照图1，重新路由的方法包括：
步骤101，基于受影响业务的业务信息更新网络分片信息。
在步骤101中，当网络发生动荡时，如链路或节点发生故障时，与该链路和节点相关的业务将受到影响，因此，可以根据故障点确定受影响业务的业务信息。其中，受影响业务的业务信息至少包括带宽资源，还可以包括源节点、宿节点、服务质量、当前部署路径、约束条件等信息。
其中，网络分片信息是对原始网络拓扑结构划分后获得的分片网络的信息。
在本实施例中，原始网络拓扑结构是指控制器第一次启动或重启时，转发设备协议上报给控制器的拓扑信息，其中，拓扑信息包括节点、链路带宽、链路时延、链路权值等属性。不难理解，拓扑信息是整个软件定义网络的拓扑信息，既包括正在使用的网络信息，也包括未使用的网络信息。
其中，按照原始网络的带宽资源划分原始网络。在一些实施例中，采用带宽均分或带宽随机等多种方式划分原始网络，获得多个分片网络。
图2为本公开实施例中原始网络拓扑图以及划分后的分片网络拓扑图。如图2所示，原始网络拓扑结构T0包括四个节点A、B、C、D和四条链路A-B、B-D、D-C、C-A，其中，A-B链路上剩余带宽资源为30，该链路的权值为1；B-D链路上剩余带宽资源为26，该链路的权值为3；D-C 链路上剩余带宽资源为30，该链路的权值为1；C-A链路上剩余带宽资源为30，该链路的权值为1。
原始网络拓扑结构T0被划分为分片网络T1、T2、T3，而且分片网络T1、T2、T3的拓扑结构与原始网络拓扑结构T0一致，即分片网络T1、T2、T3均包括四个节点A、B、C、D和四条链路A-B、B-D、D-C、C-A。
其中，在分片网络T1中，A-B链路上剩余带宽资源为10，该链路的权值为1；B-D链路上剩余带宽资源为5，该链路的权值为3；D-C链路上剩余带宽资源为10，该链路的权值为1；C-A链路上剩余带宽资源为10，该链路的权值为1。
在分片网络T2中，A-B链路上剩余带宽资源为10，该链路的权值为1；B-D链路上剩余带宽资源为8，该链路的权值为3；D-C链路上剩余带宽资源为10，该链路的权值为1；C-A链路上剩余带宽资源为10，该链路的权值为1。
在分片网络T3中，A-B链路上剩余带宽资源为10，该链路的权值为1；B-D链路上剩余带宽资源为13，该链路的权值为3；D-C链路上剩余带宽资源为10，该链路的权值为1；C-A链路上剩余带宽资源为10，该链路的权值为1。
在一些实施例中，原始网络拓扑结构T0被划分为分片网络T1、T2、T3，而且分片网络T1、T2、T3的拓扑结构与原始网络拓扑结构T0不一致。例如，分片网络T1中链路C-A的带宽资源为0，那么，分片网络T1的拓扑结构包括四个节点A、B、C、D和三条链路A-B、B-D、D-C。分片网络T2、T3的拓扑结构与原始网络拓扑结构T0保持一致。
不难理解，对于其中某条链路而言，将各个分片网络对应的带宽资源之和相加后，等于原始网络拓扑结构T0中该链路的带宽资源。此外，每个分片网络的带宽资源彼此隔离。
当原始网络中的链路或节点发生故障时，如某条链路发生断网时，与该链路相关的业务将受到影响。在恢复受影响的业务之前，需要确定受影 响业务占用的资源，以便释放受影响业务占用的资源，为重新规划该受影响业务的路由提供依据。
在一些实施例中，在获得受影响业务的业务信息后，追溯该受影响业务，确定受影响业务的路径，并释放受影响业务的带宽资源。同时，更新网络分片信息，即将受影响业务对应的链路、节点删除以及链路资源的恢复等。
图3为本公开实施例中发生断纤情况时原始网络拓扑图以及划分后的网络分片拓扑图。如图3所示，假设链路A->C发生断纤情况，此时，三个分片网络T1、T2、T3中的链路A->C均发生断纤，拓扑结构随之发生改变。链路A->C的带宽资源被分配给链路B->A、链路C->D和链路D->B。例如，当链路A->C的带宽资源为30时，链路B->A、链路C->D和链路D->B将获得总额为30的带宽资源。链路B->A和链路C->D的剩余带宽在分片网络T1和分片网络T2中分别相应增加10和7，即，分片网络T1中的链路B->A和链路C->D，在断纤之前的带宽资源均为10，在断纤之后的带宽资源均变为20；分片网络T2中的链路B->A和链路C->D，在断纤之前的带宽资源均为10，在断纤之后的带宽资源均变为17。
具体地，在分片网络T1中，A-B链路上剩余带宽资源为20，该链路的权值为1；B-D链路上剩余带宽资源为5，该链路的权值为3；D-C链路上剩余带宽资源为20，该链路的权值为1。
在分片网络T2中，A-B链路上剩余带宽资源为17，该链路的权值为1；B-D链路上剩余带宽资源为8，该链路的权值为3；D-C链路上剩余带宽资源为17，该链路的权值为1。
在分片网络T3中，A-B链路上剩余带宽资源为10，该链路的权值为1；B-D链路上剩余带宽资源为13，该链路的权值为3；D-C链路上剩余带宽资源为10，该链路的权值为1。
在本实施例中，基于受影响业务的业务信息更新网络分片信息，例如，将链路A->C删除，变更链路B->A和链路C->D的带宽资源。
需要说明的是，虽然本实施例是将原始网络拓扑结构T0划分三个分片网络T1、T2、T3进行介绍，但这并不表示是对分片网络的数量进行限制。实际上，分片网络的数量可以根据原始网络拓扑结构T0的复杂程度，或其它原因任意设定，本实施例对分片网络的数量不作限定。
步骤102，基于受影响业务的业务信息和更新后的网络分片信息获得受影响业务的分片网络索引。
在步骤102中，在确定受影响业务后，重新规划该受影响业务的路由。为了使多个受影响的业务能够快速地获得新的路由，本实施例是基于受影响业务的业务信息和更新后的网络分片信息获得受影响业务的分片网络索引，然后在此基础上，重新规划路由。其中，分片网络索引是受影响业务与该业务重新归属的分片网络的对应关系。
在本实施例中，分片网络索引是指在更新后的网络分片信息中得到受影响业务的索引，即该受影响业务被重新分配在哪个分片网络的索引。
结合参阅图2、图3和表1，例如，当前存在六个业务，其中，业务1为节点B到节点D，占用的带宽资源为10，当前归属于分片网络T1，路径为B->D；业务2为节点B到节点D，占用的带宽资源为10，当前归属于分片网络T1，路径为B->A->C->D；业务3为节点B到节点D，占用的带宽资源为7，当前归属于分片网络T2，路径为B->D；业务4为节点B到节点D，占用的带宽资源为7,当前归属分片网络T2，路径为B->A->C->D；业务5为节点B到节点D，占用的带宽资源为1,当前归属于分片网络T3，路径为B->D；业务6为节点B到节点D，占用的带宽资源为1,当前归属于分片网络T3，路径为B->D，此时分片网络中剩余的带宽信息如图2所示。
表1 为本公开实施例中的六个业务的业务信息
业务编号 源节点 宿节点 带宽 原业务的归属 断纤后业务的归属
1 B D 10 分片网络T1 分片网络T1
2 B D 10 分片网络T1 分片网络T3
3 B D 7 分片网络T2 分片网络T2
4 B D 7 分片网络T2 分片网络T2
5 B D 1 分片网络T3 分片网络T3
6 B D 1 分片网络T3 分片网络T3
假设，链路A->C发生断纤情况，即原始网络的物理拓扑中发生断纤情况，三个分片网络中拓扑结构也随之发生改变。经过链路A->C的业务2和业务4为受影响的业务。因此，在更新网络分片信息时，将分片网络T1、T2和T3中的链路A->C删除，将链路B->A和链路C->D在分片网络T1和分片网络T2中的剩余带宽相应增加10和7。图3为更新网络分片信息后的网络分片拓扑图。基于更新后的网络分片信息，获得受影响业务的分片网络索引。例如，业务2需要带宽资源为10，业务4需要带宽资源为7，分片网络T1中B->D的带宽资源为5，不能满足业务2和业务4的需求，因此，分片网络T1不能分配新业务。分片网络T2中B->D的带宽资源为8和分片网络T3中B->D的业务资源为13，分片网络T2的带宽资源只能满足业务4，分片网络T3的带宽资源可以满足业务4和业务2，但只有业务2被分配至分片网络T3，业务4被分配至分片网络T2才能同时恢复业务2和业务4，因此，业务2被分配至分片网络T3，业务4被分配至分片网络T2。
步骤103，基于受影响业务和分片网络索引获得受影响业务的新路由。
在步骤103中，在分片网络T1、分片网络T2、分片网络T3内同时对新分配的业务进行路径计算，得到业务2的新路由为B->D，业务4的新路由为B->D；至此，控制器完成了业务2和业务4的重新路由。不难理解，在计算新路由时，分片网络T1其实并不参与，因为分片网络T1未分配到待恢复业务。
图4为本公开实施例提供的重新路由的方法中基于受影响业务的业务信息更新网络分片信息的流程图。如图4所示，基于受影响的业务信息更新网络分片信息，包括：
步骤401，基于受影响业务的业务信息追溯受影响业务。
在步骤401中，在获得受影响业务的业务信息后，即获得受影响业务的源节点、宿节点、带宽资源，追溯该受影响业务，以确定受影响业务的业务信息。
参阅表1，假设业务2为受影响业务，追溯业务2，获得业务2的业务信息包括源节点B、源节点D、带宽资源为10。
步骤402，释放受影响业务的带宽资源。
在步骤402中，将业务2的带宽资源释放。
步骤403，基于受影响业务的业务信息和释放的受影响业务的带宽资源更新网络分片信息。
在步骤403中，基于受影响业务的业务信息和释放的带宽资源更新网络分片信息。例如，更新业务2的源节点B、源节点D，以及更新链路A-B、链路B-D、链路D-C和链路A-C的带宽资源。将链路B-D释放的带宽资源为10分配给链路A-B、链路D-C和链路A-C。
在一些实施例中，在步骤103中，受影响业务的分片网络索引是通过AI模型获得的，然后基于受影响的业务信息和更新后的网络分片信息，利用AI模型获得受影响业务的分片网络索引。
图5为本公开实施例提供的重新路由的方法中获得AI模型的流程图。如图5所示，AI模型的训练步骤包括：
步骤501，获取原始网络拓扑信息。
其中，原始网络拓扑信息包括节点、链路带宽资源、链路时延和链路权值。
在一些实施例中，原始网络拓扑信息来自于与控制器相连的转发设备。通常控制器在第一次启动或重启时，转发设备通过转发协议上报给控制器。
步骤502，按照网络资源划分原始网络拓扑信息，获得网络分片信息。
其中，网络资源可以是带宽资源，也可以是其它资源。按照网络资源 划分原始网络拓扑信息，获得网络分片信息。
在一些实施例中，按照带宽资源均分方式或带宽资源随机方式对原始网络拓扑信息划分，获得网络分片信息。其中，网络分片信息包括多个分片网络的信息。分片网络的信息包括拓扑结构、链路属性等。
在一些实施例中，分片网络的拓扑结构与原始网络的拓扑结构一致。对于某条链条，每个分片网络T1、T2、T3的对应的带宽资源之和相加后等于原始网络拓扑结构T0中该链条的带宽资源。例如，假设链路B-D在分片网络T1、T2、T3中对应的带宽资源之和分别为5、8、13，这些带宽资源相加后得到的数值为26，等于原始网络拓扑结构T0中链路B-D的带宽资源。这也就是切片的特点。
步骤503，以受影响业务的业务信息、更新后的网络分片信息和受影响业务的归属分片网络作为训练样本，利用深度学习算法训练模型，获得AI模型。
在步骤503中，使用贪婪式和启发式等方式构造训练样本，建立机器学习模型，以受影响业务的业务信息和网络分片信息为输入，以受影响业务的归属分片网络为输出。在AI模型训练后，将其保存，以供后续调用。
在一些实施例中，利用深度学习训练模型，模拟物理网络T0的断纤现象，将断纤后受影响业务的业务信息、以断纤后更新的网络分片信息为输入，受影响业务的归属分片网络为输出。本实施例可以离线方式训练AI模型。利用AI模型，可以得到所有受影响业务的分片网络索引。
例如，业务4为受影响业务，占用的带宽资源为7,因此，更新的网络分片信息中，分片网络T1和分片网络T2中的剩余带宽相应增加7，业务4被重新分配到分片网络T2，即受影响业务4重新归属于分片网络T2。
再如，如表1所示，业务1可以被重新归属于分片网络T1中，业务2被重新归属于分片网络T3中。
图6为本公开实施例提供的重新路由的方法中基于受影响业务和分片网络索引获得受影响业务的新路由的流程图。如图6所示，基于受影响业 务和对应的分片网络索引获得受影响业务的新路由，包括：
步骤601，根据受影响业务的分片网络索引确定受影响业务的分片网络。
通过以受影响业务的业务信息、更新后的网络分片信息，利用AI模型获得了受影响业务的分片网络索引，按照分片网络索引确定受影响业务的分片网络。
步骤602，在受影响业务对应的分片网络内计算受影响业务的路径，获得受影响业务的新路由。
在步骤602中，以并行方式在每个分片网络内对受影响业务的路径进行计算，获得受影响业务的新路由。将受影响业务的新路由返回转发设备，转发设备以分布方式下发新路由，以恢复受影响业务。
本公开实施例提供的重新路由方法，获取受影响业务的业务信息，基于受影响业务的业务信息更新网络分片信息，基于受影响业务的业务信息和更新后的网络分片信息获得受影响业务的分片网络索引，基于受影响业务和分片网络索引确定受影响业务的新路由，该方法基于受影响业务的业务信息更新网络分片信息，然后通过受影响业务的业务信息和更新后的网络分片信息获得受影响业务的分片网络索引，因此，可以并行方式对批量的受影响业务确定新路由，降低了重新路由的时间，提高了重新路由的效率，而且，提高重新路由的成功率，提升用户体验。
另外，本公开实施例提供的重新路由的方法，考虑了业务类型与网络分片资源，基于受影响业务的业务信息更新网络分片信息，使有限的网络资源尽可能保证业务的传输成功率及传输质量。
此外，本公开实施例提供的重新路由的方法，按照网络资源划分原始网络，然后基于受影响业务的业务信息和更新后的所述网络分片信息获得所述受影响业务的分片网络索引，使得可以通过并行方式重新计算受影响业务的路由，大大降低了重新路由的耗时，理论上，重新路由的耗时与分片网络的数量成正比。
图7为本公开实施例提供的另一重新路由方法的流程图。在该方法是以节点发生故障而导致网络发生动荡为例，与该节点相关的业务受到影响，这些受影响业务需要重新路由。
如图7所示，重新路由方法包括：
步骤701，确定受影响业务的业务信息。
在本实施例中，原始网络和分片网络的拓扑结构如图2所示，原始网络和分片网络的拓扑结构包括四个节点A、B、C、D和四条链路A-B、B-D、D-C、C-A，其中，A-B链路上剩余带宽资源为30，该链路的权值为1；B-D链路上剩余带宽资源为26，该链路的权值为3；D-C链路上剩余带宽资源为30，该链路的权值为1；C-A链路上剩余带宽资源为30，该链路的权值为1。
假设节点C发生故障而导致网络发生动荡，因此，与节点C相关的业务受到影响，需要重新路由。例如，业务7的链路为A->C->D，当节点C发生故障时，追溯该业务7，获得业务7的业务信息，即获得业务7的源节点为A，宿节点为D，占用的带宽资源为15。
步骤702，释放受影响业务的带宽资源。
将步骤701中确定的受影响业务的带宽资源释放。如释放链路A->C和链路C->D的带宽资源，释放的带宽资源为15。
步骤703，基于释放的受影响业务的带宽资源更新网络分片信息。
在步骤703中，将释放的链路A->C和链路C->D的带宽资源分配给链路A-B和链路B-D，如链路A-B获得的带宽资源为15，链路B-D获得的带宽资源为15。将释放的带宽资源分配到分片网络T1、T2和T3，并更新网络分片信息。例如，在分片网络T1中，链路A-B获得的带宽资源为8，链路B-D获得的带宽资源为10，在分片网络T2中，链路A-B获得的带宽资源为7，链路B-D获得的带宽资源为5。在分片网络T3中，链路A-B和链路B-D未分配带宽资源。
在本实施例中，如图8所示，更新后的网络分片信息为：分片网络的 拓扑结构包括三个节点A、B、D和两条链路A-B、B-D，其中，A-B链路上剩余带宽资源为45，该链路的权值为1；B-D链路上剩余带宽资源为41，该链路的权值为3。
其中，在分片网络T1中，A-B链路上剩余带宽资源为18，该链路的权值为1；B-D链路上剩余带宽资源为15，该链路的权值为3。
在分片网络T2中，A-B链路上剩余带宽资源为17，该链路的权值为1；B-D链路上剩余带宽资源为13，该链路的权值为3。
在分片网络T3中，A-B链路上剩余带宽资源为10，该链路的权值为1；B-D链路上剩余带宽资源为13，该链路的权值为3。
步骤704，基于受影响业务的业务信息和更新后的网络分片信息获得受影响业务的分片网络索引。
在步骤704中，根据受影响的业务7的业务信息和更新后的网络分片信息利用AI模型，获得受影响业务7的分片网络索引。例如，业务7的分片网络索引为分片网络T1，即受影响业务7重新归属于分片网络T1。因为在分片网络T2中，链路A-B的剩余带宽资源为17，但链路B-D的剩余带宽资源只有13，而业务7需要的带宽资源为15，因此，分片网络T2无法满足业务7。在分片网络T3中，链路A-B的剩余带宽资源为10，链路B-D的剩余带宽资源只有13，也无法满足业务7所需的带宽资源。在分片网络T1中，链路A-B的剩余带宽资源为18，但链路B-D的剩余带宽资源为15，能够满足业务7所需的带宽资源。
需要说明的是，本公开实施例是以链路和节点发生故障为例介绍重新路由的方法，但本公开并不局限于此。在实际应用中，当进行站点维修或改造时，也采用本公开实施例提供的重新路由方法对受影响的业务重新路由。
步骤705，在受影响业务对应的分片网络内计算受影响业务的路径，获得受影响业务的新路由。
在分片网络内对受影响业务的路径进行计算，获得受影响业务的新路 由。将受影响业务的新路由返回转发设备，转发设备以分布方式下发新路由，以恢复受影响业务。
例如，在本实施例中，在分片网络T1内获得受影响业务的新路由为B->D。
第二方面，本公开实施例提供了一种重新路由装置。图9为本公开实施例提供的一种重新路由装置的原理框图。参照图9，重新路由装置包括：
更新模块901，设置为基于受影响业务的业务信息更新网络分片信息。
其中，网络分片信息是对原始网络拓扑结构划分后获得的分片网络的信息；
其中，业务信息包括业务带宽资源、源节点、宿节点、服务质量、当前部署路径、约束条件等信息。业务信息可以通过转发设备上报给控制器，由控制器对业务的路径和网络的资源进行管理。
在一些实施例中，更新模块通过该拓扑加载单元和分片资源管理单元，其中，拓扑加载单元设置为获得原始网络的节点信息、链路信息等，并通过协议上报控制器。分片资源管理单元设置为对原始网络拓扑结构进行分片管理，将原始网络拓扑结构划分为多个分片网络，而且，每个分片网络的结构和原始网络拓扑结构一致。分片资源管理单元还设置为维护每个分片网络对应的业务信息。
更新模块901设置为对网络分片信息进行更新。如基于受影响业务的业务信息追溯受影响业务，确定删除的节点和链路，以及链路资源的恢复、释放的带宽资源，然后对网络分片信息进行更新。
分片网络索引模块902，设置为基于受影响业务的业务信息和更新后的网络分片信息获得受影响业务的分片网络索引。
其中，分片网络索引是受影响业务与该业务重新归属的分片网络的对应关系。
在一些实施例中，分片网络索引模块902可以是AI模型，将受影响业务的业务信息和更新后的网络分片信息输入AI模型，获得受影响业务的分片网络索引。
路由生成模块903，设置为基于受影响业务和分片网络索引获得受影响业务的新路由。
在路由生成模块903以并行方式在每个分片网络内对受影响业务的路径进行计算，获得受影响业务的新路由。将受影响业务的新路由返回转发设备，转发设备以分布方式下发新路由，以恢复受影响业务。
本实施例提供的重新路由装置，通过获取模块获取受影响业务的业务信息，利用更新模块基于受影响业务的业务信息更新网络分片信息，分片网络索引模块基于受影响业务的业务信息和更新后的网络分片信息获得受影响业务的分片网络索引，路由生成模块基于受影响业务和分片网络索引确定受影响业务的新路由，该装置降低了重新路由的时间，提高了重新路由的效率，而且，提高重新路由的成功率，提升用户体验。
第三方面，参照图10，本公开实施例提供了一种电子设备，其包括：
一个或多个处理器1001；
存储器1002，其上存储有一个或多个程序，当一个或多个程序被一个或多个处理器执行，使得一个或多个处理器实现上述任意一项的重新路由的方法；
一个或多个I/O接口1003，连接在处理器与存储器之间，配置为实现处理器与存储器的信息交互。
其中，处理器1001为具有数据处理能力的器件，其包括但不限于中央处理器(CPU)等；存储器1002为具有数据存储能力的器件，其包括但不限于随机存取存储器(RAM，更具体如SDRAM、DDR等)、只读存储器(ROM)、带电可擦可编程只读存储器(EEPROM)、闪存(FLASH)；I/O接口(读写接口)1003连接在处理器1001与存储器1002间，能实现 处理器1001与存储器1002的信息交互，其包括但不限于数据总线(Bus)等。
在一些实施例中，处理器1001、存储器1002和I/O接口1003通过总线相互连接，进而与计算设备的其它组件连接。
第四方面，本公开实施例提供了一种计算机可读介质，其上存储有计算机程序，程序被处理器执行时实现上述任意一种重新路由的方法。
本领域普通技术人员可以理解，上文中所公开方法中的全部或某些步骤、***、装置中的功能模块/单元可以被实施为软件、固件、硬件及其适当的组合。在硬件实施方式中，在以上描述中提及的功能模块/单元之间的划分不一定对应于物理组件的划分；例如，一个物理组件可以具有多个功能，或者一个功能或步骤可以由若干物理组件合作执行。某些物理组件或所有物理组件可以被实施为由处理器，如中央处理器、数字信号处理器或微处理器执行的软件，或者被实施为硬件，或者被实施为集成电路，如专用集成电路。这样的软件可以分布在计算机可读介质上，计算机可读介质可以包括计算机存储介质(或非暂时性介质)和通信介质(或暂时性介质)。如本领域普通技术人员公知的，术语计算机存储介质包括在用于存储信息(诸如计算机可读指令、数据结构、程序模块或其它数据)的任何方法或技术中实施的易失性和非易失性、可移除和不可移除介质。计算机存储介质包括但不限于RAM、ROM、EEPROM、闪存或其它存储器技术、CD-ROM、数字多功能盘(DVD)或其它光盘存储、磁盒、磁带、磁盘存储或其它磁存储装置、或者可以用于存储期望的信息并且可以被计算机访问的任何其它的介质。此外，本领域普通技术人员公知的是，通信介质通常包含计算机可读指令、数据结构、程序模块或者诸如载波或其它传输机制之类的调制数据信号中的其它数据，并且可包括任何信息递送介质。
本文已经公开了示例实施例，并且虽然采用了具体术语，但它们仅用于并仅应当被解释为一般说明性含义，并且不用于限制的目的。在一些实 例中，对本领域技术人员显而易见的是，除非另外明确指出，否则可单独使用与特定实施例相结合描述的特征、特性和/或元素，或可与其它实施例相结合描述的特征、特性和/或元件组合使用。因此，本领域技术人员将理解，在不脱离由所附的权利要求阐明的本公开的范围的情况下，可进行各种形式和细节上的改变。

Claims (10)

  1. 一种重新路由方法，其包括：
    基于受影响业务的业务信息更新网络分片信息；其中，所述业务信息包括业务带宽资源，所述网络分片信息是对原始网络拓扑结构划分后获得的分片网络的信息；
    基于所述受影响业务的业务信息和更新后的所述网络分片信息获得所述受影响业务的分片网络索引；其中，所述分片网络索引是所述受影响业务与分片网络的对应关系；
    基于所述受影响业务和所述分片网络索引确定所述受影响业务的新路由。
  2. 根据权利要求1所述的方法，其中，所述基于所述受影响的业务信息更新网络分片信息，包括：
    释放所述受影响业务的带宽资源；
    基于释放的所述受影响业务的带宽资源更新网络分片信息。
  3. 根据权利要求1所述的方法，其中，所述基于所述受影响的业务信息和更新后的所述网络分片信息获得所述受影响业务的分片网络索引，包括：
    基于所述受影响的业务信息和更新后的所述网络分片信息，利用AI模型获得所述受影响业务的分片网络索引。
  4. 根据权利要求3所述的方法，其中，所述AI模型通过以下步骤获得：
    获取原始网络拓扑信息；其中，所述原始网络拓扑信息包括节点、链路带宽资源、链路时延和链路权值；
    划分所述原始网络拓扑信息，获得网络分片信息；其中，所述网络分片信息包括多个分片网络的信息；
    以受影响业务的业务信息、更新后的网络分片信息和受影响业务的归属分片网络作为训练样本，利用深度学习算法训练模型，获得AI模型。
  5. 根据权利要求4所述的方法，其中，所述划分所述原始网络拓扑信息，获得网络分片信息，包括：
    按照带宽均分方式或带宽随机方式对所述原始网络拓扑信息进行划分，获得网络分片信息。
  6. 根据权利要求4所述的方法，其中，所述分片网络的拓扑结构与所述原始网络的拓扑结构一致。
  7. 根据权利要求1-6任意一项所述的方法，其中，所述基于所述受影响业务和对应的所述分片网络索引获得所述受影响业务的新路由，包括：
    根据所述受影响业务的分片网络索引确定所述受影响业务的分片网络；
    在所述受影响业务对应的分片网络内计算所述受影响业务的路径，获得所述受影响业务的新路由。
  8. 一种重新路由装置，其包括：
    更新模块，设置为基于受影响业务的业务信息更新网络分片信息；其中，所述业务信息包括业务带宽资源，所述网络分片信息是对原始网络拓扑结构划分后获得的分片网络的信息；
    分片网络索引模块，设置为基于所述受影响业务的业务信息和更新后的所述网络分片信息获得所述受影响业务的分片网络索引；其中，所述分片网络索引是所述受影响业务与该业务重新归属的分片网络的对应关系；
    路由生成模块，设置为基于所述受影响业务和所述分片网络索引获得所述受影响业务的新路由。
  9. 一种电子设备，其包括：
    一个或多个处理器；
    存储装置，其上存储有一个或多个程序，当所述一个或多个程序被所述一个或多个处理器执行，使得所述一个或多个处理器实现根据权利要求1-7任意一项所述的方法；
    一个或多个I/O接口，连接在所述处理器与存储器之间，配置为实现所述处理器与存储器的信息交互。
  10. 一种计算机可读介质，其上存储有计算机程序，所述程序被处理器执行时实现根据权利要求1-7任意一项所述的方法。
PCT/CN2020/130650 2020-04-23 2020-11-20 一种重新路由方法及装置、电子设备、计算机可读介质 WO2021212823A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/920,500 US20230155920A1 (en) 2020-04-23 2020-11-20 Rerouting Method and Apparatus, Electronic Device, and Computer Readable Medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010328422.7A CN113556281A (zh) 2020-04-23 2020-04-23 一种重新路由方法及装置、电子设备、计算机可读介质
CN202010328422.7 2020-04-23

Publications (1)

Publication Number Publication Date
WO2021212823A1 true WO2021212823A1 (zh) 2021-10-28

Family

ID=78101110

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/130650 WO2021212823A1 (zh) 2020-04-23 2020-11-20 一种重新路由方法及装置、电子设备、计算机可读介质

Country Status (3)

Country Link
US (1) US20230155920A1 (zh)
CN (1) CN113556281A (zh)
WO (1) WO2021212823A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115225561A (zh) * 2022-08-15 2022-10-21 南京邮电大学 一种基于图结构特征的路由优化方法与***

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016165422A1 (zh) * 2015-09-21 2016-10-20 中兴通讯股份有限公司 业务动态恢复的方法、装置及软件定义网络sdn控制器
CN108075982A (zh) * 2017-12-14 2018-05-25 中国科学技术大学苏州研究院 软件定义网络中实时的路由更新方法
CN109429277A (zh) * 2017-07-05 2019-03-05 中兴通讯股份有限公司 网络切片的选择方法、装置及***
CN110167094A (zh) * 2018-02-12 2019-08-23 中兴通讯股份有限公司 一种路由选择方法、装置、设备、***和存储介质
US20190394132A1 (en) * 2017-03-31 2019-12-26 Huawei Technologies Co., Ltd. System and Method for Network Slicing for Service-Oriented Networks

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101286892B (zh) * 2007-04-12 2012-09-05 华为技术有限公司 进行业务恢复的装置和方法
CN102223310A (zh) * 2011-07-12 2011-10-19 无锡北邮感知技术产业研究院有限公司 一种分布式智能光网络中基于并行信令的组播恢复机制
CN105721192A (zh) * 2014-12-05 2016-06-29 中兴通讯股份有限公司 实现容量规划的方法和装置
CN104980351A (zh) * 2014-12-26 2015-10-14 深圳市新格林耐特通信技术有限公司 一种基于sdn网络的快速重路由方法
US10129894B2 (en) * 2016-03-04 2018-11-13 Huawei Technologies Co., Ltd. Systems and methods for performing traffic engineering through network slices
US10637725B2 (en) * 2016-11-01 2020-04-28 Huawei Technologies Co., Ltd. System and method for network slice management in a management plane
US10581666B2 (en) * 2017-03-21 2020-03-03 Huawei Technologies Co., Ltd. Method and apparatus for operating network slices using shared risk groups
CN112368977B (zh) * 2018-06-15 2023-10-13 瑞典爱立信有限公司 配置网络切片
US10892958B2 (en) * 2018-08-03 2021-01-12 Huawei Technologies Co., Ltd. Methods and functions of network performance monitoring and service assurance
US11019528B2 (en) * 2018-10-29 2021-05-25 Verizon Patent And Licensing Inc. Method and system for admission control with network slice capability
US11424977B2 (en) * 2018-12-10 2022-08-23 Wipro Limited Method and system for performing effective orchestration of cognitive functions in distributed heterogeneous communication network
CN109768924B (zh) * 2019-02-14 2021-06-08 山东省计算中心（国家超级计算济南中心） 一种面向多流共存的sdn网络多链路故障恢复方法及***
US11246088B2 (en) * 2020-04-10 2022-02-08 Cisco Technology, Inc. Failover and management of multi-vendor network slices

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016165422A1 (zh) * 2015-09-21 2016-10-20 中兴通讯股份有限公司 业务动态恢复的方法、装置及软件定义网络sdn控制器
US20190394132A1 (en) * 2017-03-31 2019-12-26 Huawei Technologies Co., Ltd. System and Method for Network Slicing for Service-Oriented Networks
CN109429277A (zh) * 2017-07-05 2019-03-05 中兴通讯股份有限公司 网络切片的选择方法、装置及***
CN108075982A (zh) * 2017-12-14 2018-05-25 中国科学技术大学苏州研究院 软件定义网络中实时的路由更新方法
CN110167094A (zh) * 2018-02-12 2019-08-23 中兴通讯股份有限公司 一种路由选择方法、装置、设备、***和存储介质

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115225561A (zh) * 2022-08-15 2022-10-21 南京邮电大学 一种基于图结构特征的路由优化方法与***
CN115225561B (zh) * 2022-08-15 2022-12-06 南京邮电大学 一种基于图结构特征的路由优化方法与***

Also Published As

Publication number Publication date
US20230155920A1 (en) 2023-05-18
CN113556281A (zh) 2021-10-26

Similar Documents

Publication Publication Date Title
US10374956B1 (en) Managing a hierarchical network
CN111165019B (zh) 接入网中的控制器
US20190394132A1 (en) System and Method for Network Slicing for Service-Oriented Networks
CN109905251B (zh) 网络管理方法、装置、电子设备和存储介质
US20150234674A1 (en) Method, System and Apparatus for Creating Virtual Machine
US9110884B2 (en) Message publishing and subscribing method and apparatus
CN108243106A (zh) 控制网络切片的方法、转发设备、控制设备和通信***
CN112738820A (zh) 一种服务功能链的动态部署方法、装置及计算机设备
WO2021216613A4 (en) System, method and apparatus for generating and searching a topology of resources among multiple cloud computing environments
EP3682597B1 (en) Modeling access networks as trees in software-defined network controllers
US10250484B2 (en) Network service computation system
WO2016086709A1 (zh) 实现容量规划的方法和装置
CN110855737B (zh) 一种一致性级别可控的自适应数据同步方法和***
WO2021022944A1 (zh) 针对具有栈深约束的路径计算方法及装置
WO2016165422A1 (zh) 业务动态恢复的方法、装置及软件定义网络sdn控制器
CN115514696B (zh) 一种传递算力资源信息的方法、装置及设备
CN107276916B (zh) 基于协议无感知转发技术的交换机流表管理方法
CN113645146B (zh) 基于新流密度的软件定义网络控制器负载均衡方法及***
US20190349288A1 (en) Data Forwarding Method and Apparatus
CN106713378A (zh) 实现多个应用服务器提供服务的方法和***
CN108476175A (zh) 使用对偶变量的传送sdn流量工程方法与***
WO2021212823A1 (zh) 一种重新路由方法及装置、电子设备、计算机可读介质
CN106817306B (zh) 一种确定目标路由的方法及装置
CN103795606A (zh) 一种光网络中基于sdn的服务迁移控制方法
CN107302849B (zh) 一种光路径的分配方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20932603

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 13/03/2023)

122 Ep: pct application non-entry in european phase

Ref document number: 20932603

Country of ref document: EP

Kind code of ref document: A1