CN111327525A - Network routing method and device based on segmented routing - Google Patents

Abstract

Embodiments of the present invention provide a method and device for network routing based on segment routing, an SDN deployment architecture constructed based on data streams in different high-traffic service scenarios, comprehensively considering grid information constructed based on test data between access routers and from The port state information collected by the router is used as the metric value to calculate the optimal path method. For the SDN deployment architecture, the present invention uses the existing network equipment as much as possible to adapt to different high-traffic services routing adjustment; Based on the grid information and the port status information collected from the router, the metric value is differentiated and adjusted according to the needs of the service traffic, and the PCE calculates the optimal path based on this metric value.

Description

A method and device for network routing based on segment routing

technical field

Embodiments of the present invention relate to the field of communications technologies, and more particularly, to a method and apparatus for network routing based on segment routing.

Background technique

After more than 40 years of rapid development, the Internet has developed from a dedicated academic network to a wider commercial platform, and has gradually penetrated into every aspect of people's lives. However, with the explosive growth of global user scale and network applications, the existing Internet has exposed various shortcomings and defects of the original design. For example, the original design idea of the existing network is mainly that a network is mainly oriented to a specific service, such as a telecommunication network mainly oriented to voice services, a TV network mainly oriented to video services, the Internet mainly oriented to data transmission services, and support for other services. Poor sex. To this end, it is urgent to create a brand-new Internet architecture to meet the needs of economic and social development for the future Internet.

The traditional network constantly adjusts the network architecture and configuration passively, which cannot match the rapid development of services, and makes the network deployment more complex and difficult to maintain. Dynamic adjustment to quickly meet business changing needs. The network architecture needs to evolve from "network adaptation services" to "service-driven networks".

Software-Defined Networking (SDN) is an emerging network architecture that separates the control plane of the existing highly integrated network equipment from the complex underlying equipment, allowing network managers to Configuration, management and control become smart and efficient, and the network can be modified programmatically to cope with the ever-expanding network size and rapidly increasing traffic demands.

At present, the existing research schemes based on SDN routing mainly include: (1) Use the SDN controller to collect the information data files of the networking connection, and update the network nodes and network adjacency table according to the information data read from the information data files; The obtained network nodes and adjacency lists use the balanced global load balancing algorithm to perform route calculation and route selection; the obtained optimal path is saved and written to a file. It can improve the stability of the entire system and the optimization of path selection, and improve the utilization rate of the overall link; (2) divide the network into multiple smaller subnets, elect a traffic aggregation node within the subnet, and The communication between common nodes in the network is forwarded by the traffic convergence node, and the communication between common nodes across subnets is forwarded by the traffic convergence node of this subnet. Compared with the prior art, the positive effect is: by splitting a large network into multiple subnets, and applying different routing calculation and control methods within and between subnets, a routing entry based on the OpenFlow protocol is proposed at the same time. The aggregation method effectively reduces the computational complexity when calculating large-scale network routes, improves the network utilization of high-bandwidth, low-latency, and low-jitter links, reduces the system's flow table entry resource requirements, and enhances SDN. The performance and adaptability of the technology in a large-scale networking environment; (3) The ingress provider edge device (ProviderEdge, PE) converts the first CE according to the label of the second customer edge device (Customer Edge, CE) and the label of the target PE The sent information is encapsulated into a data packet containing label information; the ingress PE sends the encapsulated data packet to the core layer device P, and the core layer device establishes a forwarding path according to the label of the target PE in the data packet, and sends the encapsulated data packet to the core layer device P. The data packet is routed to the target PE, and the target PE routes the information sent by the first CE in the data packet to the second CE according to the label of the second CE. The method, device and system provided by the present invention use a centralized deployment SDN controller to publish VPN (Virtual Private Network, virtual private network) routing information for forwarding, reduce operation and maintenance workload, and improve operator business operation efficiency.

In the above method (1), the network node and adjacency table adopts the SDN network routing calculation method of balanced global load balancing, which does not consider the port status and link quality information, and also cannot select paths for specific services. Architecture evolution requirements of service-driven networks. In the above method (2), the SDN technology based on OpenFlow is the most widely known, and it is also the most influential protocol in the current SDN framework. However, the concept of multi-level flow table is proposed in OpenFlowv1.1, and there is no limit to the number of levels, which makes the flow table entries of the control plane very large, and the requirements for hardware performance are getting higher and higher. At the same time, the speed of matching the table lookup is also Affected to a certain extent, it is suitable for small-scale deployment, but difficult to adapt to large networks. The OpenFlow controller needs to configure all switches on the data plane, including the distribution of flow tables and the distribution of related matching rules, which greatly increases the load on the controller. In the above method (3), in order to solve the shortcoming of the large flow table of OpenFlow large-scale network application, the large-scale network is divided into multiple smaller subnets, and a traffic convergence node is elected in the subnet, and the common network in the subnet is selected. The communication between nodes is forwarded by the traffic aggregation node, and the communication between common nodes across subnets is forwarded by the traffic aggregation node of this subnet, but different routing calculation and control methods need to be applied within the network and between subnets, and the data needs to be All switches on the plane are configured, including the distribution of flow tables and related matching rules, which greatly increases the load on the controller.

SUMMARY OF THE INVENTION

Embodiments of the present invention provide a method and apparatus for network routing based on segment routing that overcomes the above problem or at least partially solves the above problem.

In a first aspect, an embodiment of the present invention provides a network routing method based on segment routing, including:

Based on the packet loss rate, round-trip delay and jitter of each port of the router in segment routing to adjacent routers, the mesh weight calculation parameters representing mesh state information are obtained; based on the remaining bandwidth of the link and the CRC error of the cyclic redundancy check The number of packets obtains the port weight calculation parameter representing the port state information;

The PCE path calculation metric value of the path calculation unit is obtained based on the Mesh weight calculation parameter and the port weight calculation parameter, and the shortest path from the source point to the destination node is obtained based on the PCE path calculation metric value.

Preferably, before acquiring the Mesh weight calculation parameters representing the mesh state information, the method further includes:

Perform TraceRoute test and ping test based on different city edge router node probes to obtain the packet loss rate, round-trip delay and jitter between direct link ports.

Preferably, the calculation parameters of mesh weights representing mesh state information are obtained, specifically including:

Based on the packet loss rate, round-trip delay, and jitter from the M ports of router x to neighboring routers, respectively, the mesh weight calculation parameter θ _Mesh (pcc(m) _x ) representing mesh state information is obtained as:

In the above formula, L _d (pcc(m) _x ), D _d (pcc(m) _x ) and J _d (pcc(m) _x ) represent the packet loss rate, round-trip delay and jitter of port m of router x, respectively Euclidean distance to the average value of M ports; σ _L , σ _D and σ _J are the standard deviation of port m in packet loss rate, round-trip delay and jitter to the average value of M ports.

Preferably, before obtaining the port weight calculation parameter representing the port state information based on the remaining bandwidth of the link and the number of CRC error packets, the method further includes:

Based on the deployment of Simple Network Management Protocol SNMP and NetStream on existing network devices, collect the remaining bandwidth of the link and the number of CRC error packets received by the port within the set time.

Preferably, the port weight calculation parameter θ _Port (pcc(m) _x ) is:

In the above formula, RBW _d (pcc(m) _x ) and CRC _d (pcc(m) _x ) respectively represent the Euclidean distance between the remaining bandwidth and the number of CRC error packets of port m of router x to the average value of M ports, σ _RBW and σ _CRC are the standard deviation of the residual bandwidth of port m and the number of CRC error packets to the average value of M ports, respectively.

Preferably, the path calculation unit PCE path calculation metric value is obtained based on the Mesh weight calculation parameter and the port weight calculation parameter, specifically including:

Mesh weight calculation parameter θ _Mesh (pcc(m) _x ) and port weight calculation parameter θ _Port (pcc(m) _x ) of router x port m, and weights α and θ _Port of θ _Mesh (pcc(m) _x ) The weight β of (pcc(m) _x ) is used to construct a weighting function for calculating the metric value of the PCE path of the mth port of router x;

The values of the weight α and the weight β are obtained based on the service traffic type to obtain the weighting coefficient of the PCE path calculation metric value, and the PCE path calculation metric value is obtained based on the weighting coefficient.

Preferably, the shortest path from the source point to the destination node is obtained by calculating the metric value based on the PCE path, specifically including:

The metric value is calculated based on the obtained PCE path, and the PCE calculates the shortest path from the source point to the destination node according to the Dijkstra algorithm.

In a second aspect, an embodiment of the present invention provides a network routing device based on segment routing, including:

The first module is used to obtain the mesh weight calculation parameters representing mesh state information based on the packet loss rate, round-trip delay and jitter of each port of the router in segment routing to adjacent routers; The number of error packets in the redundancy check CRC obtains the port weight calculation parameter representing the port status information;

The second module is configured to obtain the PCE path calculation metric value of the path calculation unit based on the Mesh weight calculation parameter and the port weight calculation parameter, and obtain the shortest path from the source point to the destination node based on the PCE path calculation metric value.

In a third aspect, an embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, the processor implementing the program as described in the first aspect when the processor executes the program Steps of the provided method.

In a fourth aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, implements the steps of the method provided in the first aspect.

The embodiment of the present invention proposes a network routing method and device based on segment routing, an SDN deployment architecture constructed based on data streams in different large-traffic service scenarios, comprehensively considering grid information constructed based on test data between access routers, and The port state information collected from the router is used as the metric value to calculate the optimal path method. For the SDN deployment architecture, the present invention uses the existing network equipment as much as possible to adapt to different high-traffic services routing adjustment; based on the test data between the access routers. Grid information and port status information collected from routers, metric values that are differentiated and adjusted according to service traffic requirements, and PCE calculates the optimal path based on this metric value.

Description of drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description These are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to these drawings without creative efforts.

1 is a schematic diagram of a network routing method based on segment routing according to an embodiment of the present invention;

2 is a schematic diagram of Segment Routing applying a PCE architecture according to an embodiment of the present invention;

3 is a schematic diagram of a MeshICMPing test architecture according to an embodiment of the present invention;

4 is a schematic diagram of a network routing device based on segment routing according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of a physical structure of an electronic device according to an embodiment of the present invention.

Detailed ways

In order to make the purposes, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments These are some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

Because the network node and the adjacency table in the prior art adopt the SDN network routing calculation method of balanced global load balancing, the port status and link quality information are not considered, and the path selection cannot be performed for a specific service, and the service-driven approach cannot be achieved. The architecture evolution requirements of the network; in order to solve the shortcomings of the large-scale network application flow table of OpenFlow, the large-scale network is divided into multiple smaller subnets, and a traffic aggregation node is elected in the subnet, and the common node in the subnet is selected. The communication between them is forwarded by the traffic aggregation node, and the communication between common nodes across subnets is forwarded by the traffic aggregation node of this subnet, but different routing calculation and control methods need to be applied within the network and between subnets, and the data plane needs to be All switches are configured, including the distribution of flow tables and related matching rules, which greatly increases the load of the controller; routing selection for different services reduces the workload of operation and maintenance, and improves the operation efficiency of operators without services. Link quality and port information are not considered during the establishment of Segment Routing and MPLS tunnels.

Therefore, each embodiment of the present invention comprehensively considers a grid information constructed based on test data between access routers and port status information collected from routers by using existing network equipment as much as possible to adapt to different high-traffic service routing adjustments. The metric value for differential adjustment of service traffic requirements, and based on this metric value PCE performs optimal path calculation. The following will expand the description and introduction through multiple embodiments.

1 is a network routing method based on segment routing provided by an embodiment of the present invention, including:

S1. Based on the packet loss rate, round-trip delay and jitter from each port of the router in the segment routing to the adjacent router, obtain the mesh weight calculation parameter representing the mesh state information; based on the remaining bandwidth of the link and the cyclic redundancy check The number of CRC error packets obtains the port weight calculation parameter representing the port status information;

S2. Obtain the PCE path calculation metric value of the path calculation unit based on the Mesh weight calculation parameter and the port weight calculation parameter, and obtain the shortest path from the source node to the destination node based on the PCE path calculation metric value.

In this embodiment, the OpenFlow controller needs to configure all switches in the data plane, including the distribution of flow tables and the distribution of related matching rules, which greatly increases the load of the controller. The controller in SR (Segment Routing, segment routing) only needs to find the forwarding label stack to the ingress border router, and does not need to configure each node, and the state of the data flow is only saved in the ingress border router, and the middle Nodes only need to forward according to the label stack encapsulated in the packet header, which greatly reduces the difficulty of network deployment. Therefore, SR will greatly reduce the network forwarding delay compared with OpenFlow, and also greatly improve the scalability, making it easier to deploy in operator-type large networks.

In this embodiment, since it is considered to be deployed in the large-scale network of an actual operator, and the existing network equipment is used as much as possible, the scheme of separating the control plane and the data plane through the model of using source routing through SR is adopted. From the perspective of the overall network, the plane considers a grid state information constructed based on test data between access routers and ports collected from routers for different high-traffic service scenarios (download and video services) data streams. The state information is used as a metric to calculate the optimal path and encode the path to the data plane; the data plane is responsible for forwarding packets from the ingress node to the correct egress node. The solution can make changes to existing network equipment as little as possible for large-scale network applications at the operator level, and can provide differentiated routing and forwarding functions for different high-traffic services.

In this embodiment, the data plane of the SR uses the data plane of MPLS (Multi-Protocol Label Switching, Multi-Protocol Label Switching), and the control plane is implemented by using the PCE architecture. The PCE architecture is composed of a Path Computation Element (PCE) and a path computation A client (Path Computation Client, PCC) is formed, and the PCE and the PCC use PCEP (Path Computation Element Communication Protocol, PCEP) to communicate.

As shown in Figure 2, the PCE acts as the control plane of the SR and calculates the traffic engineering path for the SR data plane. The SR data plane uses the MPLS data plane; the router of the SR data plane in the figure acts as the PCC.

PCE collects network-wide topology information through Traffic Engineering Database (TE-DB), hosts Label Switched Path Database (LSP-DB) to collect LSP State information, globally manages and uniformly allocates network resources, and centralizes computing The LSP service path is formed by the SR to form an MPLS tunnel for data plane forwarding; the PCC is a forwarding network element. When the LSP service path needs to be calculated, a path calculation request is sent to the PCE, and the MPLS tunnel data result formed by the SR is fed back to the PCC. PCE adopts stateful PCE technology, follows strict synchronization between PCE and traffic engineering database TE-DB, maintains state when setting active paths and using reserved resources of the network, and has the following functions: (1), through IGP and The traffic engineering resource update discovers the network topology and resources; (2), collect the real-time data information of the link, forward the plane router port data and push it to the link real-time big data analysis platform, and receive the link and port data analysis results; (3) ), monitor the network status, measure real-time traffic demand and LSP statistics, re-optimize the path based on this information, and form a label stack after the calculation is completed; (4), use the PCEP protocol to accept the request from the data plane and put the calculated label stack down It is sent to the device to instruct the device to forward it according to the path specified by the label stack.

Mesh network is a "wireless mesh network", which is a "multi-hop" network, developed from an ad hoc network, and is one of the key technologies to solve the "last mile" problem. In the process of evolution to the next generation network, wireless is an indispensable technology. Wireless mesh can cooperate with other networks to communicate. It is a dynamic and scalable network architecture, and any two devices can maintain wireless interconnection.

In this embodiment, based on the packet loss rate, round-trip delay and jitter of each port of the router in segment routing to adjacent routers, the mesh weight calculation parameters representing mesh state information are obtained; The number of error packets in the redundancy check CRC obtains the port weight calculation parameter representing the port status information; the Mesh weight calculation parameter and the port weight calculation parameter are used as the PCE path calculation metric value to calculate the optimal path parameters, and differentiated according to the needs of service traffic The adjusted PCE path calculation metric value, and the PCE performs optimal path calculation based on this metric value. For large-scale network applications at the operator level, the existing network equipment can be changed as little as possible, and it can provide differentiated routing and forwarding functions for different high-traffic services.

On the basis of the above embodiment, before acquiring the Mesh weight calculation parameter representing the mesh state information, the method further includes:

Perform TraceRoute test and ping test based on different city edge router node probes to obtain the packet loss rate, round-trip delay and jitter between direct link ports.

In this embodiment, the metric values used in the path calculation are constructed by integrating the Mesh weight calculation parameters and the router port information calculation parameters. Specifically, as shown in Figure 2, it is a schematic diagram of the MeshICMPing test architecture. Based on the MeshICMPing global positioning analysis, according to different locations For the city edge router node probes, 12*24 rounds per day (5-minute granularity) are carried out in pairs, and each round of TraceRoute+Ping tests is performed N times. Obtain the packet loss rate (Packet Loss Rate) L, jitter (Jitter) J, and delay (Delay) D information between directly connected ports based on the TraceRoute test and Ping test results.

As shown in Figure 3, according to the TraceRoute test results from RT1 to RT7, the routing path from RT1 to RT7 can be obtained as RT1→RT11→RT12→RT15→RT7, and the return time of three ICMP packets from RT1 to each hop router TRT1→RT11(1) , TRT1→RT11(2), and TRT1→RT11(3).

Assuming the ping test results of RT1 to RT7, the return delay and jitter information of ICMP packets from RT1 to RT7 can be obtained, and each direct link of TraceRoute can be obtained by calculating the information collected by each router (TRT1→RT11(1),… , TRT1→RT11(N); TRT11→RT12(1),…,TRT11→RT12(N); etc.) round-trip delay (Delay, D), jitter (Jitter, J) and packet loss rate (Packet Loss Rate) , L) and other information.

On the basis of the above embodiments, the Mesh weight calculation parameters representing mesh state information are obtained, specifically including:

Based on the packet loss rate, round-trip delay, and jitter from the M ports of router x to neighboring routers, respectively, the mesh weight calculation parameter θ _Mesh (pcc(m) _x ) representing mesh state information is obtained as:

In the above formula (1), L _d (pcc(m) _x ), D _d (pcc(m) _x ) and J _d (pcc(m) _x ) represent the packet loss rate and round-trip time of port m of router x, respectively. The Euclidean distance of delay and jitter to the average value of M ports; σ _L , σ _D and σ _J are the standard deviation of the packet loss rate, round-trip delay and jitter of port m to the average value of M ports, respectively.

In this embodiment, L _d (pcc(m) _x ), J _d (pcc(m) _x ) and D _d (pcc(m) _x ), σ _L , σ _J and σ _D are as follows: (3) shows:

和

为分别为第x台路由器的M个端口的平均丢包率、平均往返时延和平均抖动。In the above formulas (2) and (3), L(pcc _x ), D(pcc _x ) and J(pcc _x ) are the packet loss rate, round-trip delay and jitter of port m of the xth router respectively;

and

are the average packet loss rate, average round-trip delay and average jitter of M ports of the xth router, respectively.

On the basis of the foregoing embodiments, before obtaining the port weight calculation parameter representing the port state information based on the remaining bandwidth of the link and the number of CRC error packets, the method further includes:

Based on the deployment of Simple Network Management Protocol SNMP and NetStream on existing network devices, collect the remaining bandwidth of the link and the number of CRC error packets received by the port within the set time.

In this embodiment, the normalized port weight calculation parameter θ _Port (pcc(m) _x ) is based on the remaining link bandwidth (Residual Band Width) RBW collected based on protocols such as SNMP and NetStream deployed on existing network devices, and Data such as the number of CRC (Cyclic Redundancy Check, Cyclic Redundancy Check) error packets received by the port in a short period of time are obtained.

On the basis of the above embodiments, the port weight calculation parameter θ _Port (pcc(m) _x ) is:

In the above formula, RBW _d (pcc(m) _x ) and CRC _d (pcc(m) _x ) respectively represent the Euclidean distance between the remaining bandwidth and the number of CRC error packets of port m of router x to the average value of M ports, σ _RBW and σ _CRC are the standard deviation of the residual bandwidth of port m and the number of CRC error packets to the average value of M ports, respectively.

In this embodiment, RBW _d (pcc(m) _x ), CRC _d (pcc(m) _x ), σ _RBW and σ _CRC are shown in the following equations (5) and (6):

和

分别为路由器x在M个端口m的平均剩余带宽以和平均CRC错包数目。In the above formula, RBW(pcc(m) _x ) and CRC(pcc(m) _x ) are the number of error packets in the remaining bandwidth and CRC of port m of router x, respectively,

and

are the average remaining bandwidth of router x on M ports m and the average number of CRC error packets, respectively.

On the basis of the above embodiments, the path calculation unit PCE path calculation metric value is obtained based on the Mesh weight calculation parameter and the port weight calculation parameter, which specifically includes:

Mesh weight calculation parameter θ _Mesh (pcc(m) _x ) and port weight calculation parameter θ _Port (pcc(m) _x ) of router x port m, and weights α and θ _Port of θ _Mesh (pcc(m) _x ) The weight β of (pcc(m) _x ) is used to construct a weighting function for calculating the metric value of the PCE path of the mth port of router x;

The values of the weight α and the weight β are obtained based on the service traffic type to obtain the weighting coefficient of the PCE path calculation metric value, and the PCE path calculation metric value is obtained based on the weighting coefficient.

In this embodiment, the PCE path calculates the metric value based on the mesh weight calculation parameter θ _Mesh (pcc(m) _x ) and the port weight calculation parameter θ _Port (pcc(m) _x ). The linear decision algorithm is mainly based on these two The parameters are used as the basis for discrimination, and the metric value of the LSP is calculated. The weighting function of the metric value of the mth port of router x is shown in the following formula (7):

where θ _Mesh (pcc(m) _x ) and θ _Port (pcc(m) _x ) are the Mesh weight calculation parameters and port weight calculation parameters of router x port m, respectively, α and β are θ _Mesh (pcc(m) _x ) and the weight of θ _Port (pcc(m) _x ), α+β=1. The metric value calculation adopts the decision rule of the minimum decision. In order to determine the weights, a judgment matrix is constructed in this embodiment, as shown in the following formula (8):

In the above formula (8), a ₁₂ is the degree of importance of β with respect to _Th , a ₂₁ is the degree of importance of α with respect to β, and a ₁₂ =1/a ₂₁ . Let a ₁₂ =a, the calculation formula of the weighting coefficient is shown in the following formula (9):

When the service traffic is download-type traffic, the influence of bandwidth and transmission quality on policy decision is relatively large, then a ₁₂ >a ₂₁ , that is, a>1, α<β; when the service traffic is video-type traffic, when Delay, packet loss rate, and jitter have a relatively large impact on policy judgment, then a ₁₂ <a ₂₁ , that is, 0<a<1, α>β.

The value of a can be obtained based on the type of service traffic passing through the current and the past, and the calculation metric value of the PCE path is obtained as shown in the following formula (10):

On the basis of the above embodiments, and based on the PCE path calculation metric value to obtain the shortest path from the source point to the destination node, which specifically includes:

The metric value is calculated based on the obtained PCE path, and the PCE calculates the shortest path from the source point to the destination node according to the Dijkstra algorithm.

In this embodiment, based on the grid information constructed by the test data between the access routers and the port state information collected from the routers, the metric value that is differentially adjusted according to the needs of the service traffic, the PCE performs the optimal path based on the metric value. calculate.

An embodiment of the present invention further provides a network routing device based on segment routing, and the network routing method based on segment routing based on the above embodiments, as shown in FIG. 4 , includes a first module 30 and a second module 40 ,in:

The first module 30 obtains the mesh weight calculation parameter representing mesh state information based on the packet loss rate, round-trip delay and jitter of each port of the router in the segment routing to the adjacent router respectively; based on the remaining bandwidth of the link and the cyclic redundancy Check the number of CRC error packets to obtain the port weight calculation parameter representing the port status information;

The second module 40 obtains the PCE path calculation metric value of the path calculation unit based on the Mesh weight calculation parameter and the port weight calculation parameter, and obtains the shortest path from the source point to the destination node based on the PCE path calculation metric value.

FIG. 5 is a schematic diagram of an entity structure of an electronic device provided by an embodiment of the present invention. As shown in FIG. 5 , the electronic device may include: a processor (processor) 810, a communications interface (Communications Interface) 820, a memory (memory) 830, and a communication The bus 840, wherein the processor 810, the communication interface 820, and the memory 830 complete the communication with each other through the communication bus 840. The processor 810 may call a computer program stored in the memory 830 and run on the processor 810 to execute the segment routing-based network routing method provided in the above embodiments, for example, including:

Based on the packet loss rate, round-trip delay and jitter of each port of the router in segment routing to adjacent routers, the mesh weight calculation parameters representing mesh state information are obtained; based on the remaining bandwidth of the link and the CRC error of the cyclic redundancy check The number of packets obtains the port weight calculation parameter representing the port state information;

The PCE path calculation metric value of the path calculation unit is obtained based on the Mesh weight calculation parameter and the port weight calculation parameter, and the shortest path from the source point to the destination node is obtained based on the PCE path calculation metric value.

In addition, the above-mentioned logic instructions in the memory 830 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solutions of the embodiments of the present invention are essentially, or the parts that make contributions to the prior art or the parts of the technical solutions can be embodied in the form of software products, and the computer software products are stored in a storage medium , including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, Read-Only Memory (ROM, Read-Only Memory), Random Access Memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program codes .

Embodiments of the present invention further provide a non-transitory computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the computer program is implemented to execute the segment routing-based network routing method provided by the foregoing embodiments , for example including:

Based on the packet loss rate, round-trip delay and jitter of each port of the router in segment routing to adjacent routers, the mesh weight calculation parameters representing mesh state information are obtained; based on the remaining bandwidth of the link and the CRC error of the cyclic redundancy check The number of packets obtains the port weight calculation parameter representing the port state information;

The PCE path calculation metric value of the path calculation unit is obtained based on the Mesh weight calculation parameter and the port weight calculation parameter, and the shortest path from the source point to the destination node is obtained based on the PCE path calculation metric value.

An embodiment of the present invention also provides a computer program product disclosed in this embodiment, where the computer program product includes a computer program stored on a non-transitory computer-readable storage medium, the computer program includes program instructions, and when the program instructions When executed by a computer, the computer can execute the above-mentioned network routing method based on segment routing, for example, including:

Based on the packet loss rate, round-trip delay and jitter of each port of the router in segment routing to adjacent routers, the mesh weight calculation parameters representing mesh state information are obtained; based on the remaining bandwidth of the link and the CRC error of the cyclic redundancy check The number of packets obtains the port weight calculation parameter representing the port state information;

The PCE path calculation metric value of the path calculation unit is obtained based on the Mesh weight calculation parameter and the port weight calculation parameter, and the shortest path from the source point to the destination node is obtained based on the PCE path calculation metric value.

To sum up, the embodiment of the present invention provides a method and device for network routing based on segment routing, an SDN deployment architecture constructed based on data streams in different high-traffic service scenarios, and a comprehensive consideration based on the construction of test data between access routers. The grid information and the port state information collected from the router are used as the metric value to calculate the optimal path method. For the SDN deployment architecture, the present invention uses the existing network equipment as much as possible to adapt to different high-traffic services routing adjustment; based on the access router The grid information constructed from the inter-test data and the port status information collected from the routers, and the metric value that is adjusted differently according to the needs of the service traffic, and the PCE calculates the optimal path based on this metric value.

The device embodiments described above are only illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in One place, or it can be distributed over multiple network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment. Those of ordinary skill in the art can understand and implement it without creative effort.

From the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and certainly can also be implemented by hardware. Based on this understanding, the above-mentioned technical solutions can be embodied in the form of software products in essence or the parts that make contributions to the prior art, and the computer software products can be stored in computer-readable storage media, such as ROM/RAM, magnetic A disc, an optical disc, etc., includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform the methods described in various embodiments or some parts of the embodiments.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, but not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that it can still be The technical solutions described in the foregoing embodiments are modified, or some technical features thereof are equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (10)

1. A network routing method based on segmented routing is characterized by comprising the following steps:

acquiring Mesh weight calculation parameters representing grid state information based on packet loss rate, round-trip delay and jitter from each port of a router to an adjacent router in the segmented routing; acquiring port weight calculation parameters representing port state information based on link residual bandwidth and Cyclic Redundancy Check (CRC) error packet number;

and acquiring a PCE path calculation metric value of a path calculation unit based on the Mesh weight calculation parameter and the port weight calculation parameter, and acquiring the shortest path from a source point to a destination node based on the PCE path calculation metric value.

2. The method of claim 1, wherein before obtaining the Mesh weight calculation parameter characterizing the Mesh state information, the method further comprises:

and performing TraceRoute test and ping test based on different city edge router node probes to obtain the packet loss rate, round-trip delay and jitter among the direct link ports.

3. The method according to claim 1, wherein obtaining the Mesh weight calculation parameter representing the Mesh state information specifically comprises:

mesh weight calculation parameter theta representing grid state information is obtained based on packet loss rate, round-trip delay and jitter from M ports of router x to adjacent routers respectively_Mesh(pcc(m)_x) Comprises the following steps:

in the above formula, L_d(pcc(m)_x)、D_d(pcc(m)_x) And J_d(pcc(m)_x) Are respectively provided withRepresenting the packet loss rate, round trip delay and Euclidean distance of jitter to the average value of M ports of the port M of the router x; sigma_L、σ_DAnd σ_JThe packet loss rate, round trip delay and standard deviation of jitter to the average value of the M ports are respectively measured at the port M.

4. The method of claim 1, wherein before obtaining port weight calculation parameters characterizing port status information based on link residual bandwidth and Cyclic Redundancy Check (CRC) number of error packets, the method further comprises:

based on the simple network management protocol SNMP and the NetStream deployed on the existing network equipment, the residual bandwidth of the link and the number of CRC error packets received by the set time port are collected.

5. A method for routing a network based on segmented routing according to claim 3, characterized in that the port weight calculation parameter θ_Port(pcc(m)_x) Comprises the following steps:

in the above formula, RBW_d(pcc(m)_x)、CRC_d(pcc(m)_x) Respectively representing the Euclidean distance, σ, of port M of router x in the residual bandwidth and the CRC error packet number for the average value of M ports_RBW、σ_CRCThe standard deviation of the residual bandwidth and the CRC error packet number of the port M to the average value of the M ports respectively.

6. The method for selecting a route based on a segment routing according to claim 5, wherein obtaining a path computation metric of a Path Computation Element (PCE) based on the Mesh weight computation parameter and the port weight computation parameter specifically comprises:

mesh weight calculation parameter theta of x port m of router_Mesh(pcc(m)_x) And port weight calculation parameter θ_Port(pcc(m)_x) And theta_Mesh(pcc(m)_x) Right of (1)Values α and θ_Port(pcc(m)_x) The weight β, a weighting function of the PCE path calculation metric value of the mth port of the router x is constructed;

and acquiring values of the weight α and the weight β based on the service traffic type to obtain a weighting coefficient of the PCE path computation metric value, and obtaining the PCE path computation metric value based on the weighting coefficient.

7. The method according to claim 1, wherein the obtaining the shortest path from the source node to the destination node based on the PCE path computation metric value comprises:

and based on the obtained PCE path calculation metric value, the PCE calculates the shortest path from the source point to the destination node according to the Dijkstra algorithm.

8. A network routing device based on segment routing, comprising:

the first module is used for acquiring Mesh weight calculation parameters representing grid state information based on the packet loss rate, round-trip delay and jitter from each port of a router in the segmented routing to an adjacent router; acquiring port weight calculation parameters representing port state information based on link residual bandwidth and Cyclic Redundancy Check (CRC) error packet number;

and the second module is used for obtaining a PCE path calculation metric value of the path calculation unit based on the Mesh weight calculation parameter and the port weight calculation parameter, and obtaining the shortest path from a source node to a destination node based on the PCE path calculation metric value.

9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method according to any of claims 1 to 7 are implemented when the processor executes the program.

10. A non-transitory computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.

Images