WO2022166348A1 - Routing method, routing apparatus, controller and computer-readable storage medium - Google Patents

Abstract

A routing method, a routing apparatus, a controller and a computer-readable storage medium, the method comprising: receiving a routing computation request, which carries first service information, of a plurality of first services; for each first service, determining, according to the first service information, a target network slice from a plurality of network slices, wherein the plurality of network slices have the same physical structure and are obtained by division of a physical network structure of a forwarding plane; according to slice topology information of the target network slice which corresponds to the first service and according to the first service information, determining first forwarding path information, which is on the target network slice, of the first service; and sending the first forwarding path information to a forwarding plane device, such that the forwarding plane device forwards the first service.

Description

Routing method, routing apparatus, controller, and computer-readable storage medium

CROSS-REFERENCE TO RELATED APPLICATIONS

This application is based on the Chinese patent application with the application number of 202110153959.9 and the filing date of February 4, 2021, and claims the priority of the Chinese patent application. The entire content of the Chinese patent application is incorporated herein by reference.

technical field

The embodiments of the present application relate to, but are not limited to, the field of communications technologies, and in particular, relate to a routing method, a routing device, a controller, and a computer-readable storage medium.

Background technique

In traditional technologies, in order to make full use of network resources for traffic scheduling, queue management and scheduling are usually adopted. In SDN (Software Defined Network), queue-based scheduling technology is widely used to ensure high QoS (Quality of Service). However, the above queue-based resource allocation or scheduling strategy has the following disadvantages: since it needs to calculate the feasible route of the service by calculating the remaining available bandwidth on the network link in real time, the time to calculate the feasible route is relatively long, and the requests in the queue need to be In order to process, that is, the traditional technology can only perform routing calculation in a serial manner. Therefore, the traditional queue-based resource allocation or scheduling strategy seriously affects routing efficiency and reduces user service quality.

SUMMARY OF THE INVENTION

The following is an overview of the topics detailed in this article. This summary is not intended to limit the scope of protection of the claims.

Embodiments of the present application provide a routing method, a routing device, a controller, and a computer-readable storage medium.

In a first aspect, an embodiment of the present application provides a routing method, including: receiving a plurality of routing calculation requests for first services, the routing calculation requests carrying first service information; for each of the first services, according to The first service information determines the target network slice corresponding to the first service from multiple network slices, wherein the multiple network slices have the same physical structure and are obtained by dividing the physical network structure of the forwarding plane; the slice topology information and the first service information of the target network slice corresponding to the first service, and determine the first forwarding path information of the first service on the target network slice; The information is sent to the forwarding plane device, so that the forwarding plane device forwards the first service according to the first forwarding path information.

In a second aspect, an embodiment of the present application further provides a routing device, comprising: a request receiving unit configured to receive a plurality of routing calculation requests of the first service, the routing calculation requests carrying first service information; service deployment a unit configured to, for each first service, determine a target network slice corresponding to the first service from multiple network slices according to the first service information, wherein the multiple network slices have the same The physical structure is obtained by dividing the physical network structure of the forwarding plane; the path determination unit is configured to determine the first service according to the slice topology information of the target network slice corresponding to the first service and the first service information the first forwarding path information of the service on the target network slice; the path sending unit is configured to send the first forwarding path information to the forwarding plane device, so that the forwarding plane device forwards according to the first The path information forwards the first service.

In a third aspect, an embodiment of the present application further provides a controller, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor executing the computer The routing method as described in the first aspect above is implemented in the program.

In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium storing computer-executable instructions, where the computer-executable instructions are used to execute the routing method described in the first aspect.

Other features and advantages of the present application will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the present application. The objectives and other advantages of the application may be realized and attained by the structure particularly pointed out in the description, claims and drawings.

Description of drawings

The accompanying drawings are used to provide a further understanding of the technical solutions of the present application, and constitute a part of the specification. They are used to explain the technical solutions of the present application together with the embodiments of the present application, and do not constitute a limitation on the technical solutions of the present application.

1 is a schematic diagram of a system architecture platform for executing a routing method provided by an embodiment of the present application;

2 is a flowchart of a routing method provided by an embodiment of the present application;

FIG. 3 is a flowchart of a service slice allocation method based on reinforcement learning in a routing method provided by an embodiment of the present application;

4 is a schematic diagram of an interaction flow between an agent and an environment in reinforcement learning provided by an embodiment of the present application;

5 is a flowchart of a method for dynamically adjusting bandwidth resources between network slices in a routing method provided by an embodiment of the present application;

6 is a flowchart of routing compensation for deployment failure services in a routing method provided by an embodiment of the present application;

7 is a flowchart of an overall routing method provided by an embodiment of the present application;

8 is a schematic diagram of dividing the physical network topology of the forwarding plane into multiple network slices according to an embodiment of the present application;

9 is a schematic diagram of the bandwidth resource status of each network slice provided by an embodiment of the present application after the first three services are deployed;

10 is a schematic diagram of the bandwidth resource status of each network slice after dynamically adjusting the bandwidth resources between network slices according to an embodiment of the present application;

FIG. 11 is a schematic diagram of a routing device provided by an embodiment of the present application.

Detailed ways

In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application will be described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application.

It should be noted that although the functional modules are divided in the schematic diagram of the device, and the logical sequence is shown in the flowchart, in some cases, the modules may be divided differently from the device, or executed in the order in the flowchart. steps shown or described. The terms "first", "second" and the like in the specification, claims or the above drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

With the rapid development of network technologies, such as SDN and NFV (Network Functions Virtualization), how to effectively utilize limited network bandwidth resources and improve service routing efficiency while ensuring network quality has become a key research topic for various equipment manufacturers. The problem. In the SDN scenario, the controller can collect global topology information. When there are batches of services that need to be deployed, in some examples, when the user's new services are activated in batches or the network is turbulent and the affected services need to be rerouted, It is necessary to decide how to deploy these services and ensure the QoS of these services.

In traditional technologies, in order to make full use of network resources for traffic scheduling, queue management and scheduling are usually adopted. Typically, packets in a queue may have different priorities, and in some examples, earlier packets may have higher priority than other packets. In Floodlight-based SDN networks, queue-based scheduling techniques are widely used to ensure high QoS. Among them, the traffic to be transmitted is divided into QoS flow and optimal flow, and is allocated to different queues according to the priority. However, the above queue-based resource allocation or scheduling strategy has the following disadvantages: since it needs to calculate the feasible route of the service by calculating the remaining available bandwidth on the network link in real time, the time to calculate the feasible route is relatively long, and the requests in the queue need to be In order to process, that is, the traditional technology can only perform routing calculation in a serial manner. Therefore, the traditional queue-based resource allocation or scheduling strategy seriously affects routing efficiency and reduces user service quality.

Based on the above situation, the embodiments of the present application provide a routing method, a routing device, a controller, and a computer-readable storage medium. The routing method includes the following steps: receiving a plurality of routing calculation requests of the first service, and the routing calculation request carries a First service information; for each first service, determine a target network slice corresponding to the first service from multiple network slices according to the first service information, wherein the multiple network slices have the same physical structure and are determined by the physical structure of the forwarding plane. The network structure is divided and obtained; the first forwarding path information of the first service on the target network slice is determined according to the slice topology information and the first service information of the target network slice corresponding to the first service; the first forwarding path information is sent to the forwarding plane The device enables the forwarding plane device to forward the first service according to the first forwarding path information. According to the technical solutions of the embodiments of the present application, in the embodiments of the present application, the network bandwidth of the physical network structure of the forwarding plane is sliced horizontally, and the bandwidth resources of the network slices are isolated from each other, so that the forwarding paths of each service can be calculated in parallel. Therefore, The embodiments of the present application can improve the network bandwidth resource utilization rate and ensure the service quality of the service, and can also improve the success rate and time-consuming performance of service routing.

The embodiments of the present application will be further described below with reference to the accompanying drawings.

As shown in FIG. 1 , FIG. 1 is a schematic diagram of a system architecture platform 100 for executing a routing method provided by an embodiment of the present application.

In the example of FIG. 1 , the system architecture platform 100 is provided with a processor 110 and a memory 120 , wherein the processor 110 and the memory 120 may be connected by a bus or in other ways. In FIG. 1 , the connection by a bus is taken as an example.

As a non-transitory computer-readable storage medium, the memory 120 can be used to store non-transitory software programs and non-transitory computer-executable programs. Additionally, memory 120 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, memory 120 may include memory located remotely from processor 110, which may be connected to the system architecture platform through a network. Examples of such networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

Those skilled in the art can understand that the system architecture platform can be applied to a 3G communication network system, an LTE communication network system, a 5G communication network system, and a subsequently evolved mobile communication network system, etc., which is not specifically limited in this embodiment.

It can be understood by those skilled in the art that the system architecture platform shown in FIG. 1 does not constitute a limitation on the embodiments of the present application, and may include more or less components than those shown in the figure, or combine some components, or different Component placement.

In the system architecture platform shown in FIG. 1 , the processor 110 can call the routing program stored in the memory 120 to execute the routing method.

Based on the above-mentioned system architecture platform, various embodiments of the routing method of the present application are proposed below.

As shown in FIG. 2 , FIG. 2 is a flowchart of a routing method provided by an embodiment of the present application. The routing method includes but is not limited to step S100 , step S200 , step S300 and step S400 .

Step S100, receiving a plurality of routing calculation requests of the first service, where the routing calculation requests carry the first service information;

Step S200, for each first service, determine a target network slice corresponding to the first service from multiple network slices according to the first service information, wherein the multiple network slices have the same physical structure and are determined by the physical network structure of the forwarding plane. divided;

Step S300, according to the slice topology information and the first service information of the target network slice corresponding to the first service, determine the first forwarding path information of the first service on the target network slice;

Step S400: Send the first forwarding path information to the forwarding plane device, so that the forwarding plane device forwards the first service according to the first forwarding path information.

According to the technical solutions of the embodiments of the present application, in the embodiments of the present application, the network bandwidth of the physical network structure of the forwarding plane is sliced horizontally, and the bandwidth resources of the network slices are isolated from each other, so that the forwarding paths of each service can be calculated in parallel. Therefore, The embodiments of the present application can improve the network bandwidth resource utilization rate and ensure the service quality of the service, and can also improve the success rate and time-consuming performance of service routing.

It should be noted that the above steps S100, S200, S300 and S400 may all be executed by the SDN controller.

It can be understood that the bandwidths of the above-mentioned multiple network slices may be set uniformly or randomly.

In addition, it can be understood that the slice topology information about the above-mentioned target network slice may include, but is not limited to, node information, link bandwidth information, and link inherent delay information.

In addition, it can be understood that the above-mentioned first service information may include, but is not limited to, source point information, sink point information, and service bandwidth information.

It is worth noting that the above network slicing refers to dividing a physical network into multiple virtual end-to-end networks. The network slicing in this embodiment of the present application is a horizontal slicing based on network bandwidth resources, that is, each network slice has The same physical structure, but the bandwidth resources are independent of each other, which is different from the traditional vertical slicing.

For the above steps S100 and S200, when it is necessary to perform routing calculation on a batch of multiple first services, this embodiment of the present application can use the slice structure to first perform slice attribution allocation for the batch of multiple first services, that is, first determine each Which network slice a first service should be deployed to. Due to the scarcity of network slicing bandwidth resources, some first services cannot be transmitted, which will affect the transmission success rate of the first service. To solve this problem, based on the above network slicing architecture, the embodiments of the present application provide the following: The collaborative method in FIG. 3 and FIG. 5 , wherein FIG. 3 is a flowchart of the reinforcement learning-based service slice allocation method in the routing method provided by an embodiment of the present application, so as to determine which network slice each first service should be allocated to Deploy on the above; FIG. 5 is a flowchart of a method for dynamically adjusting bandwidth resources between network slices in a routing method provided by an embodiment of the present application. By combining the above two coordination methods, the embodiment of the present application can ensure that the route calculation can be accelerated without sacrificing the deployment success rate and QoS.

As shown in FIG. 3 , FIG. 3 is a flowchart of a method for allocating service slices based on reinforcement learning in a routing method provided by an embodiment of the present application. Regarding the above step S200 , determining from multiple network slices according to the first service information The target network slice corresponding to the first service includes but is not limited to step S500.

Step S500: Input the slice topology information and the first service information of the multiple network slices into the service deployment model to obtain the target network slice corresponding to the first service, wherein the service deployment model is composed of the slice topology information of the multiple network slices and the multiple network slices. A second service information is obtained through training.

Specifically, before allocating network slices to the first service, the embodiment of the present application also needs to obtain the required service deployment model through machine learning training, wherein RL (Reinforcement Learning, reinforcement learning) is a branch of machine learning, mainly It is to solve how to take actions according to the environment to maximize the cumulative expected return; the reinforcement learning model includes the environmental state set State, the action set A, the state transition probability and the immediate return after the state transition. When the agent in reinforcement learning makes a decision , it can obtain rewards from the environment and learn a strategy to achieve a specific goal by continuously interacting with the environment. Based on this, in order to determine the network slice to be deployed for the first service according to the slice topology information of each network slice and the first service information, the embodiment of the present application needs to obtain a network slice equivalent to the first service before allocating the network slice. The trained business deployment model of the above agent.

It can be understood that the above reinforcement learning models can be, but are not limited to, DQN (Deep Q-Network) model, DDPG (Deep Deterministic Policy Gradient, deep deterministic policy gradient) model, PPO (Proximal Policy Optimization , proximal policy optimization) model, A3C (Asynchronous Advantage Actor-Critic, asynchronous advantage action evaluation) model, etc., among which, the interaction process between the agent and the environment in reinforcement learning can be seen in Figure 4.

Specifically, this embodiment of the present application can model the network fragmentation problem as a Markov decision problem, where the environmental state set State, the action set A, the reward function, and the state transition probability in the model are respectively as follows:

The state in the above model specifically refers to the information observed by the agent. The information observed by the agent includes two parts, one part is the slice topology information of all network slices, and the other part is the service information of the service to be deployed, namely State={O _s , _Of }, where O _s represents the _adjacency matrix of the available bandwidth of each network slice, and Of represents the source, sink and required bandwidth of the service to be deployed.

In addition, regarding the action set A in the above model, it specifically refers to the slice index number to which each service to be deployed goes. In some examples, if there are k network slices, then the action set A={1,2,..., k}.

In addition, regarding the reward function in the above model, it can be:

in

Indicates whether service i can be deployed successfully when deployed to network slice j according to the model prediction result. If successful, the value is 1; It reflects the correctness of the agent's decision.

In addition, regarding the state transition probability in the above model, the transition probability is deterministic, which depends on the deployment mode of the service, including the routing scheme and the resource dynamic adjustment strategy between bandwidths.

When the bandwidth resources of the network slicing can meet the needs of the service to be deployed, it means that the agent has made a correct decision, and the agent will be given a positive reward; otherwise, a wrong decision will always produce a negative reward.

As shown in FIG. 5 , FIG. 5 is a flowchart of a method for dynamically adjusting bandwidth resources between network slices in a routing method provided by an embodiment of the present application. When the above-mentioned multiple network slices include a first network slice and a second network slice Slicing, the first network slice includes a first link, and the second network slice includes a second link corresponding to the first link. The routing method in this embodiment of the present application includes, but is not limited to, step S600.

Step S600, when the bandwidth of the first link is less than the bandwidth of all the first services to be routed through the first link, transfer the idle bandwidth of the second link to the first link, so that the bandwidth of the first link is greater than that of the first link. or equal to the bandwidth of all first services to be routed through the first link.

According to the technical solutions of the embodiments of the present application, since the network bandwidth and the characteristics of the new services to be deployed are constantly changing, the network bandwidth in each network slice is gradually distributed unevenly. In this regard, the embodiment of the present application can reasonably utilize the idle bandwidth in each network slice according to the current network bandwidth and service characteristics and according to a certain adjustment strategy, thereby improving the utilization rate of the network bandwidth and further improving the throughput of service deployment. .

Specifically, the embodiments of the present application can enable links that are often transmitted in network slices to obtain higher bandwidth, so that network slices can dynamically adapt to changes in scenarios. Among them, for network slicing, the embodiments of this application include but are not limited to the following four operations: the first operation is to accept bandwidth supplements, which specifically means that the second network slice transfers excess bandwidth resources to the current first network slice, and the current The first network slice of the network slice expands the bandwidth of the links involved in the transfer; the second operation is routing calculation, which specifically refers to the routing calculation of the services to be allocated to the current first network slice according to its routing strategy; the third operation is routing calculation. The operation is to respond to the application for bandwidth expansion of the first network slice, specifically referring to the resource transfer request sent by the current first network slice, judging whether the transfer conditions are met, and if so, transferring the excess bandwidth of the second network slice to the current first network slice. A network slice; the fourth operation is to send a bandwidth request request. After the above routing calculation is completed, the current network slice will deduct the bandwidth resources of the service whose route has ended. Due to the reduction of bandwidth resources on some links, it can be Other slices send bandwidth expansion applications, wherein the size of the expansion resources currently applied for by the network slice may be the service bandwidth value.

It is worth noting that for the above four operations of a single network slice, the routing calculation and the other three operations are serial, that is, the routing calculation will only be performed when the network bandwidth resource status is fixed.

As shown in FIG. 6 , FIG. 6 is a flowchart of performing routing compensation for a deployment failure service in a routing method provided by an embodiment of the present application. When the first service includes a deployment failure service for which the first forwarding path information calculation fails, The routing method in this embodiment of the present application includes, but is not limited to, steps S710 and S720.

Step S710, according to the first service information of the service that fails to deploy, calculate the second forwarding path information of the service that fails to deploy on the physical network structure of the forwarding plane;

Step S720: Send the second forwarding path information to the forwarding plane device, so that the forwarding plane device forwards the deployment failure service according to the second forwarding path information.

Specifically, after the route calculation in the above step S300, it is not excluded that there are services that fail in route calculation, wherein the reasons for the failure of route calculation include: bandwidth dilution after dividing multiple network slices, so that the bandwidth of each network slice is insufficient Support big business. In this regard, the embodiment of the present application uses the original physical network structure to perform routing compensation for services that fail to be deployed, so as to further ensure that all services have reachable paths, thereby ensuring the success rate of routing services.

Based on the method steps of FIG. 2 , FIG. 3 , FIG. 5 , and FIG. 6 , an embodiment of the present application provides a flowchart of an overall routing method, as shown in FIG. 7 . The overall routing method in FIG. 7 includes but is not limited to steps S801, S802, S803, S804, S805, S806, S807, S808, S809, and S810.

Step S801, the controller collects network topology information and second service information;

Specifically, the SDN controller collects various information, where the information includes network topology information on the forwarding plane and information about the second service to be forwarded. Specifically, the forwarding plane device will report the above information to the SDN controller through a protocol.

It can be understood that the above-mentioned network topology information includes but is not limited to node information, link bandwidth information and link inherent delay information, where node information refers to information of routers, switches or other network devices.

In addition, it can be understood that the above-mentioned second service information includes but is not limited to information such as source point information, sink point information, and service bandwidth size.

Step S802, dividing the physical network structure of the forwarding plane into multiple network slices having the same physical structure according to the collected network topology information;

Specifically, according to the network topology information collected in the foregoing step S801, the embodiment of the present application will initialize the network bandwidth resources, thereby forming a plurality of mutually isolated network slices.

It should be noted that the plurality of mutually isolated network slices formed by the embodiments of the present application can provide an interactive environment for the deep reinforcement learning algorithm in the subsequent steps.

In addition, it can be understood that the manner of dividing the network slice in this embodiment of the present application may be, but not limited to, uniform bandwidth division or random division.

Step S803, inputting slice topology information of multiple network slices and multiple second service information into a service deployment model to obtain a trained service deployment model;

According to the technical solutions of the embodiments of the present application, the embodiments of the present application will perform deep reinforcement learning algorithm model training. Specifically, the embodiments of the present application will simulate a real network slicing scene, and interact with the network slicing environment in the above step S802 to make the service The deployment model can learn a service deployment strategy, wherein the service deployment model in the embodiment of the present application is equivalent to the above-mentioned agent. Specifically, the input of the service deployment model is the topology information of multiple network slices and the second service information of a batch of second services to be deployed, and the output is the slice attribution index of the batch of second services. Know the network slice to which each second service belongs. Through a certain number of interactions with the environment, the business deployment model can finally learn a better allocation strategy, so as to obtain a trained business deployment model to realize the deployment of new services.

It can be understood that the slice topology information about the above-mentioned multiple network slices specifically refers to an adjacency matrix of a topology graph, wherein each element in the matrix is a link bandwidth value between two nodes. In addition, it can be understood that the above-mentioned second service information includes but is not limited to information such as source point information, sink point information, and service bandwidth size.

Step S804, receiving a plurality of routing calculation requests of the first service, where the routing calculation requests carry the first service information;

Specifically, after the division of network slices and the training of the service deployment model are completed, the embodiment of the present application can implement the deployment of new services. When the embodiment of the present application needs to perform batch deployment of multiple new first services, the SDN controller will receive batch routing calculation requests of multiple first services, where the routing calculation request carries first service information.

It can be understood that the above-mentioned first service information includes but is not limited to information such as source point information, sink point information, and service bandwidth size.

Step S805, calling the trained service deployment model, and inputting slice topology information of multiple network slices and multiple first service information into the service deployment model to obtain the slice attribution index of each first service;

Specifically, after the SDN controller receives the batch routing calculation requests of multiple first services, in order to better and more intelligently deploy the multiple first services, this embodiment of the present application will call the training data obtained in the above step S803. The service deployment model is based on the service deployment model, and the first service information of the first service to be deployed and the slice topology information in each current network slice are used as input, and then the service deployment model will output the slice attribution index corresponding to each first service. The slice attribution index can determine the target network slice to be deployed for each first service.

Step S806, dynamically adjusting bandwidth resources between network slices;

Specifically, since the network bandwidth and the characteristics of the incoming services are constantly changing, the network bandwidth in each network slice will gradually be unevenly distributed. According to the current network bandwidth and service characteristics, according to a certain adjustment strategy, the idle bandwidth in each network slice is rationally utilized, thereby improving the utilization rate of the network bandwidth, thereby improving the throughput of service deployment.

In some examples, the above-mentioned multiple network slices include a first network slice and a second network slice, the first network slice includes a first link, and the second network slice includes a second link corresponding to the first link, when The bandwidth of the first link is smaller than the bandwidth of all the first services to be routed through the first link, and the idle bandwidth of the second link is transferred to the first link, so that the bandwidth of the first link is greater than or equal to the bandwidth to be routed Bandwidth of all first traffic over the first link.

Step S807, calculating the first forwarding path information of each first service on the target network slice in parallel;

Specifically, according to the slice attribution index corresponding to each first service obtained in the above step S805, the target network slice to be deployed for each first service can be determined. After dynamically adjusting the bandwidth resources between the network slices, this The application embodiment will calculate in parallel the first forwarding path information of each first service on the corresponding target network slice according to the slice topology information and the first service information of the target network slice corresponding to the first service. The first forwarding path information is sent to the forwarding plane device, so that the forwarding plane device forwards the first service according to the first forwarding path information. The above-mentioned parallel computing method can significantly reduce the routing time, and at the same time, the bandwidth resources between the various network slices will be continuously adjusted to each other according to the strategy in the above step S806, thereby maximizing the bandwidth resource utilization of the network.

Step S808, determine whether there is a first service for which the route calculation fails, and when there is a first service for which the route calculation fails, execute step S809; otherwise, execute step S810;

Specifically, dividing multiple network slices will lead to bandwidth dilution, so that the bandwidth of each network slice is insufficient to support large services. Therefore, when there is a first service that requires a large bandwidth, route calculation may fail. .

Step S809, performing routing supplementary calculation on the first service for which the routing calculation fails;

Specifically, when there is a first service for which route calculation fails, the embodiment of the present application uses the original physical network structure to perform route compensation for the first service for which route calculation fails to further ensure that all services have reachable paths, thereby enabling Guarantees the success rate of service routing.

Step S810, end and output the route calculation result.

Based on the method steps of the above figures, the present application provides a specific embodiment to describe the above method steps in detail, as shown in FIG. 8 to FIG. 10 , wherein FIG. Figure 9 is a schematic diagram of the bandwidth resource status of each network slice provided by an embodiment of the present application after the first three services are deployed, and Figure 10 is a schematic diagram provided by an embodiment of the present application. A schematic diagram of the bandwidth resource status of each network slice after dynamically adjusting the bandwidth resources between network slices.

As shown in Figure 8, the original physical network topology T0 of the forwarding plane includes four nodes (A, B, C, D) and four links between nodes, wherein each link corresponds to the current bandwidth of the link , weight, assuming that the business calculates the route according to the minimum weight strategy. In some examples, as shown in FIG. 8 , the current bandwidth of the link between node A and node B in the original physical network topology T0 is 30, and the weight is 1; node A and node B in the original physical network topology T0 The current bandwidth of the link between C is 30, and the weight is 1; the current bandwidth of the link between node B and node D in the original physical network topology T0 is 16, and the weight is 3; the original physical network topology T0 The current bandwidth of the link between node C and node D is 30, and the weight is 1.

The original physical network topology T0 in Figure 8 can be divided into three network slices (T1, T2, T3), where the current bandwidth of the link between node A and node B in network slice T1 is 12 , the weight is 1; the current bandwidth of the link between node A and node C in network slice T1 is 12, and the weight is 1; the current bandwidth of the link between node B and node D in network slice T1 is 5, and the weight is 3; the current bandwidth of the link between node C and node D in the network slice T1 is 12, and the weight is 1. The current bandwidth of the link between node A and node B in network slice T2 is 10, and the weight is 1; the current bandwidth of the link between node A and node C in network slice T2 is 10, and the weight is 1 1; the current bandwidth of the link between node B and node D in network slice T2 is 3, and the weight is 3; the current bandwidth of the link between node C and node D in network slice T2 is 10, and the weight is 3. The value is 1. The current bandwidth of the link between node A and node B in network slice T3 is 8, and the weight is 1; the current bandwidth of the link between node A and node C in network slice T3 is 8, and the weight is 1 1; the current bandwidth of the link between node B and node D in network slice T3 is 8, and the weight is 3; the current bandwidth of the link between node C and node D in network slice T3 is 8, and the weight is 8. The value is 1.

According to the original physical network topology T0, network slice T1, network slice T2 and network slice T3 in FIG. 8, it can be seen that the sum of link bandwidths corresponding to all network slices is equal to the link bandwidth in the original physical network topology. In some examples, the link bandwidth between node A and node B in the original physical network topology T0 is equal to the link bandwidth between node A and node B in network slice T1 and between node A and node B in network slice T2 and the sum of the link bandwidth between node A and node B in network slice T3.

Based on the slice division result in Figure 8 above, when four services need to be deployed, the embodiment of the present application will call the trained service deployment model to predict the deployment of the four services. It is assumed that a number "1" is used to mark a service in the network slice. The deployment in T1 is successful, the number "2" marks the successful deployment of a service in network slice T2, and the number "3" marks the successful deployment of a service in network slice T3, when there are four services in network slices T1, T2, T3, T2 is deployed, and the output of the business deployment model is {1,2,3,2}.

Among them, according to the results of the business deployment model, the deployment of the current business is obtained. If the deployment of the first three businesses is shown in Table 1:

Business ID	Source	place to stay	bandwidth	Whether the service was successfully deployed before bandwidth adjustment
Business 1	B	D	10	Successfully deployed to network slice T1
Business 2	B	D	10	Successfully deployed to network slice T2
Business 3	B	D	5	Successfully deployed to network slice T3

Table 1

As can be seen from the above Table 1, the source point of service 1 is node B, the sink point is node D, the bandwidth is 10, the current slice is network slice T1 and can be successfully allocated to network slice T1, and service 1 is in network slice T1. The forwarding path is from B through A and C to D in turn; the source point of service 2 is node B, the sink point is node D, the bandwidth is 10, the current home slice is network slice T2 and can be successfully allocated to network slice T2, The forwarding path of service 2 in network slice T2 is from B through A and C to D in sequence; the source point of service 3 is node B, the sink point is node D, the bandwidth is 5, and the current home slice is network slice T3 and can be It is successfully allocated to network slice T3, and the forwarding path of service 3 in network slice T3 is from B to D.

After the deployment of Service 1, Service 2 and Service 3 is completed, the remaining bandwidth information in each network slice can be referred to as shown in Figure 9. For network slice T1, the link bandwidth between Node A and Node B is changed from the original 12 Reduced to 2, the link bandwidth between node A and node C is reduced from the original 12 to 2, and the link bandwidth between node C and node D is reduced from the original 12 to 2; in addition, for network slice T2 , the link bandwidth between node A and node B is reduced from the original 10 to 0, the link bandwidth between node A and node C is reduced from the original 10 to 0, the link bandwidth between node C and node D The path bandwidth is reduced from the original 10 to 0; in addition, for the network slice T3, the link bandwidth between the node B and the node D is reduced from the original 8 to 3.

In addition, for the fourth business, its deployment situation is shown in Table 2:

Business ID	Source	place to stay	bandwidth	Whether the service was successfully deployed before bandwidth adjustment
Business 4	B	D	4	Deployment to network slice T2 fails

Table 2

It can be seen from the above Table 2 that the source point of service 4 is node B, the sink point is node D, the bandwidth is 4, the current home slice is network slice T2, and the forwarding path is from B through A and C to D in sequence; 1. After the deployment of services 2 and 3, the link bandwidth between node A and node B, the link bandwidth between node A and node C, and the link between node C and node D in network slice T2 The bandwidth is reduced to 0, and the bandwidth required by service 4 is 4. Therefore, in network slice T2, the link bandwidth between node A and node B, the link bandwidth between node A and node C, the link bandwidth between node C and node C, and The link bandwidth between nodes D cannot meet the bandwidth value required by service 4. Therefore, service 4 will fail to be deployed on network slice T2.

In this regard, the embodiment of the present application will dynamically adjust the bandwidth resources between network slices, because the link bandwidth between node A and node B in network slice T3 after the deployment of service 1, service 2 and service 3 is completed, The link bandwidth between node C and the link bandwidth between node C and node D are all greater than the bandwidth value required by service 4. Therefore, in this embodiment of the present application, the network slice T3 is directly connected from B through A and C to the network slice T3. Part of the bandwidth value of path D is allocated to network slice T2, so that the adjusted bandwidth of network slice T2 can meet the bandwidth requirement of service 4. The bandwidth resource status of each network slice after dynamically adjusting the bandwidth resources between network slices can refer to As shown in FIG. 10 , therefore, the embodiment of the present application can effectively avoid a situation where a large number of idle bandwidth resources exist in a certain network slice, so that service 4 can be successfully transmitted. Specifically, the deployment of service 1, service 2, service 3 and service 4 before and after dynamic adjustment is shown in Table 3:

table 3

According to the technical solutions of the foregoing embodiments, the embodiments of the present application can perform routing computations in a parallel manner based on the slice architecture, which significantly improves routing efficiency; the network resource allocation and service deployment are organically combined to obtain better network resource allocation and Business deployment strategy. Secondly, the embodiment of the present application can perform the training of the deep reinforcement learning algorithm model in an offline manner. In practical applications, it is only necessary to read the stored service deployment model, which takes a short time. In addition, the embodiment of the present application uses a network slicing architecture, combined with a deep reinforcement learning algorithm, and performs rerouting in a parallel manner, which can improve the time-consuming performance of routing several times.

Based on the routing methods of the foregoing embodiments, various embodiments of the routing apparatus of the present application are proposed below.

As shown in FIG. 11 , FIG. 11 is a schematic diagram of a routing apparatus 200 provided by an embodiment of the present application. The routing apparatus 200 includes, but is not limited to, a request receiving unit 210 , a service deploying unit 220 , a path determining unit 230 and a path sending unit 240 .

Specifically, the request receiving unit 210 is configured to receive a plurality of routing calculation requests of the first service, and the routing calculation requests carry the first service information; the service deployment unit 220 is configured to, for each first service, according to the first service information The target network slice corresponding to the first service is determined from the multiple network slices, wherein the multiple network slices have the same physical structure and are obtained by dividing the physical network structure of the forwarding plane; the path determination unit 230 is configured to correspond to the first service according to the The slice topology information and the first service information of the target network slice, and determine the first forwarding path information of the first service on the target network slice; the path sending unit 240 is configured to send the first forwarding path information to the forwarding plane device so that the The forwarding plane device forwards the first service according to the first forwarding path information.

In addition, regarding the above-mentioned service deployment unit 220, it is also configured to input the slice topology information of multiple network slices and the first service information into the service deployment model, so as to obtain the target network slice corresponding to the first service, wherein the service deployment model It is obtained by training the slice topology information of multiple network slices and multiple second service information.

Referring to FIG. 11 , regarding the above-mentioned routing device 200, when multiple network slices include a first network slice and a second network slice, the first network slice includes a first link, and the second network slice includes a link with the first network slice For the corresponding second link, the routing device 200 further includes, but is not limited to, a dynamic adjustment unit 250 .

Specifically, the dynamic adjustment unit 250 is configured to transfer the idle bandwidth of the second link to the first link when the bandwidth of the first link is smaller than the bandwidth of all the first services to be routed through the first link, so that the The bandwidth of the first link is greater than or equal to the bandwidth of all the first services to be routed through the first link.

Referring to FIG. 11 , regarding the above-mentioned routing device 200, when the first service includes a deployment failure service in which the calculation of the first forwarding path information fails, the routing device 200 further includes, but is not limited to, a service compensating unit 260 with a failure and a failure Service path sending unit 270 .

Specifically, the failed service compensation unit 260 is configured to calculate the second forwarding path information of the failed service deployment on the physical network structure of the forwarding plane according to the first service information of the failed service deployment; the failed service path sending unit 270 is configured to The second forwarding path information is successfully sent to the forwarding plane device, so that the forwarding plane device forwards the deployment failure service according to the second forwarding path information.

It should be noted that, for the technical effects of the routing device in the embodiments of the present application, reference may be made to the above-mentioned routing method embodiments.

Based on the above routing method, various embodiments of the controller and the computer-readable storage medium of the present application are respectively proposed below.

In addition, one embodiment of the present application provides a controller including: a memory, a processor, and a computer program stored on the memory and executable on the processor.

The processor and memory may be connected by a bus or otherwise.

It should be noted that the controller in this embodiment may correspond to including the memory and the processor in the embodiment shown in FIG. 1 , and can constitute a part of the system architecture platform in the embodiment shown in FIG. 1 , and the two belong to The same inventive concept, therefore, both have the same realization principle and beneficial effects, and will not be described in detail here.

Specifically, the controller in this embodiment of the present application may be an SDN controller.

The non-transitory software programs and instructions required to implement the routing method of the above-mentioned embodiment or the routing method are stored in the memory, and when executed by the processor, the routing method of the above-mentioned embodiment is executed. Method steps S100 to S400 , method step S500 in FIG. 3 , method step S600 in FIG. 5 , method steps S710 to S720 in FIG. 6 , method steps S801 to S810 in FIG. 7 .

The apparatus embodiments described above are only illustrative, and the units described as separate components may or may not be physically separated, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment.

In addition, an embodiment of the present application further provides a computer-readable storage medium, where the computer-readable storage medium stores computer-executable instructions, when the computer-executable instructions are used to execute the above routing method, for example, to execute the above description The method steps S100 to S400 in FIG. 2 , the method step S500 in FIG. 3 , the method step S600 in FIG. 5 , the method steps S710 to S720 in FIG. 6 , and the method steps S801 to S810 in FIG. 7 .

The embodiments of the present application include: the controller receives a plurality of route calculation requests of the first service, wherein the route calculation requests carry the first service information; then the controller will, for each first service, select from the plurality of first service information according to the first service information. The target network slice corresponding to the first service is determined from the network slices, wherein multiple network slices have the same physical structure and are obtained by dividing the physical network structure of the forwarding plane; then, the controller will determine the target network slice corresponding to the first service according to the target network slice of the first service. The slicing topology information and the first service information are obtained, and determine the first forwarding path information of the first service on the target network slice; finally, the controller will send the first forwarding path information to the forwarding plane device to make the forwarding plane device follow the first The forwarding path information forwards the first service. According to the technical solutions of the embodiments of the present application, in the embodiments of the present application, the network bandwidth of the physical network structure of the forwarding plane is sliced horizontally, and the bandwidth resources of the network slices are isolated from each other, so that the forwarding paths of each service can be calculated in parallel. Therefore, The embodiments of the present application can improve the utilization rate of network bandwidth resources and ensure the service quality of services, and can also improve the success rate and time-consuming performance of service routing.

Those of ordinary skill in the art can understand that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and appropriate combinations thereof. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor or microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit . Such software may be distributed on computer-readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media). As is known to those of ordinary skill in the art, the term computer storage media includes both volatile and nonvolatile implemented in any method or technology for storage of information, such as computer readable instructions, data structures, program modules or other data flexible, 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 cartridges, magnetic tape, magnetic disk storage or other magnetic storage devices, or may Any other medium used to store desired information and which can be accessed by a computer. In addition, communication media typically include computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and can include any information delivery media, as is well known to those of ordinary skill in the art .

The above is a specific description of some implementations of the application, but the application is not limited to the above-mentioned embodiments, and those skilled in the art can make various equivalent deformations or replacements under the shared conditions that do not violate the scope of the application. Equivalent modifications or substitutions are included within the scope defined by the claims of the present application.

Claims (10)

1. A routing method that includes:

   receiving a plurality of route calculation requests of the first service, where the route calculation requests carry the first service information;

   For each first service, a target network slice corresponding to the first service is determined from multiple network slices according to the first service information, wherein the multiple network slices have the same physical structure and are forwarded by The physical network structure of the plane is divided;

   determining the first forwarding path information of the first service on the target network slice according to the slice topology information and the first service information of the target network slice corresponding to the first service;

   Sending the first forwarding path information to the forwarding plane device, so that the forwarding plane device forwards the first service according to the first forwarding path information.
2. The routing method according to claim 1, wherein the determining a target network slice corresponding to the first service from a plurality of network slices according to the first service information comprises:

   Input the slice topology information of multiple network slices and the first service information into the service deployment model to obtain the target network slice corresponding to the first service, wherein the service deployment model is composed of the multiple network slices. The slice topology information and a plurality of second service information are obtained by training.
3. The routing method according to claim 1, wherein the plurality of network slices include a first network slice and a second network slice, the first network slice includes a first link, and the second network slice includes a The second link corresponding to the first link, the routing method further includes:

   When the bandwidth of the first link is smaller than the bandwidth of all the first services to be routed through the first link, the idle bandwidth of the second link is transferred to the first link, so that the The bandwidth of the first link is greater than or equal to the bandwidth of all the first services to be routed through the first link.
4. The routing method according to claim 1, wherein, when the first service includes a deployment failure service in which the calculation of the first forwarding path information fails, the routing method further comprises:

   calculating, according to the first service information of the deployment failure service, second forwarding path information of the deployment failure service on the physical network structure of the forwarding plane;

   The second forwarding path information is sent to the forwarding plane device, so that the forwarding plane device forwards the deployment failure service according to the second forwarding path information.
5. The routing method according to claim 1, wherein the bandwidths of the plurality of network slices are set uniformly or randomly.
6. The routing method according to any one of claims 1 to 5, wherein the slice topology information of the target network slice includes node information, link bandwidth information and link inherent delay information.
7. The routing method according to any one of claims 1 to 5, wherein: the first service information includes source point information, sink point information and service bandwidth information.
8. A routing device, comprising:

   a request receiving unit, configured to receive a plurality of route calculation requests of the first service, the route calculation requests carrying the first service information;

   A service deployment unit is configured to, for each of the first services, determine a target network slice corresponding to the first service from multiple network slices according to the first service information, wherein the multiple network slices have The same physical structure and is divided by the physical network structure of the forwarding plane;

   A path determination unit, configured to determine a first forwarding path of the first service on the target network slice according to the slice topology information of the target network slice corresponding to the first service and the first service information information;

   A path sending unit, configured to send the first forwarding path information to the forwarding plane device, so that the forwarding plane device forwards the first service according to the first forwarding path information.
9. A controller, comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein, when the processor executes the computer program, the implementation of claims 1 to 7 The routing method described in any of the above.
10. A computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are used to execute the routing method according to any one of claims 1 to 7.

Images