US-12621222-B2 - Method for determining network optimization policy, apparatus, and system

US12621222B2US 12621222 B2US12621222 B2US 12621222B2US-12621222-B2

Abstract

A method for determining a network optimization policy, an apparatus, and a system are provided. A network optimization device can process a performance parameter of a network and a service requirement of a target service flow by using a network optimization algorithm, to obtain a target network optimization policy, and can apply the target network optimization policy to the target service flow, to enable transmission performance of the target service flow to satisfy the service requirement. Because the network optimization device can automatically determine the network optimization policy based on the network optimization algorithm, efficiency of determining the network optimization policy is effectively improved. In addition, when determining the network optimization policy, the network optimization device further considers the performance parameter of the network.

Inventors

Huanhuan Zhang
Qi Yu
Yi Kai
Mingjie Cai
Boyuan YU
Chen Sun

Assignees

HUAWEI TECHNOLOGIES CO., LTD.

Dates

Publication Date: 20260505
Application Date: 20240306
Priority Date: 20210907

Claims (20)

1 . A method for determining a network optimization policy, wherein the method comprises: obtaining a performance parameter of a network; selecting at least one network optimization technology from a plurality of different network optimization technologies by processing an input parameter using a network optimization algorithm, to obtain a target network optimization policy comprising at least one parameter value of a parameter value range of a network optimization technology selected by the network optimization algorithm, wherein the input parameter comprises the performance parameter of the network and a service requirement of a target service flow in the network, and wherein the plurality of different network optimization technologies includes at least two of a forward error correction (FEC) technology, a compression technology, a multi-feed and selective receiving technology, a per-packet load sharing technology, a jitter buffer technology, an active lost packet retransmission technology, or a transmission control protocol (TCP) acceleration technology; and applying the target network optimization policy to the target service flow, to enable transmission performance of the target service flow in the network to satisfy the service requirement.
2 . The method according to claim 1 , wherein the obtaining the performance parameter of the network comprises: obtaining a performance parameter of a link through which the target service flow is to pass in the network.
3 . The method according to claim 1 , wherein the processing the input parameter comprises: processing the input parameter using the network optimization algorithm, to select at least one network optimization technology, including the network optimization technology, from a plurality of different network optimization technologies, to obtain the target network optimization policy.
4 . The method according to claim 3 , wherein the input parameter further comprises first temporary transmission performance of the target service flow obtained after a temporary network optimization technology is applied to the target service flow, wherein the temporary network optimization technology belongs to the plurality of different network optimization technologies; and wherein the selecting at least one network optimization technology from a plurality of different network optimization technologies, to obtain the target network optimization policy comprises: selecting, from the plurality of different network optimization technologies, in response to the first temporary transmission performance not satisfying the service requirement, at least one network optimization technology different from the temporary network optimization technology, to obtain the target network optimization policy.
5 . The method according to claim 1 , wherein the processing the input parameter comprises: processing the input parameter by using the network optimization algorithm, to select a group of parameter values from the parameter value range of the network optimization technology, to obtain the target network optimization policy.
6 . The method according to claim 5 , wherein the input parameter further comprises: second temporary transmission performance of the target service flow obtained after the network optimization technology whose parameter value is a first temporary parameter value is applied to the target service flow; and wherein the selecting the group of parameter values from the parameter value range of the network optimization technology, to obtain the target network optimization policy comprises: adjusting, in response to the second temporary transmission performance not satisfying the service requirement, the first temporary parameter value within the parameter value range of the network optimization technology to obtain the target network optimization policy.
7 . The method according to claim 5 , wherein the input parameter further comprises: third temporary transmission performance of the target service flow obtained after the network optimization technology whose parameter value is a second temporary parameter value is applied to the target service flow; and wherein the selecting the group of parameter values, including the at least one parameter value, from the parameter value range of the network optimization technology, to obtain the target network optimization policy comprises: when the third temporary transmission performance is higher than the service requirement, adjusting the second temporary parameter value within the parameter value range of the network optimization technology to obtain the target network optimization policy, wherein device resources required for executing the network optimization technology whose parameter value is the second temporary parameter value are more than device resources required for executing the target network optimization policy.
8 . The method according to claim 1 , wherein the input parameter further comprises resource use information of a network device configured to execute the target network optimization policy; and wherein the resource use information is associated with making device resources of the network device available for execution of the target network optimization policy.
9 . The method according to claim 1 , wherein the processing the input parameter comprises: inputting the performance parameter and the service requirement into an optimization model, to obtain one or more different network optimization technologies output by the optimization model, wherein the target network optimization policy is determined from the one or more different network optimization technologies.
10 . The method according to claim 9 , wherein the optimization model is obtained by performing training on training samples, and wherein the training samples comprise a reference network optimization policy, wherein a performance parameter of the network before the reference network optimization policy is applied to a reference service flow, and wherein transmission performance of the reference service flow obtained after the reference network optimization policy is applied to the reference service flow.
11 . The method according to claim 1 , wherein the method for determining the network optimization policy is applied to a network optimization device, and wherein the network optimization device is a controller, and wherein the applying the target network optimization policy to the target service flow comprises: delivering the target network optimization policy to a network device configured to execute the target network optimization policy.
12 . The method according to claim 1 , wherein the method for determining the network optimization policy is applied to a network optimization device, and wherein the applying the target network optimization policy to the target service flow comprises: executing the target network optimization policy on the target service flow.
13 . A network optimization device comprising: a non-transitory memory; a processor; and a computer program stored on the non-transitory memory and executable by the processor, wherein the computer program comprises instructions that, when executed by the processor, enable the network optimization device to: obtain a performance parameter of a network; selecting at least one network optimization technology from a plurality of different network optimization technologies by processing an input parameter using a network optimization algorithm, to obtain a target network optimization policy comprising at least one parameter value of a parameter value range of a network optimization technology selected by the network optimization algorithm, wherein the input parameter comprises the performance parameter of the network and a service requirement of a target service flow in the network, and wherein the plurality of different network optimization technologies includes at least two of a forward error correction (FEC) technology, a compression technology, a multi-feed and selective receiving technology, a per-packet load sharing technology, a jitter buffer technology, an active lost packet retransmission technology, or a transmission control protocol (TCP) acceleration technology; and apply the target network optimization policy to the target service flow, to enable transmission performance of the target service flow in the network to satisfy the service requirement.
14 . The device according to claim 13 , wherein the computer program comprises further instructions that, when executed by the processor, enable the network optimization device to: obtain a performance parameter of a link through which the target service flow is to pass in the network.
15 . The device according to claim 13 , wherein the computer program comprises further instructions that, when executed by the processor, enable the network optimization device to: process the input parameter by using the network optimization algorithm, to select at least one network optimization technology including the network optimization technology, from a plurality of different network optimization technologies, to obtain the target network optimization policy.
16 . The device according to claim 15 , wherein the input parameter further comprises first temporary transmission performance of the target service flow obtained after a temporary network optimization technology is applied to the target service flow, wherein the temporary network optimization technology belongs to the plurality of different network optimization technologies; and wherein the computer program comprises further instructions that, when executed by the processor, further enable the network optimization device to: select, from the plurality of different network optimization technologies, in response to the first temporary transmission performance not satisfying the service requirement, at least one network optimization technology different from the temporary network optimization technology, to obtain the target network optimization policy.
17 . The device according to claim 13 , wherein the computer program comprises further instructions when executed by the processor further enable the network optimization device to: process the input parameter using the network optimization algorithm, to select a group of parameter values from the parameter value range of the network optimization technology, to obtain the target network optimization policy.
18 . The device according to claim 17 , wherein the input parameter further comprises second temporary transmission performance of the target service flow obtained after the network optimization technology whose parameter value is a first temporary parameter value is applied to the target service flow; and wherein the computer program comprises further instructions that, when executed by the processor, further enable the network optimization device to: when the second temporary transmission performance does not satisfy the service requirement, adjust the first temporary parameter value within the parameter value range of the network optimization technology, to obtain the target network optimization policy.
19 . The device according to claim 17 , wherein the input parameter further comprises third temporary transmission performance of the target service flow obtained after the network optimization technology whose parameter value is a second temporary parameter value is applied to the target service flow; and wherein the computer program comprises further instructions that, when executed by the processor further, enable the network optimization device to: adjust, in response to the third temporary transmission performance being higher than the service requirement, the second temporary parameter value within the parameter value range of the network optimization technology, to obtain the target network optimization policy, wherein device resources required for executing the network optimization technology whose parameter value is the second temporary parameter value are more than device resources required for executing the target network optimization policy.
20 . The device according to claim 13 , wherein the input parameter further comprises resource use information of a network device configured to execute the target network optimization policy; and wherein the resource use information is associated with making device resources of the network device available for execution of the target network optimization policy.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS This application is a continuation of International Application No. PCT/CN2022/116441, filed on Sep. 1, 2022, which claims priorities to Chinese Patent Application No. 202111043719.X, filed on Sep. 7, 2021 and Chinese Patent Application No. 202111270140.7, filed on Oct. 29, 2021. All of the aforementioned patent applications are hereby incorporated by reference in their entireties. TECHNICAL FIELD This application relates to the communication field, and in particular, to a method for determining a network optimization policy, an apparatus, and a system. BACKGROUND A wide area network (WAN) is a communication network that connects local area networks or metropolitan area networks in different areas. To ensure transmission quality of a service flow in a WAN, the WAN usually can provide the following network optimization policies: forward error correction (FEC), compression, multi-feed and selective receiving, per-packet load sharing, jitter buffer, active lost packet retransmission, transmission control protocol (TCP) acceleration, and the like. In a related technology, an operation and maintenance engineer may configure, in a network device based on a service requirement of a service flow, for example, a requirement defined in the service level agreement (SLA), a network optimization policy that can satisfy the service requirement. Then, the network device may execute the network optimization policy on the service flow. For example, if a service requirement of a service flow is high quality and a high throughput, the network optimization policy configured by the operation and maintenance engineer for the service flow may include FEC. If a service requirement of a service flow is high efficiency, the network optimization policy configured by the operation and maintenance engineer for the service flow may include TCP acceleration. However, efficiency of the foregoing method for determining and configuring the network optimization policy by the operation and maintenance engineer based on the service requirement is low. SUMMARY This application provides a method for determining a network optimization policy, an apparatus, and a system, to alleviate low efficiency of configuring a network optimization policy. According to one aspect, a method for determining a network optimization policy is provided, applied to a network optimization device, where the method includes: obtaining a performance parameter of a network, and processing an input parameter by using a network optimization algorithm, to obtain a target network optimization policy, where the input parameter includes the performance parameter of the network and a service requirement of a target service flow; and then applying the target network optimization policy to the target service flow, to enable transmission performance of the target service flow in the network to satisfy the service requirement. Because the network optimization device can automatically determine the network optimization policy based on the network optimization algorithm, efficiency of determining the network optimization policy is effectively improved. In addition, when determining the network optimization policy, the network optimization device not only considers the service requirement of the service flow, but also considers the performance parameter of the network. Therefore, the determined network optimization policy can adapt to a dynamic change of network performance. Optionally, the process of obtaining a performance parameter of a network may include: obtaining a performance parameter of a link through which the target service flow is to pass in the network. Because the performance of the link through which the target service flow is to pass is a key factor that affects the transmission performance of the target service flow, when the target network optimization policy is determined, only the performance parameter of the link through which the target service flow is to pass in the network may be obtained. Therefore, while ensuring accuracy of the determined target network optimization policy, a data volume of data that needs to be processed by the network optimization device can be reduced, and data processing efficiency can be improved. Optionally, the process of processing the performance parameter of the network and a service requirement of a target service flow by using a network optimization algorithm, to obtain a target network optimization policy may include: processing the input parameter by using the network optimization algorithm, to select at least one network optimization technology from a plurality of different network optimization technologies, to obtain the target network optimization policy. The plurality of different network optimization technologies may include at least two of the following technologies: an FEC technology, a compression technology, a multi-feed and selective receiving technology, a per-packet load shari