A method and apparatus for adjusting configuration parameters of a network device

Abstract

The application provides a method and device for adjusting configuration parameters of network devices, aiming at balancing performance and energy consumption of a network system, so as to reduce energy consumption of the network system while maintaining normal forwarding of traffic. The method for adjusting configuration parameters of network devices comprises: obtaining network topology of the network system and information of predicted traffic of the network system, determining multiple traffic distribution relationships of the predicted traffic according to the network topology, selecting a traffic distribution relationship from the multiple traffic distribution relationships according to network performance and energy consumption corresponding to the predicted traffic of the network system based on the multiple traffic distribution relationships, and instructing at least one network device in the network system to apply configuration parameters corresponding to the selected traffic distribution relationship.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a method and apparatus for adjusting the configuration parameters of a network device. Background Technology

[0002] A network system comprises multiple network devices used to transmit traffic within the network. The energy consumption cost of these network devices constitutes a significant portion of the overall operating cost of the network system. This energy cost refers to the cost incurred by the network devices during operation. Therefore, reducing the operating cost of a network system requires reducing the energy consumption of its network devices. Summary of the Invention

[0003] This application provides a method and apparatus for adjusting the configuration parameters of network devices, aiming to balance the performance and energy consumption of the network system, thereby reducing the energy consumption of the network system while maintaining normal traffic forwarding.

[0004] Firstly, this application provides a method for adjusting the configuration parameters of a network device. This method is applied to network devices in a network system or to a control device used to control the network system. The control device can be a controller, server, or a control program running on a cloud platform. The control device acquires information about the network topology and predicted traffic of the network system. Based on the network topology, it determines multiple traffic allocation relationships for the predicted traffic. Based on network performance and energy consumption, it selects one traffic allocation relationship from these multiple relationships as the target traffic allocation relationship and instructs at least one network device in the network system to apply the configuration parameters corresponding to the target traffic allocation relationship. The predicted traffic of the network system includes the traffic of edge network devices within a first time period. The first time period is a time period later than the current time. The predicted traffic information includes the magnitude of the predicted traffic. The traffic allocation relationship can be used to describe the specific transmission of the predicted traffic in the network system, i.e., indicating the traffic sharing ratio on the forwarding path of the predicted traffic in the network system and the links included in the forwarding path. The network performance corresponding to the target traffic allocation relationship meets performance requirements, and the energy consumption corresponding to the target traffic allocation relationship meets energy consumption requirements. Thus, while ensuring the performance of the target network device, the energy consumption of the target network device is reduced, thereby reducing the total energy consumption of the entire network system. In this way, the performance and energy consumption of the network system are balanced, thereby reducing the total energy consumption of the network system while maintaining normal traffic forwarding.

[0005] In one possible design, different traffic allocation relationships can be used to indicate different forwarding paths, or to indicate different traffic sharing ratios on the same links within the same forwarding path. Specifically, assume that multiple traffic allocation relationships include a first traffic allocation relationship and a second traffic allocation relationship. Then, the forwarding paths indicated by the first traffic allocation relationship and the second traffic allocation relationship are different; that is, the forwarding path of a data flow in the network system configured according to the first traffic allocation relationship is different from the forwarding path of the same data flow in the network system configured according to the second traffic allocation relationship. Alternatively, if the forwarding path indicated by the first traffic allocation relationship is the same as the forwarding path indicated by the second traffic allocation relationship, then for the first sub-path and the second sub-path in the network system, the traffic sharing ratio of the predicted traffic indicated by the first traffic allocation relationship between the first sub-forwarding path and the second sub-forwarding path is different from the traffic sharing ratio of the predicted traffic indicated by the second traffic allocation relationship between the first sub-forwarding path and the second sub-forwarding path.

[0006] In one possible design, the control device acquires multiple sets of performance parameters. Each set of performance parameters corresponds to a traffic allocation relationship, meaning there is a one-to-one correspondence between the performance parameter sets and the traffic allocation relationships. Each performance parameter set includes one or more performance parameters. The performance parameters in a performance parameter set are the performance parameters of the network devices on the forwarding path indicated by the traffic allocation relationship corresponding to that performance parameter set. In other words, the performance parameter set corresponding to the first traffic allocation relationship includes the performance parameters of one or more network devices on the forwarding path of the predicted traffic when the network system configured according to the first traffic allocation relationship forwards the predicted traffic. Accordingly, the network performance indicated by the performance parameter set corresponding to the selected target traffic allocation relationship is superior to the network performance indicated by the performance requirement.

[0007] In one possible design, the control device acquires multiple energy consumption groups. Each energy consumption group corresponds to a traffic allocation relationship, meaning there is a one-to-one correspondence between energy consumption groups and traffic allocation relationships. Each energy consumption group includes the energy consumption values ​​of one or more network devices. The energy consumption values ​​in an energy consumption group are the energy consumption values ​​of the network devices on the forwarding path indicated by the traffic allocation relationship corresponding to that energy consumption group. That is, the energy consumption group corresponding to the first traffic allocation relationship includes the energy consumption values ​​of one or more network devices on the forwarding path of the predicted traffic when the network system configured according to the first traffic allocation relationship forwards the predicted traffic. Accordingly, the energy consumption group corresponding to the selected target traffic allocation relationship indicates an energy consumption value that is superior to the energy consumption value indicated by the energy consumption requirement.

[0008] In one possible design, a traffic allocation relationship may correspond to multiple configuration information sets, each set of which corresponds to a set of performance parameters and a set of energy consumption parameters. The control device can determine the optimal configuration information corresponding to a traffic allocation relationship using a network performance model. Specifically, the size of the traffic to be processed by multiple network devices on the forwarding path indicated by any one of the multiple traffic allocation relationships is determined. For example, assuming the multiple traffic allocation relationships include a first traffic allocation relationship, the size of the traffic to be processed by multiple network devices on the forwarding path indicated by the first traffic allocation relationship can be determined first. Then, based on the network performance prediction model, the size of the traffic to be processed by the target network device, and the first configuration information of the target network device, the first performance information and the first energy consumption information of the target network device can be determined. The first performance information indicates the performance index of the target network device within a first time period, the first energy consumption information indicates the energy consumption value of the target network device within the first time period, and the first configuration information includes multiple configuration parameters of the target network device at the current time and the value corresponding to each configuration parameter. The target network device is any one of the multiple network devices on the forwarding path indicated by any one of the traffic allocation relationships. After determining the first performance information and first energy consumption information of the target network device, it can be determined whether the performance indicators corresponding to the first performance information meet the performance conditions, and whether the energy consumption value corresponding to the first energy consumption information meets the energy consumption conditions. If the performance indicators corresponding to the first performance information meet the performance conditions and the energy consumption value corresponding to the first energy consumption information meets the energy consumption conditions, the second configuration information can be determined as the configuration parameter set of the target network device corresponding to the first traffic allocation relationship, the performance indicators corresponding to the first performance information can be determined as the performance parameters of the target network device in the performance parameter group of the first traffic allocation relationship, and the energy consumption value corresponding to the first energy consumption information can be determined as the energy consumption value of the target network device in the energy consumption group of the first traffic allocation relationship. The second configuration information corresponds to the first performance information and the first energy consumption information.

[0009] In one possible design, the control device can determine the first performance information and the first energy consumption information through multiple rounds of prediction, judgment, and adjustment. Specifically, the performance information and energy consumption information of the target network device under the first configuration information can be determined first based on the network performance prediction model, the amount of traffic to be processed by the target network device, and the first configuration information of the target network device. The performance information of the target network device under the first configuration information is called intermediate performance information, and the energy consumption information of the target network device under the first configuration information is called intermediate energy consumption information. Next, it is determined whether the performance indicators corresponding to the intermediate performance information meet the performance conditions, and whether the energy consumption value corresponding to the intermediate energy consumption information meets the energy consumption conditions. If the performance indicators corresponding to the intermediate performance information do not meet the performance conditions, and / or the energy consumption value corresponding to the intermediate energy consumption information does not meet the energy consumption conditions, the first configuration information of the target network device can be adjusted to obtain intermediate configuration information. Then, the intermediate performance information and intermediate energy consumption information of the target network device under the intermediate configuration information can be determined through the network performance prediction model, the amount of traffic to be processed by the target network device, and the intermediate configuration information. Then, it is determined again whether the performance indicators corresponding to the intermediate performance information under the intermediate configuration information meet the performance conditions, and whether the energy consumption value corresponding to the intermediate energy consumption information under the intermediate configuration information meets the energy consumption conditions. Thus, through multiple rounds of prediction, judgment, and adjustment, intermediate configuration information is obtained that ensures the performance indicators corresponding to the intermediate performance information meet the performance conditions, and the energy consumption value corresponding to the intermediate energy consumption information meets the energy consumption conditions. In this way, the intermediate performance information corresponding to this intermediate configuration information is determined as the aforementioned first performance information, and the intermediate energy consumption information corresponding to this intermediate configuration information is determined as the aforementioned first energy consumption information.

[0010] In one possible design, the control device can instruct the network system to apply a target traffic allocation relationship by indicating the forwarding path and the traffic sharing ratio of the links included in the forwarding path. Specifically, it can instruct at least one network device to apply the traffic sharing ratio on the links included in the forwarding path indicated by the target traffic allocation relationship, and send corresponding configuration parameters to the network devices on the forwarding path indicated by the target traffic allocation relationship to instruct these network devices to adjust the configuration parameters.

[0011] Secondly, this application provides an apparatus for adjusting configuration parameters of network devices. The apparatus includes an acquisition unit, a processing unit, and a transmission unit. The processing unit acquires information about the network topology of a network system and predicted traffic of the network system. The predicted traffic information includes the traffic volume of data streams from edge devices of the network system during a first time period, which is later than the current time. The processing unit is further configured to determine multiple traffic allocation relationships for the predicted traffic based on the network topology. Each traffic allocation relationship indicates the forwarding path of the predicted traffic in the network system and the traffic sharing ratio on the links included in the forwarding path. The processing unit is further configured to select a traffic allocation relationship from the multiple traffic allocation relationships based on the network performance and energy consumption corresponding to forwarding the predicted traffic in the network system according to the multiple traffic allocation relationships. The network performance corresponding to the selected traffic allocation relationship meets performance requirements, and the energy consumption corresponding to the selected traffic allocation relationship meets energy consumption requirements. The transmission unit instructs at least one network device in the network system to apply configuration parameters corresponding to the selected traffic allocation relationship.

[0012] In one possible design, multiple traffic allocation relationships include a first traffic allocation relationship and a second traffic allocation relationship. The forwarding path indicated by the first traffic allocation relationship is different from the forwarding path indicated by the second traffic allocation relationship. Alternatively, the forwarding path indicated by the first traffic allocation relationship is the same as the forwarding path indicated by the second traffic allocation relationship, and this same forwarding path includes a first sub-forwarding path and a second sub-forwarding path. The traffic sharing ratio of the predicted traffic indicated by the first traffic allocation relationship between the first sub-forwarding path and the second sub-forwarding path is different from the traffic sharing ratio of the predicted traffic indicated by the second traffic allocation relationship between the first sub-forwarding path and the second sub-forwarding path.

[0013] In one possible design, the acquisition unit is also used to acquire multiple sets of performance parameters. These multiple sets of performance parameters correspond one-to-one with the aforementioned various traffic allocation relationships, and each set of performance parameters includes the performance parameters of the network devices included in the forwarding path indicated by the corresponding traffic allocation relationship.

[0014] The network performance corresponding to the selected traffic allocation relationship meets the performance requirements, including: the network performance indicated by the performance parameter group corresponding to the selected traffic allocation relationship is better than the network performance indicated by the performance requirements.

[0015] In one possible design, the acquisition unit is also used to acquire multiple energy consumption groups. The multiple energy consumption groups correspond one-to-one with the multiple traffic allocation relationships, and each energy consumption group includes the energy consumption value of the network device included in the forwarding path indicated by the corresponding traffic allocation relationship.

[0016] The energy consumption corresponding to the selected traffic allocation relationship meets the energy consumption requirements, including: the sum of the energy consumption values ​​included in the energy consumption group corresponding to the selected traffic allocation relationship is the minimum of the sum of the energy consumption values ​​included in the energy consumption groups corresponding to the multiple traffic allocation relationships, or the energy consumption value included in the energy consumption group corresponding to the selected traffic allocation relationship is less than the energy consumption threshold.

[0017] In one possible design, the processing unit is configured to: determine the amount of traffic to be processed by multiple network devices on the forwarding path indicated by any one of the multiple traffic allocation relationships; and determine the first performance information and first energy consumption information of the target network device based on a network performance prediction model, the amount of traffic to be processed by the target network device, and the first configuration information of the target network device. The first performance information indicates the performance index of the target network device within the first time period, the first energy consumption information indicates the energy consumption value of the target network device within the first time period, and the first configuration information includes multiple configuration parameters of the target network device at the current time and the value corresponding to each configuration parameter. The target network device is any one of the multiple network devices on the forwarding path indicated by the multiple traffic allocation relationships. The processing unit is further configured to: when the performance index corresponding to the first performance information meets the performance condition and the energy consumption value corresponding to the first energy consumption information meets the energy consumption condition, determine the second configuration information as the set of configuration parameters of the target network device corresponding to the multiple traffic allocation relationships. The first performance information indicates the network performance of the target network device corresponding to any of the traffic allocation relationships, the first energy consumption information indicates the energy consumption of the target network device corresponding to any of the traffic allocation relationships, and the second configuration information corresponds to the first performance information and the first energy consumption information.

[0018] In one possible design, the processing unit is configured to determine intermediate performance information and intermediate energy consumption information of the target network device based on the network performance prediction model, the amount of traffic to be processed by the target network device, and the first configuration information of the target network device; when the performance index corresponding to the intermediate performance information does not meet the performance conditions and / or the energy consumption value corresponding to the intermediate energy consumption information does not meet the energy consumption conditions, adjust the first configuration information of the target network device to obtain intermediate configuration information; and determine the first performance information and first energy consumption information of the target network device based on the network performance prediction model, the amount of traffic to be processed by the target network device, and the intermediate configuration information.

[0019] In one possible design, the sending unit is configured to instruct the at least one network device to apply the traffic sharing ratio on the links included in the forwarding path indicated by the selected traffic allocation relationship; and to send the configuration parameters of the network devices included in the forwarding path indicated by the selected traffic allocation relationship to the corresponding network devices, so as to instruct the corresponding network devices to adjust the configuration parameters.

[0020] Thirdly, this application provides a network device. The network device includes a processor and a memory. The memory stores instructions or program code, and the processor retrieves and executes the instructions or program code from the memory to cause the device to implement the method for adjusting the configuration parameters of the network device as described in the first aspect or any possible implementation thereof.

[0021] Fourthly, this application provides a network system. The network system includes at least one network device and a control device. The at least one network device is used for forwarding traffic. The control device is used to implement the method for adjusting the configuration parameters of the network device as described in the first aspect or any possible implementation thereof.

[0022] Fifthly, this application provides a computer-readable storage medium. The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method for adjusting configuration parameters of a network device as described in the first aspect or any possible implementation thereof.

[0023] Sixthly, this application provides a computer program product. This computer program product includes a program or code that, when run on a computer, implements the method for adjusting configuration parameters of a network device as described in the first aspect or any possible implementation thereof. Attached Figure Description

[0024] Figure 1-A A system schematic diagram provided for an embodiment of this application;

[0025] Figure 1-B Another system schematic diagram provided for an embodiment of this application;

[0026] Figure 1-C Another system schematic diagram provided for an embodiment of this application;

[0027] Figure 1-D Another system schematic diagram provided for embodiments of this application;

[0028] Figure 2 A flowchart illustrating a method for adjusting configuration parameters of a network device, as provided in this application embodiment;

[0029] Figure 3A flowchart illustrating a method for determining performance and energy consumption corresponding to traffic allocation relationships, provided in an embodiment of this application;

[0030] Figure 4 A flowchart illustrating a method for determining first performance information and first energy consumption information provided in an embodiment of this application;

[0031] Figure 5 A schematic diagram of a device for adjusting the configuration parameters of a network device provided in an embodiment of this application;

[0032] Figure 6 This is a schematic diagram of the structure of a device provided in an embodiment of this application;

[0033] Figure 7 This is a schematic diagram of the structure of a device provided in an embodiment of this application. Detailed Implementation

[0034] The technical solutions provided in the embodiments of this application will now be described in conjunction with the accompanying drawings.

[0035] To reduce the overall energy consumption of a network system, some network devices can be shut down. These shut-down devices no longer consume power, thus reducing the overall energy consumption and saving on operating costs. In this embodiment, the energy consumption of a network device can refer to its power consumption. The energy consumption value can include the total power consumption of the network device over a period of time, or it can include the average power consumption of the network device over a period of time.

[0036] For example, a control device used to manage network devices can acquire traffic information for multiple links in the network system. Based on the link traffic information, it can shut down some network devices on those links and transfer the traffic carried by those devices to other network devices. Specifically, the control device can acquire traffic information for each link in the network system and allocate the traffic carried by underutilized links to other links. Then, it can shut down the network devices on those underutilized links. In this way, the number of network devices in operation in the network system is reduced, thereby reducing the overall energy consumption of the network system.

[0037] The following is a detailed description with reference to the accompanying drawings. See also... Figure 1-A This figure is a schematic diagram of a system provided in an embodiment of this application. Figure 1-AThe system shown includes a network system 110, terminal devices 121, 122, and 123, and a control device 130. The network system 110 includes network devices 111, 112, 113, 114, 115, 116, and 117. Network device 111 is connected to terminal devices 121, 112, and 115; network device 113 is connected to network devices 112, 114, and 116; network device 117 is connected to network devices 114, 116, and terminal device 123; terminal device 122 is connected to network device 114; network device 116 is also connected to network device 115; and the management device 130 can be connected to each network device in the network system 110. It is understood that the network system 110 may include more or fewer network devices, which will not be elaborated here.

[0038] As shown in Figure 1, terminal device 121 sends data stream A to terminal device 122 and data stream B to terminal device 123 through network system 110. Data stream A is transmitted through the path "network device 111 → network device 112 → network device 113 → network device 114", and data stream B is transmitted through the path "network device 111 → network device 115 → network device 116 → network device 117".

[0039] To reduce the overall power consumption of network system 110, control device 130 can acquire the traffic status of each link in network system 110 and shut down one or more network devices based on the link traffic status. For example, assuming that data flow A has a larger traffic volume and data flow B has a smaller traffic volume, control device 130 can integrate data flow B into the link that transmits data flow A, thereby shutting down some of the network devices that originally transmitted data flow B. Specifically, control device 130 can control data flow B to be transmitted via the path "network device 111 → network device 112 → network device 113 → network device 114 → network device 117". In this way, network devices 115 and 116 are no longer used to transmit data flows, and control device 130 can shut down network devices 115 and / or network device 116, thereby reducing the overall power consumption of network system 110. Alternatively, control device 130 can also reduce the configuration of network devices 115 and / or network device 116, which can also reduce the overall power consumption of network system 110.

[0040] However, shutting down certain network devices or downgrading the configuration of some network devices may increase the load on other network devices, affecting traffic transmission performance. For example, if a network device in the network system carries a lot of traffic, and the traffic carried by other network devices is consolidated onto that network device, then the actual traffic carried by that network device is too much, which may affect the normal forwarding of traffic.

[0041] See Figure 1-B ,exist Figure 1-A Based on this, control device 130 shuts down network devices 115 and 116. Data stream B is transmitted via the path "network device 111 → network device 112 → network device 113 → network device 114 → network device 117". If network device 114 is also used to transmit data streams other than data streams A and B, the actual traffic carried by network device 114 may exceed the theoretically achievable traffic, leading to performance degradation of the data streams carried by network device 114. This may result in packet loss, latency degradation, or other issues with data streams A and / or B, affecting normal data transmission.

[0042] In view of this, embodiments of this application provide a method for adjusting the configuration parameters of a network device, aiming to balance the performance and energy consumption of the network system, thereby reducing the energy consumption of the network system while maintaining normal traffic forwarding.

[0043] The method for adjusting the configuration parameters of a network device provided in this application embodiment can be applied to... Figure 1-A The network system shown. Among them, Figure 1-A Each network device can be a network device with forwarding capabilities, such as a router or switch, or a server or terminal device. The control device 130 can be a standalone physical device such as a server or controller, or it can be integrated into the network devices. For example, the control device 130 can be integrated into network device 111 within network system 110. Alternatively, the control device 130 can be a software module running on a cloud platform or cloud server; that is, the control device 130 is a virtual software device.

[0044] In this embodiment, devices 121 and 122 can be terminal devices. These terminal devices can be referred to as user equipment (UE), mobile station (MS), mobile terminal (MT), or terminals, etc. A terminal device is a device that provides voice and / or data connectivity to a user, or a chip located within that device. For example, a terminal device can be a handheld device with wireless connectivity, an in-vehicle device, etc. Terminal devices can be mobile phones, desktop computers, tablets, laptops, PDAs, mobile internet devices (MID), wearable devices, virtual reality (VR) devices, augmented reality (AR) devices, wireless or wired terminals in industrial control, wireless terminals in self-driving, wireless terminals in remote medical surgery, wireless terminals in smart grids, wireless terminals in transportation safety, wireless terminals in smart cities, wireless terminals in smart homes, or 5G residential gateways (5G RG) that support 5G access, etc.

[0045] The following is combined Figure 2 The technical solutions provided in the embodiments of this application are described below. See also... Figure 2 The figure is a flowchart of a method for adjusting the configuration parameters of a network device according to an embodiment of this application, specifically including the following S201-S203.

[0046] S201: Obtain information on the network topology of the network system and the predicted traffic of the network system.

[0047] In order to reduce the total power consumption of the network system while maintaining normal traffic forwarding, the control device can obtain information on the network topology and predicted traffic of the network system, so as to adjust the configuration parameters of one or more network devices in the network system.

[0048] The network topology of a network system indicates the physical connections between various network devices within the system. Predicted traffic information for a network system refers to information about data flows entering and / or leaving the network system over a future period. Specifically, predicted traffic information may include the volume of data flows from edge devices within a first time period. It is understood that if the network system is used to carry multiple data flows, the predicted traffic may include the size of each of the multiple data flows.

[0049] The first time period is a period of time later than the current time. Optionally, the first network device can periodically execute S201, in which case the first time period can be the time period corresponding to the next cycle.

[0050] The volume of a data stream can be represented by its throughput value, which can be the throughput, measured in units such as bits or bytes per second. The throughput value can also be the amount of data that a network device needs to process, measured in units such as bits, bytes, or megabytes.

[0051] The following example, using S201 executed by a control device, illustrates a specific method for obtaining information about the network topology and predicted traffic of a network system. It is understood that the control device can be a physical device such as a server or controller, or a software module running on a cloud platform.

[0052] In this embodiment of the application, the network topology of the network system can be pre-stored by the control device. Specifically, when deploying the network system, the connection relationships between various network devices in the network system can be recorded and saved in the control device.

[0053] Alternatively, the network topology of the network system can also be reconstructed by the control device based on the forwarding path information. Specifically, the control device first obtains the forwarding path information of the network system, which can be, for example, reported by the network devices to the control device. The forwarding path information includes information about at least one forwarding path in the network system. A forwarding path is the path that a data stream takes during transmission, representing the network devices that the data stream passes through during transmission. Since the network devices on the forwarding path are neighbors, the neighbor relationships between network devices in the network system can be determined based on the forwarding path information, thereby obtaining the network system topology. In one possible implementation, if the control device is a software module running on a cloud platform, the software module used to determine the network system topology and the control device can be different software modules.

[0054] In this embodiment, the predicted traffic information of the network system can be obtained based on the historical traffic information of the edge devices of the network system. Here, an edge device is a network device in the network system that connects to network devices in other network systems. The network system can be an autonomous system (AS) network or a user-defined network. For example, assuming the network is divided into an access network and an aggregation network, the edge device in the access network system can be, for example, an edge device between the access network and the aggregation network, or a network device in the access network connected to the aggregation network. Alternatively, the edge device can also be the egress router of a local area network (LAN).

[0055] Specifically, assuming the edge devices of the network equipment include a first network device, the control device can first obtain the historical traffic information reported by the first network device. The historical traffic information reported by the first network device includes the traffic values ​​of one or more interfaces of the first network device within a second time period. The second time period is earlier than the current time. Then, the control device can determine the traffic pattern of the first network device based on the historical traffic information, and determine the traffic value of the first network device within the first time period based on the traffic pattern, thus obtaining the predicted traffic information of the first network device.

[0056] For example, a control device can learn the traffic patterns of a first network device through a machine learning model, thereby obtaining predicted traffic information for the first network device. The machine learning model that has learned the traffic patterns of the first network device can be called a traffic prediction model. Traffic prediction models can be, for example, deep learning models, long short-term memory (LSTM) models, or recurrent neural network (RNN) models, and other artificial intelligence models.

[0057] During network system operation, the control device can periodically acquire historical traffic information from edge devices and determine the predicted traffic information for the edge devices in the next period based on this historical traffic information. Optionally, at the end of the period, the control device can adjust the traffic prediction model based on the traffic information of the edge devices in the current period and the predicted traffic information obtained in the previous period.

[0058] It is understood that the aforementioned topology reconstruction model and traffic prediction model can be pre-trained on other devices and configured on the control device, or they can be trained on the control device itself. Optionally, if the control device is a software module running on a cloud platform or cloud server, the topology reconstruction model and traffic prediction model can run on other software modules of the cloud platform or cloud server. The software module running the topology reconstruction model is used to obtain the network system topology, and the software module running the traffic prediction model is used to obtain predicted traffic information for multiple network devices in the network system.

[0059] S202: Based on the network topology, determine various traffic allocation relationships for the predicted traffic.

[0060] After obtaining the network topology, various traffic allocation relationships for predicted traffic can be determined. These traffic allocation relationships indicate the traffic distribution ratio of predicted traffic along the forwarding paths within the network system and the links included in those paths; in other words, they represent how the predicted traffic is specifically transmitted within the network system.

[0061] The predicted forwarding path in the network system indicates the forwarding path of the data flow within the network system during the first time period, representing which forwarding path the data flow takes during that first time period. In other words, based on the traffic allocation relationship, the forwarding path of each data flow in the predicted traffic can be determined within the first time period.

[0062] The traffic distribution relationship on the links included in the forwarding path represents the proportion of traffic on each link in the network system to the total traffic within the first time period. In other words, the traffic distribution relationship on the links included in the forwarding path indicates the proportion of traffic sent by each outgoing port of the network device to the total traffic received by the network device.

[0063] In other words, if a network device in a network system includes multiple outgoing ports, the device can determine the proportion of traffic allocated to each outgoing port based on traffic distribution rules. For example, suppose network device A includes outgoing ports A1 and A2, and the traffic distribution rule instructs network device A to send 80% of its traffic from outgoing port A1 and 20% from outgoing port A2. Then, during data forwarding, if network device A receives 10MB of data per second, it will send 8MB of data per second from outgoing port A1 and 2MB of data per second from outgoing port A2.

[0064] In the embodiments of this application, multiple traffic allocation relationships can be determined. Different traffic allocation relationships can indicate different forwarding paths or different traffic sharing ratios.

[0065] For example, suppose there are multiple traffic allocation relationships, including a first traffic allocation relationship and a second traffic allocation relationship. If the forwarding paths indicated by the first and second traffic allocation relationships are different, then the forwarding paths of data flows in a network system configured according to the first traffic allocation relationship are different from those in a network system configured according to the second traffic allocation relationship. For example, in a network system configured according to the first traffic allocation relationship, data flow A is transmitted along forwarding path A, and data flow B is transmitted along forwarding path B; in a network system configured according to the second traffic allocation relationship, data flow A is transmitted along forwarding path A, and data flow B is transmitted along path C; in a network system configured according to the third traffic allocation relationship, data flow A is transmitted along path D, and data flow B is transmitted along forwarding path B; in a network system configured according to the fourth traffic allocation relationship, data flow A is transmitted along forwarding path E, and data flow B is transmitted along forwarding path F. Here, forwarding paths AF are different forwarding paths, meaning that any two forwarding paths in forwarding path AF include at least one different network device.

[0066] When different traffic allocation relationships indicate the same forwarding path in a network system, the traffic sharing ratios indicated by these different traffic allocation relationships are different. For example, if the first traffic allocation relationship indicates that the predicted traffic forwarding path of the network system includes path A and path B (which can be called sub-paths), and the second traffic allocation relationship indicates that the predicted traffic forwarding path of the network system also includes path A and path B, then the traffic sharing ratios indicated by the first and second traffic allocation relationships are different. For example, the first traffic allocation relationship indicates that the predicted traffic sharing ratio on path A and path B is 5:5, that is, 50% of the predicted traffic is transmitted through path A and 50% of the predicted traffic is transmitted through path B; the second traffic allocation relationship indicates that the predicted traffic sharing ratio on path A and path B is 3:7, that is, 30% of the predicted traffic is transmitted through path A and 70% of the predicted traffic is transmitted through path B.

[0067] The traffic allocation ratio can be obtained by the control device based on the historical traffic allocation ratio of the network system. Specifically, the control device can periodically record the traffic allocation ratio of sub-forwarding paths on multiple forwarding paths in the network system. In this way, when it is necessary to adjust the configuration information of network devices, the control device can obtain the historical traffic allocation ratio and determine the traffic distribution relationship based on the historical traffic allocation ratio.

[0068] In the embodiments of this application, the various traffic allocation relationships can be preset in the control device, or they can be obtained by the control device through analysis and prediction of traffic information.

[0069] Optionally, the control device can also acquire routing information from the network system. This routing information reflects the routing within the network system at the current moment, indicating the forwarding direction of data flows within the network system. Correspondingly, the traffic prediction information also includes tuples representing the data flows from the edge devices of the network system within a first time period. A tuple is a set of information describing the data flow, including the destination Internet Protocol (IP) address. Optionally, the tuple may also include other information; for example, it may be a five-tuple representing the data flow, specifically including the protocol number, source IP address, source port number, destination IP address, and destination port number. The IP address can be an Internet Protocol version 4 (IPv4) address or an Internet Protocol version 6 (IPv6) address.

[0070] Similar to network topology, the routing information of a network system can be pre-stored in the control device, or it can be obtained by the control device based on the forwarding path information.

[0071] Accordingly, the control device can also determine various traffic allocation relationships for predicted traffic based on the network system topology, network routing information, and predicted traffic information. The control device can determine which forwarding paths the data streams included in the predicted traffic information can take based on the tuples and routing information, thereby more accurately determining the distribution ratio of predicted traffic on each forwarding path, or reducing the configuration requirements for network devices.

[0072] S203: Based on the network performance and energy consumption corresponding to the traffic forwarding prediction based on the multiple traffic allocation relationships of the network system, select one traffic allocation relationship from the multiple traffic allocation relationships.

[0073] After determining multiple traffic allocation relationships for predicted traffic, the performance and energy consumption information of the network system forwarding the predicted traffic under each of these relationships can be determined. Based on this performance and energy consumption information, a traffic allocation relationship is selected from the multiple relationships. The configuration parameters of at least one network device in the network system are then adjusted according to the selected traffic allocation relationship. For ease of explanation, the selected traffic allocation relationship will be referred to as the target traffic allocation relationship in the following text.

[0074] Specifically, the network performance corresponding to the target traffic allocation relationship meets the performance requirements, and the energy consumption corresponding to the target traffic allocation relationship meets the energy consumption requirements. The energy consumption value indicated by the energy consumption requirements can be an energy consumption threshold pre-stored in the control device, or it can be an energy consumption threshold calculated by the control device based on the predicted traffic information. In other words, in the network system configured according to the target traffic allocation relationship, the sum of the energy consumption values ​​of all network devices is less than the aforementioned energy consumption threshold.

[0075] For details on performance requirements, please refer to the following text; they will not be repeated here.

[0076] To select a target traffic allocation relationship from multiple options, we can first determine the performance and energy consumption corresponding to each relationship, and then select the appropriate relationship based on these parameters. Specifically, the performance of a traffic allocation relationship can be represented by a set of performance parameters, and the energy consumption can be represented by a set of energy consumption parameters.

[0077] Each performance parameter group corresponds to a traffic allocation relationship, meaning there is a one-to-one correspondence between performance parameter groups and traffic allocation relationships. A performance parameter group includes multiple performance parameters. The performance parameters in a performance parameter group are the performance parameters of network devices in the network system configured according to the traffic allocation relationship corresponding to that performance parameter group. Optionally, the multiple performance parameters in a performance parameter group can be the performance parameters of the same network device in the network system, or they can be the performance parameters of different network devices in the network system. For example, when a performance parameter group includes latency parameters, the latency in the performance parameter group can be the latency of a single network device, or it can be the end-to-end latency. When a performance parameter group includes packet loss rate, the packet loss rate in the performance parameter group can include the maximum packet loss rate of the network device, or it can include the end-to-end packet loss rate. When a performance parameter group includes throughput, the throughput in the performance parameter group can be the throughput of the network system's egress network device.

[0078] Each energy consumption group corresponds to a traffic allocation relationship, meaning there is a one-to-one correspondence between energy consumption groups and traffic allocation relationships. An energy consumption group includes one or more energy consumption values. These energy consumption values ​​can include the sum of the energy consumption values ​​of network devices on the forwarding paths configured according to the traffic allocation relationship corresponding to that energy consumption group.

[0079] For specific methods on obtaining performance parameter sets and energy consumption sets, please refer to [link / reference]. Figure 4 The description of the illustrated embodiments will not be repeated here.

[0080] The following examples illustrate methods for control devices to determine target flow distribution relationships.

[0081] In the first possible implementation, the control device first determines the candidate flow allocation relationship based on the performance parameter group, and then determines the target flow allocation relationship from the candidate flow allocation relationship based on the energy consumption group.

[0082] Specifically, after determining multiple performance parameter groups, the control device can select a traffic allocation relationship from these groups where the network performance indicated by the corresponding performance parameter group is superior to the performance indicated by the performance requirement. If only one performance parameter group indicates network performance superior to the performance requirement, the control device can determine the traffic allocation relationship corresponding to that performance parameter group as the target traffic allocation relationship. If multiple performance parameter groups indicate network performance superior to the performance requirement, the control device further determines the target traffic allocation relationship based on the energy consumption parameter group. In other words, the network performance indicated by the performance parameter group corresponding to the target traffic allocation relationship is superior to the network performance indicated by the performance requirement.

[0083] As mentioned above, performance parameter sets can include parameters such as latency, packet loss rate, and jitter. Accordingly, if the performance parameter set includes latency, the network performance indicated by the performance parameter set being superior to the network performance indicated by the performance requirement means that the latency indicated by the performance parameter set is less than the latency indicated by the performance requirement. If the performance parameter set includes packet loss rate, the network performance indicated by the performance parameter set being superior to the network performance indicated by the performance requirement means that the packet loss rate indicated by the performance parameter set is less than the packet loss rate indicated by the performance requirement. If the performance parameter set includes jitter, the network performance indicated by the performance parameter set being superior to the network performance indicated by the performance requirement means that the jitter indicated by the performance parameter set is less than the jitter indicated by the performance requirement.

[0084] If multiple performance parameter groups indicate network performance superior to that indicated by the performance requirements, the control device can determine the traffic allocation relationships corresponding to these performance parameter groups as candidate traffic allocation relationships. Then, based on the energy consumption group, the control device can select one traffic allocation relationship from the multiple candidate traffic allocation relationships as the target traffic allocation relationship.

[0085] For example, the control device can determine the traffic allocation relationship that minimizes the total energy consumption of the network system indicated by the energy consumption group as the target traffic allocation relationship. That is, the sum of the energy consumption values ​​of all network devices in the network system configured according to the target traffic allocation relationship is less than the sum of the energy consumption values ​​of all network devices in the network system configured according to any non-target traffic allocation relationship.

[0086] In a second possible implementation, the control device first determines the candidate flow allocation relationship based on the energy consumption group, and then determines the target flow allocation relationship from the candidate flow allocation relationships based on the performance parameter group. Specific implementation details can be found above and will not be repeated here.

[0087] In a third possible implementation, the control device can assign performance weights to the performance of the flow allocation relationship and energy consumption weights to its energy consumption, thereby determining the target flow allocation relationship by comprehensively considering performance and energy consumption. Specifically, the control device can calculate the comprehensive parameters corresponding to the flow allocation relationship by comprehensively considering its performance, performance weights, energy consumption, and energy consumption weights. These comprehensive parameters reflect the merits of the flow allocation relationship's performance and energy consumption. In this way, the control device can select the flow allocation relationship with the optimal comprehensive parameters from multiple flow allocation relationships as the target flow allocation relationship.

[0088] In a fourth possible implementation, the control device can determine the target traffic allocation relationship based on its performance, taking into account the energy consumption of the traffic allocation relationship. Specifically, the aforementioned multiple energy consumption groups may include multiple groups whose sum of energy consumption values ​​is less than an energy consumption threshold. To ensure the normal operation of the network system, the control device can identify traffic allocation relationships in the corresponding energy consumption groups whose sum of energy consumption values ​​is less than the energy consumption threshold as candidate traffic allocation relationships. These multiple traffic allocation relationships include multiple candidate traffic allocation relationships. Then, the control device can select the candidate traffic allocation relationship with the best performance from the multiple candidate traffic allocation relationships as the target traffic allocation relationship. Similarly, the control device can also determine the target traffic allocation relationship based on the energy consumption of the traffic allocation relationship in addition to its performance. That is, it first determines multiple candidate traffic allocation relationships that meet the performance requirements, and then selects the traffic allocation relationship whose network performance meets the performance requirements from the multiple candidate traffic allocation relationships as the target traffic allocation relationship. The control device can also determine target traffic allocation relationships one by one. For example, the control device first determines a traffic allocation relationship. When the network performance and energy consumption corresponding to this traffic allocation relationship meet the requirements, the control device can directly determine this traffic allocation relationship as the target traffic allocation relationship, without attempting to determine other traffic allocation relationships and their corresponding performance and energy consumption. This will not be elaborated further here.

[0089] In the fifth possible implementation, the control device can also determine the target traffic allocation relationship through machine learning models or other methods. This will not be elaborated upon here.

[0090] It is understandable that the above five implementation methods can be combined with each other, and the control device can also use other implementation methods to determine the target flow distribution relationship.

[0091] S204: Instructs at least one network device in the network system to apply configuration parameters corresponding to the selected traffic allocation relationship.

[0092] After selecting a traffic allocation relationship from multiple options, at least one network device in the network system can be instructed to apply configuration information corresponding to the selected traffic allocation relationship. In other words, the control device can determine the configuration information corresponding to the target traffic allocation relationship and instruct at least one network device in the network system to apply that configuration information. Here, the at least one network device is a network device whose configuration parameters after applying the target traffic allocation relationship differ from its current configuration parameters. For example, the at least one network device can be a single network device in the network system, multiple network devices in the network system, or all network devices in the network system.

[0093] As mentioned above, traffic allocation relationships can be used to indicate forwarding paths or the traffic sharing ratio on sub-paths of a forwarding path.

[0094] If traffic allocation relationships are used to indicate forwarding paths, then when instructing network devices to apply configuration parameters, the control device can issue new traffic redirection rules to the at least one network device, so that the network device can adjust the traffic forwarding path according to the traffic redirection rules. Specifically, the control device can modify the routing table of the at least one network device, for example, by modifying the routing table of the network device through a network configuration protocol (netconf), or by issuing flow specifications (flowspec) to change the traffic path.

[0095] If the traffic allocation relationship is used to indicate the traffic sharing ratio of different sub-forwarding paths on a forwarding path, then when instructing network devices to apply configuration parameters, the control device can instruct the at least one network device to apply the traffic sharing ratio on the links included in the forwarding path indicated by the target traffic allocation relationship. Then, the control device can send the configuration parameters corresponding to the target traffic allocation relationship to the network devices on the forwarding path indicated by the target traffic allocation relationship, so that these network devices can adjust according to the configuration parameters sent by the control device.

[0096] In this embodiment, multiple traffic allocation relationships for predicted traffic can be determined based on the network system topology. The traffic allocation relationship that satisfies both network performance and energy consumption requirements is selected as the target traffic allocation relationship. Next, network devices in the network system can be instructed to apply the configuration parameters corresponding to the target traffic allocation relationship. Then, during the first time period, the traffic flowing through the network system corresponds to the predicted traffic, and the forwarding behavior of the traffic within each network device matches the target traffic allocation relationship. Therefore, the network performance of each network device in the network system meets the performance requirements, and the energy consumption of the network system also meets the energy consumption requirements. Thus, by adjusting the traffic allocation relationship, for each network device in the network system, the traffic flowing through that network device during the first time period matches the configuration parameters of that network device, ensuring normal traffic transmission. Furthermore, the energy consumption of the network system configured according to the target traffic allocation relationship also meets the energy consumption requirements. While ensuring the performance of the target network device, the energy consumption of the target network device is reduced, thereby reducing the total energy consumption of the entire network system. In this way, the performance and energy consumption of the network system are balanced, thereby reducing the total energy consumption of the network system while maintaining normal traffic forwarding.

[0097] As described above, a traffic allocation relationship can be selected as the target traffic allocation relationship based on the performance and energy consumption of the network system corresponding to that relationship. However, under the same traffic allocation relationship, the configuration parameters of network devices in the network system may differ. These different configuration parameters will lead to different network system performance and / or energy consumption. In other words, one traffic allocation relationship may correspond to multiple performance parameter groups and / or multiple energy consumption groups.

[0098] To ensure that the selected target traffic allocation relationship is the optimal one in terms of performance and energy consumption, the performance parameter set and energy consumption set corresponding to each of the various traffic allocation relationships can be determined separately. The following section combines... Figure 3 Taking the first traffic allocation relationship included in multiple traffic allocation relationships as an example, this application describes a method for determining the performance parameter group and energy consumption group of the traffic allocation relationship in its embodiments. It is understood that... Figure 3 The first traffic allocation relationship and in the embodiment shown Figure 2 In the illustrated embodiment, the first traffic allocation relationship can be the same or different.

[0099] See Figure 3 The figure is a flowchart of a method for determining the performance and energy consumption corresponding to the first flow allocation relationship provided in the embodiments of this application, including the following S301-S303.

[0100] S301: Determine the amount of traffic to be processed by multiple network devices on the forwarding path indicated by the first traffic allocation relationship based on the first traffic allocation relationship and the predicted traffic information.

[0101] As described above, the performance parameter group for the first traffic allocation relationship can include the performance parameters of one or more network devices in the network system. To determine the performance parameter group and energy consumption group corresponding to the first traffic allocation relationship, the traffic volume of multiple network devices in the network system during the first time period can be determined first, so that the operation of the network system can be simulated based on the traffic volume in subsequent steps.

[0102] Specifically, the control device can first determine, based on the first traffic allocation relationship, the traffic volume of multiple network devices in the network system within a first time period under the forwarding path and traffic sharing ratio indicated by the first traffic allocation relationship. In one possible implementation, in order to fully simulate the working conditions of the network system, the control device can determine the traffic volume to be processed by each network device in the network system within the first time period based on the first traffic allocation relationship and the predicted traffic information.

[0103] S302: Based on the network performance prediction model, the amount of traffic to be processed by the target network device, and the first configuration information of the target network device, determine the first performance information and the first energy consumption information of the target network device.

[0104] After determining the amount of traffic to be processed by multiple network devices within a first time period, the control device uses a network performance prediction model to predict the performance and energy consumption information of the target network device within the first time period under the first configuration information. Here, the target network device is any one of the multiple network devices on the forwarding path indicated by the first traffic allocation relationship. In some possible implementations, the number of target network devices is multiple; that is, the control device determines the performance and energy consumption information of multiple network devices in the network system based on the network performance prediction model.

[0105] Network performance prediction models are used to simulate the operation of network devices, thereby determining the performance and energy consumption of these devices. Specifically, a network performance prediction model can simulate the operation of a network device over a given period based on its configuration and the amount of traffic it forwards, thus determining its performance and energy consumption information during that period. In other words, for a target network device, the control device can determine its performance and traffic information within a first time period based on the network performance prediction model, the amount of traffic to be processed by the target network device in that first time period, and the target network device's initial configuration information. The initial configuration information of the target network device includes multiple configuration parameters of the target network device at the current moment, and the corresponding values ​​for each parameter.

[0106] After determining the performance and energy consumption information of the target network device in the first time period, the control device can determine whether the performance indicators corresponding to the performance information of the target network device in the first time period meet the performance conditions, and whether the energy consumption value corresponding to the energy consumption information of the target network device in the first time period meets the energy consumption conditions.

[0107] The performance conditions correspond to the performance requirements described in S203 above, and the energy consumption conditions correspond to the energy consumption requirements described in S203 above. Specifically, the performance conditions can be determined based on the performance requirements. If the network system meets the performance requirements, then the target network device must meet the performance conditions. For example, if the performance requirements include end-to-end jitter less than A on the forwarding path, and the target network device is a network device on the forwarding path, then the performance conditions can include: the jitter of the target network device is less than A. Similarly, the energy consumption conditions can be determined based on the energy consumption requirements. For example, if the energy consumption requirement is that the total energy consumption of the network devices is less than an energy consumption threshold, the control device can determine the maximum energy consumption of the target network device that makes the total energy consumption of the network system less than the energy consumption threshold based on the energy consumption threshold and the topology of the network system, and determine this energy consumption as a sub-energy consumption threshold. Then the energy consumption conditions can include: the energy consumption value of the target network device is not greater than the sub-energy consumption threshold. Alternatively, the control device can first determine the energy consumption threshold of each network device, and then sum the energy consumption thresholds of each network device on the forwarding path as the energy consumption threshold of a forwarding path. Alternatively, the control device can first determine the latency threshold for each network device, and then add the latency thresholds of each network device on the forwarding path to the transmission time between the network devices as the latency threshold for the forwarding path. Alternatively, the control device can determine packet loss thresholds, jitter thresholds, etc., for each network device that are not greater than the packet loss thresholds and jitter thresholds indicated by the performance requirements of the network system.

[0108] If the performance indicators corresponding to the performance information of the target network device within the first time period do not meet the performance conditions, and / or the energy consumption value corresponding to the energy consumption information of the target network device within the first time period does not meet the energy consumption conditions, it indicates that the performance indicators of the target network device configured according to the first configuration information are poor or the energy consumption is high within the first time period. The control device can adjust the values ​​of the configuration parameters of the target network device and redetermine the performance and energy consumption of the target network device. In this way, the control device can adjust the configuration of the target network device once or multiple times until the performance indicators corresponding to the performance information of the target network device meet the performance requirements, and the energy consumption value corresponding to the energy consumption information of the target network device meets the energy consumption requirements.

[0109] If the performance indicators corresponding to the performance information of the target network device in the first time period meet the performance conditions, and the energy consumption value corresponding to the energy consumption information of the target network device in the first time period meets the energy consumption conditions, it indicates that the performance indicators of the target network device configured according to the first configuration information are superior or the energy consumption is lower in the first time period. Then the control device can determine the performance information obtained from the network performance prediction model as the first performance information, and the energy consumption information obtained from the network performance prediction model as the first energy consumption information, and continue to execute S303. In some possible implementations, the control device obtains the first performance information and the first energy consumption information after multiple rounds of prediction and adjustment. For more information on this, please refer to... Figure 4 This will not be elaborated upon here.

[0110] Understandably, if there are multiple target network devices, the control device will only determine the performance information as the first performance information and the energy consumption information as the first energy consumption information if the performance indicators corresponding to the performance information of each target network device meet the performance conditions and the energy consumption value corresponding to the energy consumption information of each target network device meets the energy consumption conditions. In other words, the control device will only execute S303 after determining that each target network device meets the performance and energy consumption conditions.

[0111] In some possible implementations, under the forwarding path and traffic sharing ratio corresponding to the first traffic allocation relationship, there may not be configuration information that ensures all target network devices meet the energy consumption and performance conditions. In this case, the control device selects the configuration parameter with the optimal performance metric and / or minimum energy consumption value from the various performance and energy consumption information corresponding to the first traffic allocation relationship, and determines the performance information corresponding to this configuration parameter as the first performance information and the energy consumption information corresponding to this configuration parameter as the first energy consumption information. Alternatively, the control device may discard the first traffic allocation relationship. In this case, the first traffic allocation relationship will not be selected when subsequently determining the target traffic allocation relationship.

[0112] S303: Determine that the performance parameter corresponding to the first performance information belongs to the performance parameter group of the first traffic allocation relationship, determine that the energy consumption value corresponding to the second performance information belongs to the energy consumption parameter group of the first traffic allocation relationship, and determine that the second configuration information is the configuration parameter set of the first traffic allocation relationship of the target network device.

[0113] In step S302, first performance information indicating that the corresponding performance indicators meet the performance conditions, and first energy consumption information indicating that the corresponding energy consumption values ​​meet the energy consumption conditions, are determined. Next, the performance parameter group and energy consumption parameter group corresponding to the first traffic allocation relationship can be determined based on the first performance information and the first energy consumption information. That is, the performance parameter group of the first traffic allocation relationship includes the performance parameters corresponding to the first performance information, and the energy consumption group of the first traffic allocation relationship includes the energy consumption values ​​corresponding to the first performance information. If there are multiple target network devices, the performance parameter group of the first traffic allocation relationship includes the performance parameters corresponding to the first performance information of each target network device, and the energy consumption group of the first traffic allocation relationship includes the energy consumption values ​​corresponding to the first energy consumption information of each target network device.

[0114] In addition, the second configuration information can be defined as the set of configuration parameters for the target network device under the first traffic allocation relationship. The second configuration information corresponds to the first performance information and the first energy consumption information, including multiple configuration parameters and the value corresponding to each parameter. That is, the target network device configured according to the second configuration information has performance parameters corresponding to the first performance information within a first time period, and energy consumption values ​​corresponding to the first energy consumption information within the first time period.

[0115] It is understandable that if the first traffic allocation relationship is determined as the target traffic allocation relationship in S203, then the aforementioned second configuration information belongs to the configuration information corresponding to the target traffic allocation relationship. That is to say, in S204, the target network device applies the second configuration information.

[0116] As described above, the control device can determine the initial performance and initial power consumption information of the target network device through multiple predictions and adjustments. The following section will combine... Figure 4 This process will be described in detail.

[0117] See Figure 4 , Figure 4 A flowchart of a method for determining first performance information and first energy consumption information provided in this application embodiment specifically includes the following S401-S405.

[0118] S401: The control device determines the intermediate performance information and intermediate energy consumption information corresponding to the first configuration information through the network performance prediction model.

[0119] The network performance prediction model is generated based on network performance training samples. Each training sample includes configuration parameters, the amount of traffic processed based on those parameters, corresponding network performance information, and corresponding energy consumption information. The network performance information and energy consumption information in the training samples serve as labels. The network performance prediction model can be a convolutional neural network (CNN) model. The model simulates the operation of network devices under specific traffic conditions to determine their performance and energy consumption. In other words, the control device can use the network performance prediction model to predict the performance and energy consumption of a target network device when configured with the first configuration information and the traffic volume is the same as the target network device's processing volume. Optionally, the performance information predicted based on the first configuration information can be referred to as intermediate performance information, and the performance information predicted based on the first configuration information can be referred to as intermediate energy consumption information.

[0120] In some possible implementations, intermediate performance information and intermediate energy consumption information can be obtained through different models. That is, the aforementioned network performance prediction model can include a transmission performance prediction model and an energy consumption prediction model. The transmission performance prediction model is used to determine the transmission performance of the network device within a first time period under intermediate configuration information, and the energy consumption prediction model is used to determine the energy consumption of the network device within the first time period under intermediate configuration information. The transmission performance prediction model includes, for example, one or more of latency prediction models, jitter prediction models, and packet loss prediction models. Each training sample of the transmission performance prediction model includes configuration parameters, the amount of traffic processed based on the configuration parameters, and corresponding network performance information, where the network performance information is the label of each training sample. Each training sample of the energy consumption prediction model includes configuration parameters, the amount of traffic processed based on the configuration parameters, and corresponding energy consumption information, where the energy consumption information is the label of each training sample.

[0121] S402: The control device determines whether the performance indicators corresponding to the intermediate performance information meet the preset conditions, and determines whether the energy consumption value corresponding to the intermediate energy consumption information is less than the energy consumption threshold.

[0122] After obtaining the intermediate performance information, the control device determines whether the performance indicators corresponding to the intermediate performance information meet the performance conditions, and determines whether the energy consumption value corresponding to the intermediate energy consumption information meets the energy consumption conditions. If the performance indicators corresponding to the intermediate performance information do not meet the preset conditions, and / or the energy consumption value corresponding to the intermediate energy consumption information is not less than the energy consumption threshold, the control device executes S403; if the performance indicators corresponding to the intermediate performance information meet the preset conditions, and the energy consumption value corresponding to the intermediate energy consumption information is less than the energy consumption threshold, the control device executes S405.

[0123] In some possible implementations, performance conditions include one or more performance parameters, and energy consumption conditions include energy consumption thresholds. Correspondingly, the performance metric corresponding to intermediate performance information satisfies the performance conditions if any performance parameter included in the performance metric corresponding to the intermediate energy consumption information is superior to the performance parameter corresponding to the performance condition. The energy consumption value corresponding to intermediate energy consumption information satisfies the energy consumption conditions if the energy consumption value is less than the energy consumption threshold. The energy consumption threshold can be preset or determined based on the amount of traffic to be processed by the target network device.

[0124] S403: The control device determines intermediate configuration information.

[0125] If the performance indicators corresponding to the intermediate performance information do not meet the preset conditions, and / or the energy consumption values ​​corresponding to the intermediate energy consumption information do not meet the energy consumption conditions, the control device can adjust the configuration parameters of the target network device to determine the intermediate configuration information. The intermediate configuration information includes at least one configuration parameter of the target network device, and the value corresponding to each configuration parameter. The configuration parameters included in the intermediate configuration information are different from the configuration parameters included in the first configuration information, and / or, there exists a configuration parameter whose value in the intermediate configuration information differs from its value in the first configuration information.

[0126] Specifically, if the performance indicators corresponding to the intermediate performance information do not meet the preset conditions, it indicates that the target network device running according to the first configuration information has poor performance in the first time period. Therefore, the control device adjusts the configuration of the target network device to improve its performance indicators. The performance of the target network device configured according to the intermediate configuration information is better than the performance of the target network device configured according to the first configuration information. Specifically, the control device can control the target network device to perform any one or more of the following actions: power on, wake from sleep, enable the processor, enable the processor core, enable the interface, and increase the maximum bandwidth of the interface. The intermediate configuration information is the configuration information of the target network device after performing the above actions.

[0127] If the energy consumption value corresponding to the intermediate energy consumption information is greater than or equal to the energy consumption threshold, it indicates that the target network device operating according to the first configuration information has a high energy consumption value within the first time period. Therefore, the control device adjusts the configuration of the target network device to reduce its energy consumption. The energy consumption value of the target network device after configuration based on the intermediate configuration information is less than the energy consumption value of the target network device after configuration based on the first configuration information. Specifically, the control device can control the execution of any one or more of the following actions: shutdown, hibernation, processor shutdown, interface shutdown, and reduction of maximum interface bandwidth. The intermediate configuration information is the configuration information of the target network device after performing the above actions.

[0128] Understandably, if the performance indicators corresponding to the intermediate performance information do not meet the preset conditions, and the energy consumption value corresponding to the intermediate energy consumption information is greater than or equal to the energy consumption threshold, in order to ensure the normal operation of the network system, the control device can prioritize improving the performance of the network device, and then select the configuration information with the lowest energy consumption from multiple configuration information that meet the performance requirements.

[0129] The following describes the specific methods for determining intermediate configuration information in control equipment.

[0130] In one possible implementation, the control device can determine a first policy based on intermediate performance information and intermediate energy consumption information, and then determine first configuration information based on the first policy. Specifically, the control device can be configured with multiple preset policies, each preset policy being used to adjust one or more configuration parameters of the target network device. After determining the intermediate performance information and intermediate energy consumption information, the control device can select one or more preset policies from the multiple preset policies, and use the selected policy as the first policy. Next, the control device adjusts the values ​​of the configuration parameters in the first configuration information according to the first policy, and determines intermediate configuration information based on the adjusted configuration parameters.

[0131] It is understandable that the adjustment of the configuration parameters in the first configuration information is performed by the control device, while the configuration parameters of the target network device are not adjusted by the control device.

[0132] In some possible implementations, the control device determines a first strategy through an intelligent decision-making model. This intelligent decision-making model selects one or more preset strategies from multiple preset strategies based on intermediate performance and energy consumption information. The intelligent decision-making model represents the impact of the target network device's configuration parameter values ​​on its energy consumption and performance. In other words, the intelligent decision-making model can determine which configuration parameters need to be modified to improve the target network device's performance or reduce its energy consumption. Optionally, the intelligent decision-making model is a reinforcement learning model.

[0133] After determining the intermediate configuration information, the control device can use a network performance prediction model to predict the performance and energy consumption information of the target network device under the intermediate configuration information within a first time period. The performance information of the target network device under the intermediate configuration information within the first time period is referred to as intermediate performance information, and the energy consumption information of the target network device under the intermediate configuration information within the first time period is referred to as intermediate energy consumption information. For specific methods of determining the intermediate performance information and intermediate energy consumption information, please refer to the description in S302 above, which will not be repeated here.

[0134] It is understandable that the intermediate performance information determined in S404 based on the intermediate configuration information may be different from the intermediate performance information determined in S401 based on the first configuration information; the intermediate energy consumption information determined in S404 based on the intermediate configuration information may be different from the intermediate energy consumption information determined in S401 based on the first configuration information.

[0135] After determining the intermediate performance information and intermediate energy consumption information, the control device can return to execute S402 to determine whether the performance indicators corresponding to the intermediate performance information corresponding to the intermediate configuration information meet the performance conditions, and whether the energy consumption value corresponding to the intermediate energy consumption information meets the energy consumption conditions.

[0136] As described in S402, if the performance indicators corresponding to the intermediate performance information corresponding to the intermediate configuration information meet the performance conditions, but the energy consumption value corresponding to the intermediate energy consumption information does not meet the energy consumption conditions, the control device executes S405. If the performance indicators corresponding to the intermediate performance information corresponding to the intermediate configuration information do not meet the performance conditions, and / or the energy consumption value corresponding to the intermediate energy consumption information does not meet the energy consumption conditions, the processor continues to execute S403. In this way, the control device can predict and adjust the configuration parameters of the target network device multiple times until the performance indicators of the target network device under the intermediate configuration information meet the performance conditions and the energy consumption value meets the energy consumption conditions in the first time period.

[0137] Understandably, to reduce the time required to determine the second configuration information, the control device can record the number of times intermediate configuration information is determined and set a maximum number of times the intermediate configuration information is calculated on the control device. If the number of times the control device determines the intermediate configuration information reaches the maximum number, the control device will not return to execute S402, but will continue to execute S405.

[0138] S405: The control device determines the intermediate performance information as the first performance information and the intermediate energy consumption information as the first energy consumption information.

[0139] If the control device performs multiple calculations, it indicates that the intermediate configuration information from the first calculation is insufficient to ensure that the target network device's performance indicators meet the preset conditions and that its energy consumption is less than the energy consumption threshold. To reduce the time required to determine the first performance information and the first energy consumption information, the control device can adjust the intelligent decision-making model based on the second configuration information. For example, the control device can use the intermediate performance information corresponding to the first configuration information, the intermediate energy consumption information corresponding to the first configuration information, and the second configuration information as training samples to retrain the intelligent decision-making model.

[0140] The following is combined Figure 1-A The network system shown will be used to further describe the method for controlling network devices provided in the embodiments of this application.

[0141] Assume that network devices 111, 112, 113, 114, 115, 116, and 117 each include two processors, each of which can process 10 megabytes of traffic per second. The traffic size of data stream A is 9 MB / s, and the size of data stream B is 8 MB / s.

[0142] In addition to data streams A and B, network system 110 also carries data stream C. Data stream C flows through network device 114 and has a size of 7 MB / s. Besides network device 114, network devices 111, 112, 113, 115, 116, and 117 are not used to carry data streams other than data streams A and / or B. That is, network devices 111 and 114 are edge network devices in network system 110.

[0143] If the configuration of each network device in network device 110 is not adjusted, network devices 111, 112, 113, 114, 115, 116 and 117 will always be in working state, resulting in high total power consumption of network system 110.

[0144] To reduce the overall power consumption of network system 110, control device 130 can acquire the network topology of network system 110, predicted traffic information of network device 111, predicted traffic information of network device 114, and first configuration information of at least one network device in network system 110. For ease of explanation, it is assumed that control device 130 stores the first configuration information of network devices 111, 112, 113, 114, 115, 116, and 117.

[0145] Next, the control device 130 can determine various traffic allocation relationships for predicted traffic based on the network topology of the network system 110, as well as the performance parameter group and energy consumption group corresponding to each traffic allocation relationship.

[0146] For example, suppose the control device 130 determines a first flow allocation relationship and a second flow allocation relationship.

[0147] The forwarding paths indicated by the first traffic allocation relationship are as follows: Data flow A is transmitted through the forwarding path "Network Device 111 → Network Device 112 → Network Device 113 → Network Device 114", data flow B is transmitted through the forwarding path "Network Device 111 → Network Device 112 → Network Device 113 → Network Device 116 → Network Device 117", and data flow C is transmitted along the original path. The forwarding paths of traffic in network system 110 during the first time period are as follows: Figure 1-C As shown.

[0148] The forwarding paths indicated by the second traffic allocation relationship are as follows: Data flow A is transmitted through the forwarding path "Network Device 111 → Network Device 112 → Network Device 113 → Network Device 114", data flow B is transmitted through the forwarding path "Network Device 111 → Network Device 112 → Network Device 115 → Network Device 116 → Network Device 117", and data flow C is transmitted along the original path. The forwarding paths of traffic in network system 110 during the first time period are as follows: Figure 1-D As shown.

[0149] The performance parameter group corresponding to the first traffic allocation relationship indicates that each network device in network system 110 is operating normally. The energy consumption group corresponding to the first traffic allocation relationship includes the energy consumption values ​​of network device 111, network device 112, network device 113, network device 114, network device 115, network device 116, and network device 117. The performance parameter group corresponding to the second traffic allocation relationship indicates that each network device in network system 110 is operating normally. The energy consumption group corresponding to the second traffic allocation relationship includes the energy consumption values ​​of network device 111, network device 112, network device 113, network device 114, network device 115, network device 116, and network device 117.

[0150] In the first traffic allocation relationship, the energy consumption values ​​of network devices 111, 112, 114, 116, and 117 in the energy consumption group are the same as those of network devices 111, 112, 114, 116, and 117 in the energy consumption group corresponding to the second traffic allocation relationship. The energy consumption value of network device 113 in the first traffic allocation relationship is equal to the sum of its base energy consumption value and the energy consumption values ​​of the two processors; the energy consumption value of network device 115 is zero. In the second traffic allocation relationship, the energy consumption value of network device 113 is the sum of its base energy consumption value and the energy consumption value of one processor; the energy consumption value of network device 115 is the sum of its base energy consumption value and the energy consumption value of one processor.

[0151] Based on the performance parameter groups and energy consumption groups of the first traffic allocation relationship and the second traffic allocation relationship, the control device 130 determines that the total energy consumption of the network system configured according to the first traffic allocation relationship is less than the total energy consumption of the network system configured according to the second traffic allocation relationship. Then, the control device 130 determines the first traffic allocation relationship as the target traffic allocation relationship and instructs each network device in the network system 110 to apply the configuration parameters corresponding to the first traffic allocation relationship. Specifically, the control device 130 can control network device 115 to shut down and instruct network devices 111, 112, and 113 to forward the predicted traffic according to the forwarding path indicated by the first traffic allocation relationship.

[0152] In this way, network device 115 is in a powered-off or sleep state, reducing the total power consumption of network system 110. Additionally, with Figure 1-B Compared to the proposed control method, data flow B is transmitted via the path "network device 111 → network device 112 → network device 113 → network device 116 → network device 117", bypassing network device 114. This avoids performance degradation of network device 114 and does not affect the normal transmission of other data flows. In this way, the performance and energy consumption of the network system are balanced, thereby reducing the overall energy consumption of the network system while maintaining normal traffic forwarding.

[0153] See Figure 5 This application embodiment also provides a device 500 for adjusting the configuration parameters of a network device, which can achieve... Figure 2 , Figure 3 or Figure 4 The illustrated embodiment controls the function of the device. The device 500 includes an acquisition unit 510, a processing unit 520, and a sending unit 530. The acquisition unit 510 is used to implement… Figure 2 S201 in the illustrated embodiment. Processing unit 520 is used to implement Figure 2 In the illustrated embodiment, S202 and S203, the transmitting unit 530 is used to implement... Figure 2 S204 in the illustrated embodiment.

[0154] Specifically, the acquisition unit 510 is used to acquire information about the network topology of the network system and the predicted traffic of the network system. The predicted traffic information includes the data flow volume of the edge devices of the network system during a first time period, which is later than the current time.

[0155] Processing unit 520 is configured to determine multiple traffic allocation relationships for the predicted traffic based on the network topology. Each traffic allocation relationship indicates the forwarding path of the predicted traffic in the network system and the traffic sharing ratio on the links included in the forwarding path. Processing unit 520 is further configured to select a traffic allocation relationship from the multiple traffic allocation relationships based on the network performance and energy consumption corresponding to the forwarding of the predicted traffic by the network system based on the multiple traffic allocation relationships. The network performance corresponding to the selected traffic allocation relationship meets the performance requirements, and the energy consumption corresponding to the selected traffic allocation relationship meets the energy consumption requirements.

[0156] The sending unit 530 is used to instruct at least one network device in the network system to apply configuration parameters corresponding to the selected traffic allocation relationship.

[0157] Please refer to the above for the specific execution process. Figure 2 , Figure 3 or Figure 4 The detailed descriptions of the corresponding steps in the illustrated embodiments will not be repeated here.

[0158] It should be noted that the division of units in this embodiment is illustrative and represents only one logical functional division; in actual implementation, other division methods may be used. The functional units in this embodiment can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. For example, in the above embodiment, the processing unit and the sending unit can be the same unit or different units. The integrated unit can be implemented in hardware or as a software functional unit.

[0159] Figure 6 This is a schematic diagram of the structure of a device 600 provided in an embodiment of this application. The device 500 for adjusting the configuration parameters of the network device mentioned above can be... Figure 6 This is achieved using the device shown. See also Figure 6 The device 600 includes at least one processor 601, a communication bus 602 and at least one communication interface 604. Optionally, the device 600 may also include a memory 603.

[0160] The processor 601 can be a general-purpose central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits (ICs) used to control the execution of the program in this application. The processor 601 can be used to process messages or parameters to implement the method for adjusting the configuration parameters of the network device provided in the embodiments of this application.

[0161] For example, when Figure 2 The control equipment in the middle is through Figure 6 When implemented using the device shown, the processor acquires information about the network topology of the network system and the predicted traffic of the network system. The predicted traffic information includes the traffic volume of data streams from edge devices of the network system within a first time period, which is later than the current time. The processor is also configured to determine multiple traffic allocation relationships for the predicted traffic based on the network topology. Each traffic allocation relationship indicates the forwarding path of the predicted traffic in the network system and the traffic sharing ratio on the links included in the forwarding path. The processor is further configured to select a traffic allocation relationship from the multiple traffic allocation relationships based on the network performance and energy consumption corresponding to forwarding the predicted traffic according to the multiple traffic allocation relationships. The network performance corresponding to the selected traffic allocation relationship meets performance requirements, and the energy consumption corresponding to the selected traffic allocation relationship meets energy consumption requirements. The processor is also configured to instruct at least one network device in the network system to apply configuration parameters corresponding to the selected traffic allocation relationship.

[0162] The communication bus 602 is used to transfer information between the processor 601, the communication interface 604, and the memory 603.

[0163] The memory 603 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions. It may also be a random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions. Furthermore, it may be a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media, or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures that can be accessed by a computer, but is not limited thereto. The memory 603 may exist independently and be connected to the processor 601 via a communication bus 602. Alternatively, the memory 603 may be integrated with the processor 601.

[0164] Optionally, the memory 603 stores program code or instructions for executing the technical solutions provided in the embodiments of this application, and the processor 601 controls the execution of these instructions. The processor 601 executes the program code or instructions stored in the memory 603. The program code may include one or more software modules. Optionally, the processor 601 may also store program code or instructions for executing the technical solutions provided in the embodiments of this application, in which case the processor 601 does not need to read the program code or instructions from the memory 603.

[0165] The communication interface 604 can be a transceiver or similar device used to communicate with other devices or communication networks. The communication network can be Ethernet, a radio access network (RAN), or a wireless local area network (WLAN), etc. The communication interface 604 can be an Ethernet interface, a Fast Ethernet (FE) interface, or a Gigabit Ethernet (GE) interface, etc.

[0166] In a specific implementation, as one example, device 600 may include multiple processors, for example... Figure 6 The processors 601 and 605 are shown. Each of these processors can be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. A processor here can refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0167] Figure 7This is a schematic diagram of the structure of a device 700 provided in an embodiment of this application. Figure 2 , Figure 3 and Figure 4 The control device in the middle can be Figure 7 This is achieved using the device shown. See also Figure 7The illustrated device structure diagram shows that device 700 includes a main control board and one or more interface boards. The main control board is communicatively connected to the interface boards. The main control board, also called a main processing unit (MPU) or route processor card, includes a CPU and memory. It is responsible for controlling and managing the various components in device 700, including routing calculations, device management, and maintenance functions. The interface boards, also called line processing units (LPUs) or line cards, are used to receive and send messages. In some embodiments, the main control board and interface boards, or interface boards themselves, communicate via a bus. In some embodiments, the interface boards communicate via a switch fabric unit (SFU). In this case, device 700 also includes a switch fabric unit, which is communicatively connected to the main control board and interface boards. The switch fabric unit is used to forward data between the interface boards and can also be called a switch fabric unit (SFU). Each interface board includes a CPU, memory, a forwarding engine, and an interface card (IC), where the interface card may include one or more communication interfaces. The communication interface can be an Ethernet interface, FE interface, or GE interface, etc. The CPU communicates with the memory, forwarding engine, and interface card respectively. The memory stores the forwarding table. The forwarding engine forwards received packets based on the forwarding table stored in the memory. If the destination address of the received packet is the IP address of device 700, the packet is sent to the CPU of the main control board or interface board for processing; if the destination address of the received packet is not the IP address of device 700, the forwarding table is consulted based on the destination. If the next hop and outgoing interface corresponding to the destination address are found in the forwarding table, the packet is forwarded to the outgoing interface corresponding to the destination address. The forwarding engine can be a network processor (NP). The interface card, also called a daughter card, can be installed on the interface board and is responsible for converting photoelectric signals into data frames, performing validity checks on the data frames, and forwarding them to the forwarding engine for processing or the interface board CPU. In some embodiments, the CPU can also perform the functions of the forwarding engine, such as implementing soft forwarding based on a general-purpose CPU, thus eliminating the need for a forwarding engine on the interface board. In some embodiments, the forwarding engine can be implemented using an ASIC or a field-programmable gate array (FPGA). In some embodiments, the memory storing the forwarding table can also be integrated into the forwarding engine as part of the forwarding engine.

[0168] This application also provides a chip system, including: a processor coupled to a memory, the memory being used to store programs or instructions, wherein when the program or instructions are executed by the processor, the chip system performs the aforementioned functions. Figure 2 or Figure 3 The illustrated embodiment describes a method for adjusting the configuration parameters of a network device, executed by the control device.

[0169] Optionally, the chip system may include one or more processors. These processors can be implemented in hardware or software. When implemented in hardware, the processor can be a logic circuit, integrated circuit, etc. When implemented in software, the processor can be a general-purpose processor that reads software code stored in memory. Optionally, the chip system may also include one or more memories. These memories can be integrated with the processor or separated from it; this application does not limit this. For example, the memory can be a non-transient processor, such as a read-only memory (ROM), which can be integrated with the processor on the same chip or disposed on different chips. This application does not specifically limit the type of memory or the arrangement of the memory and processor.

[0170] For example, the chip system can be an FPGA, an ASIC, a system on chip (SoC), a CPU, an NP, a digital signal processor (DSP), a microcontroller unit (MCU), a programmable logic device (PLD), or other integrated chips.

[0171] It should be understood that each step in the above method embodiments can be completed by integrated logic circuits in the processor hardware or by instructions in software form. The method steps disclosed in the embodiments of this application can be directly manifested as being executed by a hardware processor, or being executed by a combination of hardware and software modules in the processor.

[0172] This application also provides a computer-readable storage medium including instructions that, when run on a computer, cause the computer to perform the method described in the above method embodiments, which is executed by a control device, for adjusting the configuration parameters of a network device.

[0173] This application also provides a computer program product containing instructions that, when run on a computer, causes the computer to execute the method described in the above method embodiments, which is executed by a control device, for adjusting the configuration parameters of a network device.

[0174] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0175] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0176] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical module division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0177] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be obtained according to actual needs to achieve the purpose of this embodiment.

[0178] Furthermore, the module units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software module unit.

[0179] If the integrated unit is implemented as a software module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0180] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method of adjusting a configuration parameter of a network device, the method comprising: The method comprises: obtaining network topology of a network system and predicted traffic information of the network system, the predicted traffic information comprising traffic size of data flow of an edge device of the network system in a first time period, the first time period being later than a current time; determining multiple traffic distribution relationships of the predicted traffic according to the network topology, each traffic distribution relationship indicating traffic distribution proportion of the predicted traffic on a forwarding path of the network system and a link included in the forwarding path; selecting a traffic distribution relationship from the multiple traffic distribution relationships according to network performance and energy consumption of the network system based on the multiple traffic distribution relationships, the selected traffic distribution relationship corresponding to network performance satisfying a performance requirement, wherein the traffic carried by a network device in the forwarding path indicated by the selected traffic distribution relationship satisfies a transmission performance requirement; and the energy consumption corresponding to the selected traffic distribution relationship satisfies an energy consumption requirement; indicating at least one network device in the network system to apply a configuration parameter corresponding to the selected traffic distribution relationship.

2. The method of claim 1, wherein, The multiple traffic distribution relationships comprise a first traffic distribution relationship and a second traffic distribution relationship; the forwarding path indicated by the first traffic distribution relationship is different from the forwarding path indicated by the second traffic distribution relationship; or, the forwarding path indicated by the first traffic distribution relationship is the same as the forwarding path indicated by the second traffic distribution relationship, the same forwarding path comprises a first sub-forwarding path and a second sub-forwarding path, and the traffic distribution proportion of the predicted traffic between the first sub-forwarding path and the second sub-forwarding path indicated by the first traffic distribution relationship is different from the traffic distribution proportion of the predicted traffic between the first sub-forwarding path and the second sub-forwarding path indicated by the second traffic distribution relationship.

3. The method of claim 1, wherein, The method further comprises: obtaining multiple performance parameter groups, a corresponding relationship between the multiple performance parameter groups and the multiple traffic distribution relationships being one-to-one correspondence, each performance parameter group in the multiple performance parameter groups comprising performance parameters of network devices included in a forwarding path indicated by a corresponding traffic distribution relationship; the network performance corresponding to the selected traffic distribution relationship satisfying the performance requirement comprises: the network performance indicated by the performance parameter group corresponding to the selected traffic distribution relationship being better than the network performance indicated by the performance requirement.

4. The method of claim 1, wherein, The method further comprises: obtaining multiple energy consumption groups, a corresponding relationship between the multiple energy consumption groups and the multiple traffic distribution relationships being one-to-one correspondence, each energy consumption group in the multiple energy consumption groups comprising energy consumption values of network devices included in a forwarding path indicated by a corresponding traffic distribution relationship; the energy consumption corresponding to the selected traffic distribution relationship satisfying the energy consumption requirement comprises: a sum of the energy consumption values included in the energy consumption group corresponding to the selected traffic distribution relationship being a minimum value of sums of energy consumption values respectively included in the energy consumption groups corresponding to the multiple traffic distribution relationships, or the energy consumption value included in the energy consumption group corresponding to the selected traffic distribution relationship being less than an energy consumption threshold.

5. The method according to claim 3 or 4, characterized in that, The method further comprises: determine, according to the network performance prediction model, the traffic size to be processed by the target network device, and first configuration information of the target network device, first performance information and first energy consumption information of the target network device, the first performance information indicating a performance index of the target network device in the first time period, the first energy consumption information indicating an energy consumption value of the target network device in the first time period, and the first configuration information including a plurality of configuration parameters of the target network device at the current moment and a value corresponding to each configuration parameter, the target network device being any one of the network devices on the forwarding path indicated by the selected traffic distribution relationship; determine, according to the network performance prediction model, the traffic size to be processed by the target network device, and first configuration information of the target network device, first performance information and first energy consumption information of the target network device, the first performance information indicating a performance index of the target network device in the first time period, the first energy consumption information indicating an energy consumption value of the target network device in the first time period, and the first configuration information including a plurality of configuration parameters of the target network device at the current moment and a value corresponding to each configuration parameter, the target network device being any one of the network devices on the forwarding path indicated by the selected traffic distribution relationship; determine, according to the network performance prediction model, the traffic size to be processed by the target network device, and first configuration information of the target network device, first performance information and first energy consumption information of the target network device, the first performance information indicating a performance index of the target network device in the first time period, the first energy consumption information indicating an energy consumption value of the target network device in the first time period, and the first configuration information including a plurality of configuration parameters of the target network device at the current moment and a value corresponding to each configuration parameter, the target network device being any one of the network devices on the forwarding path indicated by the selected traffic distribution relationship; 6. The method of claim 5, wherein, determine, according to the network performance prediction model, the traffic size to be processed by the target network device, and first configuration information of the target network device, first performance information and first energy consumption information of the target network device, the first performance information indicating a performance index of the target network device in the first time period, the first energy consumption information indicating an energy consumption value of the target network device in the first time period, and the first configuration information including a plurality of configuration parameters of the target network device at the current moment and a value corresponding to each configuration parameter, the target network device being any one of the network devices on the forwarding path indicated by the selected traffic distribution relationship; determine, according to the network performance prediction model, the traffic size to be processed by the target network device, and first configuration information of the target network device, first performance information and first energy consumption information of the target network device, the first performance information indicating a performance index of the target network device in the first time period, the first energy consumption information indicating an energy consumption value of the target network device in the first time period, and the first configuration information including a plurality of configuration parameters of the target network device at the current moment and a value corresponding to each configuration parameter, the target network device being any one of the network devices on the forwarding path indicated by the selected traffic distribution relationship; determine, according to the network performance prediction model, the traffic size to be processed by the target network device, and first configuration information of the target network device, first performance information and first energy consumption information of the target network device, the first performance information indicating a performance index of the target network device in the first time period, the first energy consumption information indicating an energy consumption value of the target network device in the first time period, and the first configuration information including a plurality of configuration parameters of the target network device at the current moment and a value corresponding to each configuration parameter, the target network device being any one of the network devices on the forwarding path indicated by the selected traffic distribution relationship; determine, according to the network performance prediction model, the traffic size to be processed by the target network device, and first configuration information of the target network device, first performance information and first energy consumption information of the target network device, the first performance information indicating a performance index of the target network device in the first time period, the first energy consumption information indicating an energy consumption value of the target network device in the first time period, and the first configuration information including a plurality of configuration parameters of the target network device at the current moment and a value corresponding to each configuration parameter, the target network device being any one of the network devices on the forwarding path indicated by the selected traffic distribution relationship; 7. The method according to any one of claims 1 to 4, characterized in that, the apparatus comprises: ​ ​ 8. An apparatus for adjusting a configuration parameter of a network device, the apparatus comprising: ​ The acquisition unit is configured to acquire information of a network topology of a network system and predicted traffic of the network system, the information of the predicted traffic including traffic sizes of data flows of edge devices of the network system in a first time period, the first time period being later than a current time point. The processing unit is configured to determine a plurality of traffic distribution relationships of the predicted traffic according to the network topology, each traffic distribution relationship indicating a traffic distribution proportion of the predicted traffic on a forwarding path of the network system and links included in the forwarding path. The processing unit is further configured to select a traffic distribution relationship from the plurality of traffic distribution relationships according to network performance and energy consumption of the network system based on the plurality of traffic distribution relationships, the network performance of the selected traffic distribution relationship satisfying a performance requirement, wherein the traffic carried by network devices in the forwarding path indicated by the selected traffic distribution relationship satisfies a transmission performance requirement, and the energy consumption of the selected traffic distribution relationship satisfies an energy consumption requirement. The sending unit is configured to instruct at least one network device in the network system to apply a configuration parameter corresponding to the selected traffic distribution relationship.

9. The apparatus of claim 8, wherein, The plurality of traffic distribution relationships include a first traffic distribution relationship and a second traffic distribution relationship. The forwarding path indicated by the first traffic distribution relationship is different from the forwarding path indicated by the second traffic distribution relationship. Alternatively, The forwarding path indicated by the first traffic distribution relationship is the same as the forwarding path indicated by the second traffic distribution relationship, the same forwarding path includes a first sub-forwarding path and a second sub-forwarding path, and the traffic distribution proportion of the predicted traffic between the first sub-forwarding path and the second sub-forwarding path indicated by the first traffic distribution relationship is different from the traffic distribution proportion of the predicted traffic between the first sub-forwarding path and the second sub-forwarding path indicated by the second traffic distribution relationship.

10. The apparatus of claim 8, wherein The acquisition unit is further configured to acquire a plurality of performance parameter groups, a correspondence between the plurality of performance parameter groups and the plurality of traffic distribution relationships being one-to-one, and each performance parameter group in the plurality of performance parameter groups including performance parameters of network devices included in a forwarding path indicated by a corresponding traffic distribution relationship. The network performance of the selected traffic distribution relationship satisfying the performance requirement includes that network performance indicated by a performance parameter group corresponding to the selected traffic distribution relationship is better than network performance indicated by the performance requirement.

11. The apparatus of claim 8, wherein The acquisition unit is further configured to acquire a plurality of energy consumption groups, a correspondence between the plurality of energy consumption groups and the plurality of traffic distribution relationships being one-to-one, and each energy consumption group in the plurality of energy consumption groups including energy consumption values of network devices included in a forwarding path indicated by a corresponding traffic distribution relationship. The energy consumption corresponding to the selected flow distribution relationship satisfies an energy consumption requirement, including that a sum of energy consumption values included in an energy consumption group corresponding to the selected flow distribution relationship is a minimum value of a sum of energy consumption values included in energy consumption groups corresponding to the multiple flow distribution relationships, or the energy consumption group corresponding to the selected flow distribution relationship includes energy consumption values less than an energy consumption threshold.

12. The apparatus of claim 10 or 11, wherein, The processing unit is further configured to determine, according to any one of the multiple flow distribution relationships, a size of traffic to be processed by a plurality of network devices on a forwarding path indicated by the any one of the multiple flow distribution relationships. The processing unit is further configured to determine, according to a network performance prediction model, the size of traffic to be processed by a target network device, and first configuration information of the target network device, first performance information and first energy consumption information of the target network device, the first performance information indicating a performance indicator of the target network device in the first time period, the first energy consumption information indicating an energy consumption value of the target network device in the first time period, and the first configuration information including a plurality of configuration parameters of the target network device at the current time and a value corresponding to each configuration parameter, the target network device being any one of the plurality of network devices on the forwarding path indicated by the any one of the multiple flow distribution relationships. The processing unit is further configured to determine, when a performance indicator corresponding to the first performance information satisfies a performance condition and an energy consumption value corresponding to the first energy consumption information satisfies an energy consumption condition, second configuration information as a set of configuration parameters of the target network device corresponding to the any one of the multiple flow distribution relationships, the first performance information indicating network performance of the target network device corresponding to the any one of the multiple flow distribution relationships, the first energy consumption information indicating energy consumption of the target network device corresponding to the any one of the multiple flow distribution relationships, and the second configuration information corresponding to the first performance information and the first energy consumption information.

13. The apparatus of claim 12, wherein, The processing unit is configured to determine, according to the network performance prediction model, the size of traffic to be processed by the target network device, and the first configuration information of the target network device, intermediate performance information and intermediate energy consumption information of the target network device. The processing unit is further configured to adjust the first configuration information of the target network device to obtain intermediate configuration information, when a performance indicator corresponding to the intermediate performance information does not satisfy the performance condition and / or an energy consumption value corresponding to the intermediate energy consumption information does not satisfy the energy consumption condition. The processing unit is further configured to determine, according to the network performance prediction model, the size of traffic to be processed by the target network device, and the intermediate configuration information, the first performance information and the first energy consumption information of the target network device.

14. The apparatus of any one of claims 8 to 11, wherein, The sending unit is configured to instruct the at least one network device to apply a traffic distribution ratio on a link included in a forwarding path indicated by the selected flow distribution relationship. The sending unit is further configured to send configuration parameters of network devices included in the forwarding path of the selected traffic distribution relationship indication to the corresponding network devices, so as to instruct the corresponding network devices to adjust the configuration parameters.

15. A network device, comprising: The device comprises a memory and a processor, the memory is configured to store instructions, and the processor is configured to run the instructions, so that the network device executes the method for adjusting the configuration parameters of the network device according to any one of claims 1-7.

16. A network system, characterized by The network system comprises a plurality of network devices and a control device, the plurality of network devices are configured to forward traffic, and the control device is configured to implement the method for adjusting the configuration parameters of the network device according to any one of claims 1-7.

17. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the method for adjusting the configuration parameters of the network device according to any one of claims 1-7.

18. A computer program product, characterised in that, The computer program product comprises a program or code, and the program or code is run on a computer to implement the method for adjusting the configuration parameters of the network device according to any one of claims 1-7.

Images