WO2024040976A1 - 网络设备的能耗控制方法及装置 - Google Patents

网络设备的能耗控制方法及装置 Download PDF

Info

Publication number
WO2024040976A1
WO2024040976A1 PCT/CN2023/087055 CN2023087055W WO2024040976A1 WO 2024040976 A1 WO2024040976 A1 WO 2024040976A1 CN 2023087055 W CN2023087055 W CN 2023087055W WO 2024040976 A1 WO2024040976 A1 WO 2024040976A1
Authority
WO
WIPO (PCT)
Prior art keywords
network
energy
network device
network equipment
time series
Prior art date
Application number
PCT/CN2023/087055
Other languages
English (en)
French (fr)
Inventor
王宁诚
何力
徐泽华
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2024040976A1 publication Critical patent/WO2024040976A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/02—Details
    • H04L12/12—Arrangements for remote connection or disconnection of substations or of equipment thereof
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0803—Configuration setting
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0803—Configuration setting
    • H04L41/0823—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability
    • H04L41/0833—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability for reduction of network energy consumption
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W52/00—Power management, e.g. TPC [Transmission Power Control], power saving or power classes
    • H04W52/02—Power saving arrangements

Definitions

  • Embodiments of the present invention relate to the field of communications, and specifically, to a method and device for controlling energy consumption of network equipment.
  • Embodiments of the present invention provide an energy consumption control method and device for network equipment, so as to at least solve the problems in related technologies.
  • a method for controlling energy consumption of network equipment includes: using a time series prediction algorithm to predict the protection bandwidth of the network equipment according to network configuration information; and using a time series based on the operating status information of the network equipment.
  • the prediction algorithm predicts changes in traffic volume of the network device; plans a business processing protection strategy based on the predicted protection bandwidth of the network device and the predicted change in traffic volume of the network device, and obtains energy-saving action decisions for the network device .
  • an energy consumption control device for network equipment including: a traffic load prediction module configured to use a time series prediction algorithm to predict the protection bandwidth of the network equipment based on network configuration information; predict The business volume change module is configured to use a time series prediction algorithm to predict the business volume changes of the network equipment based on the network equipment operating status information; the energy-saving decision-making module is configured to predict the business volume and maximum business volume carried by the network equipment based on the prediction According to the traffic volume changes of the network equipment, the business processing protection strategy is planned, and the energy-saving action decision is obtained.
  • a computer-readable storage medium is also provided.
  • a computer program is stored in the computer-readable storage medium, wherein the computer program is configured to execute any of the above methods when running. Steps in Examples.
  • an electronic device including a memory and a processor.
  • a computer program is stored in the memory, and the processor is configured to run the computer program to perform any of the above. Steps in method embodiments.
  • Figure 1 is a network equipment hardware structure block diagram of a method for controlling energy consumption of network equipment according to an embodiment of the present invention
  • Figure 2 is a flow chart of an energy consumption control method for network equipment according to an embodiment of the present invention
  • Figure 3 is a flow chart of an energy consumption control method for network equipment according to an embodiment of the present invention.
  • Figure 4 is a flow chart of an energy consumption control method for network equipment according to an embodiment of the present invention.
  • Figure 5 is a flow chart of an energy consumption control method for network equipment according to an embodiment of the present invention.
  • FIG. 6 is a structural block diagram of an energy consumption control device for network equipment according to an embodiment of the present invention.
  • FIG. 7 is a structural block diagram of an energy consumption control device for network equipment according to an embodiment of the present invention.
  • FIG. 8 is a structural block diagram of an energy consumption control device for network equipment according to an embodiment of the present invention.
  • FIG. 9 is a structural block diagram of an energy consumption control device for network equipment according to an embodiment of the present invention.
  • Figure 10 is a network architecture block diagram of an energy consumption control method for network equipment implemented according to a scenario embodiment of the present invention.
  • Figure 11 is a schematic diagram of the protection strategy principle according to the scenario embodiment of the present invention.
  • Figure 12 is a flow chart of network-level information sensing and transmission according to a scenario embodiment of the present invention.
  • Figure 13 is a schematic diagram of the principle of network element port aggregation according to a scenario embodiment of the present invention.
  • Figure 14 is a flow chart of energy saving processing on the network device side according to a scenario embodiment of the present invention.
  • Figure 15 is a schematic diagram of the assessment and correction principle of the time series prediction algorithm according to the scenario embodiment of the present invention.
  • FIG. 1 is a hardware structure block diagram of a network device according to an energy consumption control method of a network device according to an embodiment of the present invention.
  • the network device may include one or more (only one is shown in Figure 1) processors 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the above-mentioned network device may also include a transmission device 106 and an input and output device 108 for communication functions.
  • processors 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA
  • a memory 104 for storing data
  • the above-mentioned network device may also include a transmission device 106 and an input and output device 108 for communication functions.
  • a network device may also include more or fewer components than shown in FIG. 1 , or have a different configuration than shown in FIG. 1 .
  • the memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the energy consumption control method of network equipment in the embodiment of the present invention.
  • the processor 102 runs the computer program stored in the memory 104 , thereby executing various functional applications and data processing, that is, implementing the above method.
  • Memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory.
  • the memory 104 may further include memory located remotely relative to the processor 102, and these remote memories may be connected to the management terminal through a network. Examples of the above-mentioned networks include but are not limited to the Internet, intranets, local area networks, mobile communication networks and combinations thereof.
  • the transmission device 106 is used to receive or send data via a network.
  • Specific examples of the above-mentioned network may include a wired network provided by a communication provider of the network device.
  • the transmission device 106 includes a network adapter (Network Interface Controller, NIC for short), which can be connected to other network devices through optical fibers to communicate with the Internet.
  • the transmission device 106 may be an optical module, which is used to communicate with the Internet through wired means.
  • FIG. 2 is a flow chart of an energy consumption control method for network equipment according to an embodiment of the present invention. As shown in Figure 2, the The process includes the following steps:
  • Step S202 According to the network configuration information, use a time series prediction algorithm to predict the protection bandwidth of the network device;
  • Step S204 Use a time series prediction algorithm to predict changes in traffic volume of the network device based on the operating status information of the network device;
  • Step S206 Plan a service processing protection strategy based on the predicted protection bandwidth of the network device and the predicted traffic volume change of the network device, and obtain an energy-saving action decision for the network device.
  • a time series prediction algorithm is used to predict the protection bandwidth of the network device according to the network configuration information; according to the operating status information of the network device, a time series prediction algorithm is used to predict the traffic change of the network device; according to the predicted Describe the business volume and maximum business volume carried by network equipment as well as the predicted traffic volume changes of the network equipment, plan business processing protection strategies, obtain energy-saving action decisions for the network equipment, and solve the lag in response and poor reliability in related technologies. , the problem of easy interruption of network services has been achieved, and the effect of improving the sensitivity and reliability of network business processing has been achieved.
  • the execution subject of the above steps may be a base station, a terminal, a bearer transmission device, etc., but is not limited thereto.
  • the process of using the time series prediction algorithm to predict the protection bandwidth of the network device is: using the time series prediction algorithm to predict the traffic volume carried by the network equipment port and the maximum traffic volume of the port set, based on the port set
  • the maximum business volume is used as the protection bandwidth of the network device port
  • the protection bandwidth of the network device is obtained based on the protection bandwidth of the network device port.
  • the protection bandwidth of the network device can be obtained by summing up the protection bandwidth of the network device port.
  • the network element ports of each network device are aggregated to obtain a corresponding port set.
  • Figure 3 is a flow chart of an energy consumption control method for network equipment according to an embodiment of the present invention. As shown in Figure 3, the process includes the following steps:
  • Step S302 obtain network configuration information
  • Step S304 Construct a network connection topology diagram based on the network configuration information, and aggregate network element ports;
  • Step S306 According to the network configuration information, use a time series prediction algorithm to predict the protection bandwidth of the network device;
  • Step S308 Use a time series prediction algorithm to predict changes in traffic volume of the network device based on the operating status information of the network device;
  • Step S310 Plan a service processing protection strategy based on the predicted protection bandwidth of the network device and the predicted traffic volume change of the network device, and obtain an energy-saving action decision for the network device.
  • the network device operating status information includes: network device performance data and network device device status information; wherein the network device performance data includes at least one of the following: port traffic information, CPU utilization, memory utilization ;
  • the network equipment device status information includes at least one of the following: single board working status, chip working status, inter-board communication connection status, and device performance indicators.
  • the content of the above network device operating status information is for illustration. During the specific implementation process, the content of the network device operating status information can be adjusted according to the actual situation, and is not limited here.
  • the method further includes: dynamically adjusting the weight coefficient of the time series prediction algorithm based on the predicted network equipment traffic volume changes and the real traffic volume.
  • Figure 4 is a flow chart of an energy consumption control method for network equipment according to an embodiment of the present invention. As shown in Figure 4, the process includes the following steps:
  • Step S402 According to the network configuration information, use a time series prediction algorithm to predict the protection bandwidth of the network device;
  • Step S404 Use a time series prediction algorithm to predict changes in traffic volume of the network device based on the operating status information of the network device;
  • Step S406 Dynamically adjust the weight coefficient of the time series prediction algorithm according to the predicted network equipment business volume changes and the real business volume;
  • Step S408 Plan a business processing protection strategy based on the predicted protection bandwidth carried by the network device and the predicted traffic volume change of the network device, and obtain an energy-saving action decision for the network device.
  • the protection policy includes at least one of the following: network device active and backup protection reservation; network device balancing redundancy reservation; network-level maximum service load reservation.
  • planning a business processing protection strategy and obtaining an energy-saving action decision includes: sequentially calculating the number of boards, modules, and chips that can handle the load of the device, and comparing the calculation results with the device operating resource status to obtain the Energy-saving action decisions.
  • the method further includes: converting the energy-saving action decision into a communication message and sending it to the corresponding device for energy-saving operation.
  • Figure 5 is a flow chart of an energy consumption control method for network equipment according to an embodiment of the present invention. As shown in Figure 5, the process includes the following steps:
  • Step S502 According to the network configuration information, use a time series prediction algorithm to predict the protection bandwidth of the network device;
  • Step S504 Use a time series prediction algorithm to predict changes in traffic volume of the network device based on the operating status information of the network device;
  • Step S506 Dynamically adjust the weight coefficient of the time series prediction algorithm based on the predicted network equipment business volume changes and the real business volume;
  • Step S508 Plan a business processing protection strategy based on the predicted protection bandwidth carried by the network device and the predicted change in traffic volume of the network device, and obtain an energy-saving action decision for the network device.
  • Step S510 Convert the energy-saving action decision into a communication message and send it to the corresponding device for energy-saving operation.
  • the method according to the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is Better implementation.
  • the technical solution of the present invention can be embodied in the form of a software product in essence or the part that contributes to the existing technology.
  • the computer software product is stored in a storage medium (such as ROM/RAM, disk, CD), including several instructions to cause a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in various embodiments of the present invention.
  • This embodiment also provides an energy consumption control device for network equipment.
  • the device is used to implement the above embodiments and preferred implementations. What has been described will not be described again.
  • the term "module” may be a combination of software and/or hardware that implements a predetermined function.
  • the apparatus described in the following embodiments is preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.
  • FIG. 6 is a structural block diagram of an energy consumption control device for network equipment according to an embodiment of the present invention.
  • the energy consumption control device 60 includes: a prediction protection bandwidth module 610, which is configured to use time series according to network configuration information.
  • the prediction algorithm predicts the traffic volume and the maximum traffic volume carried by the network equipment, and uses the maximum traffic volume as the protection bandwidth of the link;
  • the prediction traffic volume change module 620 is configured to use a time series prediction algorithm to predict the network equipment based on the operating status information of the network equipment. Changes in business volume;
  • the energy-saving decision module 630 is configured to plan business processing protection strategies and obtain energy-saving action decisions based on the predicted business volume and maximum business volume carried by the network equipment and the predicted business volume changes of the network equipment.
  • Figure 7 is a structural block diagram of an energy consumption control device for network equipment according to an embodiment of the present invention.
  • the energy consumption control device 70 includes, in addition to various modules in Figure 6, It also includes: a configuration information acquisition module 710, configured to acquire network configuration information; a port aggregation module 720, configured to construct a network connection topology diagram based on the network configuration information, and aggregate network element ports.
  • FIG. 8 is a structural block of an energy consumption control device for network equipment according to an embodiment of the present invention.
  • the energy consumption control device 80 in addition to each module in Figure 6, the energy consumption control device 80 also includes: an algorithm optimization module 810, which is configured to calculate time based on the predicted traffic changes of the network equipment and the real traffic. The weight coefficient of the sequence prediction algorithm is dynamically adjusted.
  • Figure 9 is a structural block diagram of an energy consumption control device for network equipment according to an embodiment of the present invention.
  • the energy consumption control device 90 includes, in addition to various modules in Figure 8, It also includes: a decision execution module 910 configured to convert the energy-saving action decision into a communication message and send it to the corresponding network device for energy-saving operation.
  • each of the above modules can be implemented through software or hardware.
  • it can be implemented in the following ways, but is not limited to this: the above modules are all located in the same processor; or the above modules can be implemented in any combination.
  • the forms are located in different processors.
  • Embodiments of the present invention also provide a computer-readable storage medium that stores a computer program, wherein the computer program is configured to execute the steps in any of the above method embodiments when running.
  • the computer-readable storage medium may include but is not limited to: U disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM) , mobile hard disk, magnetic disk or optical disk and other media that can store computer programs.
  • ROM read-only memory
  • RAM random access memory
  • mobile hard disk magnetic disk or optical disk and other media that can store computer programs.
  • An embodiment of the present invention also provides an electronic device, including a memory and a processor.
  • a computer program is stored in the memory, and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.
  • the above-mentioned electronic device may further include a transmission device and an input-output device, wherein the transmission device is connected to the above-mentioned processor, and the input-output device is connected to the above-mentioned processor.
  • modules or steps of the present invention can be implemented using general-purpose computing devices, and they can be concentrated on a single computing device, or distributed across a network composed of multiple computing devices. They may be implemented in program code executable by a computing device, such that they may be stored in a storage device for execution by the computing device, and in some cases may be executed in a sequence different from that shown herein. or the described steps, or they are respectively made into individual integrated circuit modules, or multiple modules or steps among them are made into a single integrated circuit module. As such, the invention is not limited to any specific combination of hardware and software.
  • the embodiment of the present invention proposes an energy consumption control method for network equipment to predict the operating status of the network equipment and dynamically adjust the power consumption of the equipment through two-level service protection of the network and network elements.
  • the types of network equipment in the embodiment of the present invention include routers, switches, PTN/SPN and other equipment.
  • the threshold control method sets thresholds based on the volume of business transmitted by the device, and controls energy-saving actions based on one or more thresholds. .
  • This method needs to set thresholds and corresponding energy-saving behaviors in advance, has poor flexibility, lags in response, and cannot respond to business changes in the network in a timely manner. It is easy to cause business loss when the business volume changes drastically; for example, the external equipment control method is controlled by the management or Other control terminals that can communicate with the device predict the status of the network, issue energy-saving instructions to the device, and control it. Energy saving. This method relies on external commands, has delayed response, and has poor reliability. When network communication fails, it cannot receive a timely response, leading to increased losses.
  • Figure 10 is a network device according to a scenario embodiment of an embodiment of the present invention.
  • the network architecture block diagram for implementing the energy consumption control method is shown in Figure 10.
  • the network architecture includes a network-level sensing unit 1010 and a device-level control unit 1020.
  • the network-level sensing unit 1010 is deployed on the management and control side and is responsible for the overall group control.
  • the network service load information is sensed and calculated to sense the overall network situation and provide network-level supplementary information for device energy saving.
  • the network-level sensing unit 1010 includes:
  • the network awareness module 10101 is configured to obtain overall network information, including network element networking relationships, network element configuration information, and bearer service status in the network;
  • the network load prediction module 10102 constructs a network connection topology diagram based on the overall network information, automatically aggregates all network element ports based on the system location of the network element and the connection relationship between the network element and the network element, and automatically collects the network element ports according to the collected Based on the historical traffic volume of the network element, the time series prediction algorithm is used to predict the traffic volume that the network element may carry in the future. Furthermore, combined with the link protection relationship, it calculates the traffic volume that the network element on the protected link may carry in the future when service switching occurs. The maximum traffic volume is used as the protection bandwidth of the link.
  • the data transmission module 10103 is configured as a data transmission channel between control and equipment, responsible for data interaction with network elements, and delivers the predicted protection bandwidth to the equipment side in units of network elements.
  • the device-level control unit 1020 is deployed on the device side and is responsible for sensing internal information of the device, predicting business conditions and making specific decisions on energy-saving actions, including:
  • the information sensing module 10201 is set to obtain the operating status of the device, which is divided into two categories.
  • the first is device performance data, including port traffic information, CPU utilization, and memory utilization;
  • the second is device device status information, including but not limited to, Single board working status, chip working status, inter-board communication connection status, and device performance indicators.
  • the prediction module 10202 predicts changes in equipment business volume in the next time period through historical business volume data.
  • the prediction unit contains multiple time series prediction algorithms, including but not limited to machine learning time series prediction algorithms, neural network time series prediction algorithms, and algorithm
  • the model can be trained in real time or pre-trained in advance and loaded directly when used. Each algorithm has a certain weight coefficient, and the prediction results are calculated by weighting multiple algorithms.
  • Algorithm KPI assessment module 10203 When the algorithm model is running for a long time, due to changes in business data characteristics, the prediction accuracy of the algorithm model will decrease. In order to prevent the deterioration of the algorithm model from reducing the prediction accuracy, the embodiment of the present invention designs an algorithm KPI assessment module. 10203 is the KPI assessment and correction of the information prediction unit based on the comparison between predicted data and real data. After obtaining new business data from the information sensing module 10201, it is calculated with the data obtained from the last prediction, and by comparing the prediction accuracy of each algorithm, the degree of deterioration of the algorithm model is judged, and the weight coefficient of the algorithm is dynamically adjusted. Reduce the weight coefficient of the degraded algorithm model, and use the newly acquired business data to iteratively update the original model parameters to ensure that the algorithm can adapt to changes in business data.
  • the decision-making module 10204 is configured to conduct a comprehensive analysis of predicted business information, equipment operating status information, and network-level sensing information, plan business processing protection strategies, and make decisions that can save energy.
  • the above-mentioned protection strategy refers to the processing capacity reservation of equipment operating resources to prevent protection switching and service path recalculation caused by communication network failures, which will cause the network business volume to surge in a short period of time, exceeding the current equipment's ability to withstand capabilities, thereby causing business damage.
  • Figure 11 is a schematic diagram of the protection strategy principle according to the scenario embodiment of the present invention. As shown in Figure 11, the protection strategy includes equipment active and backup protection reservation, equipment balancing redundancy reservation, and network-level maximum service load reservation. level protection.
  • the equipment operation resources in the embodiment of the present invention include three levels of granularity: single board level, module level, and chip level.
  • a single board refers to a switching board that handles business volume distribution.
  • a single board contains multiple devices.
  • a module refers to a circuit system composed of multiple chips and connecting devices that jointly complete related functions.
  • a module contains multiple chips and devices. Refers to independent integrated chips in the circuit, including but not limited to processors, memories, and forwarding chips. The three granularities can complement each other to complete refined management.
  • active and backup protection refers to the use of the active active resource for the resources in the device and the standby resource that is switched in case of emergency.
  • the active resource When the active resource is abnormal, the standby resource can be switched to continue working to ensure the normal operation of the device.
  • Device balancing redundancy reservation means that the adjustable resources in the device are load balanced. Just like all ports on a single board will share the same processor, storage and other resources, the processing capabilities of this type of resources have greater redundancy for a single port. When the traffic on a single port increases abnormally, the impact on the overall performance of the device is small.
  • the maximum service load reservation at the network level is calculated by management and control. By analyzing the overall link status and business conditions of the network, the maximum load situation that may occur in the future of the network element equipment can be reserved in advance.
  • the action module 10205 is configured to convert the energy-saving action information obtained by the decision-making unit into corresponding communication messages and forward them to execution modules such as equipment electromechanical management and drive management.
  • the execution module 10206 is configured to receive energy saving messages, control corresponding device resources to perform power-off/sleep operations, and so on.
  • FIG. 12 is a flow chart of network-level information sensing and transmission according to a scenario embodiment of the present invention. As shown in Figure 12, the process includes the following steps:
  • Step S1201 obtain network configuration information
  • Network configuration information should exist in the managed asset data, including network element port information, network element location, connection relationships between network elements, network element system IDs, etc., as well as network performance data, including the business volume of each network element port.
  • Step S1202 Construct a network connection topology diagram based on the network configuration information, and aggregate network element ports;
  • FIG. 13 is a schematic diagram of the principle of network element port aggregation according to the scenario embodiment of the present invention. As shown in Figure 13, network elements A-F are on a unified link ring. For network element D, its links D-C-B-A and D-E-F-G-H are in a mutual protection relationship. , all ports on the link can be classified into the same set.
  • the basic principle for port aggregation is that the ports at both ends of the same link must belong to the same system. It is necessary to distinguish the levels of the network elements at both ends of a link. If the levels of the network elements at both ends of the link are the same, the two ports belong to the same set. If the levels of the network elements at both ends of the link are different, the levels of the network elements at both ends of the link need to be different. Network elements and peer network element information are divided, which will be explained in detail below:
  • the network port is divided into user network interface (UNI) and network node interface (Network to Network Interface, NNI). If the current port is on the UNI side, its port belonging set ID is the system ID of the network element where it is located. If the current port is on the NNI side and the opposite network element does not belong to any system, the port's belonging set ID is the system ID of this network element. If the current port is on the NNI side and the opposite network element belongs to only one system, the port's belonging set ID is the system ID of the opposite network element. If the current port is on the NNI side, and the peer network element belongs to multiple systems, and if the peer network element and the local network element have the same system ID, the port belonging set ID is the system ID, otherwise it is the peer network element system ID. .
  • UNI user network interface
  • NNI Network to Network Interface
  • Step S1203 Use a time series prediction algorithm to predict the traffic volume carried by the network device port and the maximum service volume of the port set.
  • the maximum service volume of the port set is used as the protection bandwidth of the device port, and the device protection bandwidth is obtained by summarizing the device port protection bandwidth;
  • a timing prediction algorithm is selected to predict the business load of each network element in the next period of time.
  • the interpolation method can be used, which can be used here to assign nearby data to the missing part of the data.
  • filter processing such as median filtering is used.
  • step signals which may include normal traffic surges, service switching, etc.
  • a nonlinear prediction algorithm is used to identify mutation points that appear in the sequence, making the prediction data more accurate.
  • the port service carrying future of the network element can be obtained.
  • the maximum port service load predicted in the port belonging set is used as the protection bandwidth of the port belonging set.
  • the protection bandwidth of all ports in the belonging set is equal to the protection bandwidth of the port belonging set.
  • Step S1204 Deliver the predicted service volume and maximum service volume carried by the network device to the network device;
  • the predicted protection bandwidth information is transmitted to the device based on the network element device.
  • FIG 14 is a flow chart of energy saving processing on the network device side according to a scenario embodiment of the present invention. As shown in Figure 14, the process includes the following steps:
  • Step S1401 obtain the device operation information of the network device
  • Step S1402 perform assessment and correction on the timing prediction algorithm
  • Figure 15 is a schematic diagram of the assessment and correction principle of the time series prediction algorithm according to the scenario embodiment of the present invention.
  • the algorithm KPI assessment is performed, and the new data and the previous prediction are calculated.
  • the mean square error of the predicted data Based on the size of the mean square error, identify whether the algorithm has deteriorated. For the deteriorated algorithm, there are two ways to deal with it. One is to reduce the weight of its prediction results, and the other is to use new data to incrementally train the algorithm model.
  • Step S1403 Predict changes in traffic volume of network equipment
  • the idle CPU core is decided, the prediction algorithm model instance is bound to the idle core, and then based on the previously collected single board load information, the timing prediction algorithm is used to predict the business size of the next stage.
  • a prediction algorithm can be used for prediction.
  • One or more time series algorithms are used to obtain the prediction results of each algorithm, and then combined with the weight coefficient, the final business volume prediction result is obtained by weighting the average.
  • Step S1404 Obtain the network protection bandwidth issued by the management and control to the device, and verify its rationality
  • the protection bandwidth should be no less than the actual traffic volume of the port and no more than the port's bearable bandwidth. If it exceeds this range, it needs to be classified within this range.
  • the actual business volume of the port can be used as the protection bandwidth of the port, and then based on The device port protection bandwidth is summarized to obtain the network-level protection bandwidth of the device.
  • Step S1405 Calculate the load required for the operation of network equipment in the future;
  • the load required for operation network-level protection bandwidth + device predicted load.
  • Step S1406 Calculate resource action information that needs to turn on/off energy saving based on the calculated future load condition of the network device.
  • the energy-saving action decision which specifically includes:
  • Step S1407 Notify the corresponding resource management module according to the energy-saving action information to perform resource energy-saving management.
  • the energy consumption control method of network equipment deployed by embodiments of the present invention deploys sensing and decision-making units within network elements, which can respond to changes in services carried in the network in the shortest time, and at the same time control the network with the help of management
  • the overall perception capability can prevent future traffic surges caused by possible network failures and ensure that losses due to energy conservation will not be further expanded in the event of network abnormalities.
  • the network element combines resource active and backup protection, equipment load balancing redundancy reservation, and network-level protection bandwidth three-level protection. It has sufficient room to absorb sudden business volume and ensures that energy-saving components are turned on before the business volume reaches the processing upper limit. .

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本发明实施例提供了一种网络设备的能耗控制方法及装置，通过根据网络配置信息，采用时间序列预测算法预测得到网络设备的保护带宽；根据网络设备运行状态信息，采用时间序列预测算法预测网络设备的业务量变化；根据预测的网络设备承载的业务量和最大业务量以及预测的网络设备的业务量变化，规划业务处理保护策略，获得对网络设备的节能动作决策。

Description

网络设备的能耗控制方法及装置 技术领域
本发明实施例涉及通信领域，具体而言，涉及一种网络设备的能耗控制方法及装置。
背景技术
目前，5G通信设备的能耗在运营商的开支中占比很大部分，在国家双碳政策节能减排的要求下，运营商已明确提出承载设备降功耗的需求，网络设备的功耗控制要求已成为其必备的能力之一。但是，节能的操作往往伴随着设备处理性能的下降，严重的甚至会导致网络业务的中断。现有的节能技术包括：人工决策法、阈值控制法和外部设备控制法，现有的节能技术均存在反应滞后，可靠性差，容易造成业务丢失甚至网络业务中断的情况。
发明内容
本发明实施例提供了一种网络设备的能耗控制方法及装置，以至少解决相关技术中的问题。
根据本发明的一个实施例，提供了一种网络设备的能耗控制方法，包括：根据网络配置信息，采用时间序列预测算法预测得到网络设备的保护带宽；根据网络设备运行状态信息，采用时间序列预测算法预测所述网络设备的业务量变化；根据预测的所述网络设备的保护带宽以及预测的所述网络设备的业务量变化，规划业务处理保护策略，获得对所述网络设备的节能动作决策。
根据本发明的另一个实施例，提供了一种网络设备的能耗控制装置，包括：预测业务量承载模块，设置为根据网络配置信息，采用时间序列预测算法预测得到网络设备的保护带宽；预测业务量变化模块，设置为根据网络设备运行状态信息，采用时间序列预测算法预测网络设备的业务量变化；节能决策模块，设置为根据预测的所述网络设备承载的业务量和最大业务量以及预测的所述网络设备的业务量变化，规划业务处理保护策略，获得节能动作决策。
根据本发明的又一个实施例，还提供了一种计算机可读存储介质，所述计算机可读存储介质中存储有计算机程序，其中，所述计算机程序被设置为运行时执行上述任一项方法实施例中的步骤。
根据本发明的又一个实施例，还提供了一种电子装置，包括存储器和处理器，所述存储器中存储有计算机程序，所述处理器被设置为运行所述计算机程序以执行上述任一项方法实施例中的步骤。
附图说明
图1是本发明实施例的一种网络设备的能耗控制方法的网络设备硬件结构框图；
图2是根据本发明实施例的网络设备的能耗控制方法的流程图；
图3是根据本发明实施例的网络设备的能耗控制方法的流程图；
图4是根据本发明实施例的网络设备的能耗控制方法的流程图；
图5是根据本发明实施例的网络设备的能耗控制方法的流程图；
图6是根据本发明实施例的网络设备的能耗控制装置的结构框图；
图7是根据本发明实施例的网络设备的能耗控制装置的结构框图；
图8是根据本发明实施例的网络设备的能耗控制装置的结构框图；
图9是根据本发明实施例的网络设备的能耗控制装置的结构框图；
图10是根据本发明场景实施例的一种网络设备的能耗控制方法实施的网络架构框图；
图11是根据本发明场景实施例的保护策略原理示意图；
图12是根据本发明场景实施例的网络级信息感知与传输的流程图；
图13是根据本发明场景实施例的网元端口归集原理示意图；
图14是根据本发明场景实施例的网络设备侧节能处理流程图；
图15是根据本发明场景实施例的时序预测算法考核修正原理示意图。
具体实施方式
下文中将参考附图并结合实施例来详细说明本发明的实施例。
需要说明的是，本发明的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。
本申请实施例中所提供的方法实施例可以在网络设备中执行。以运行在网络设备上为例，图1是本发明实施例的一种网络设备的能耗控制方法的网络设备的硬件结构框图。如图1所示，网络设备可以包括一个或多个(图1中仅示出一个)处理器102(处理器102可以包括但不限于微处理器MCU或可编程逻辑器件FPGA等的处理装置)和用于存储数据的存储器104，其中，上述网络设备还可以包括用于通信功能的传输设备106以及输入输出设备108。本领域普通技术人员可以理解，图1所示的结构仅为示意，其并不对上述网络设备的结构造成限定。例如，网络设备还可包括比图1中所示更多或者更少的组件，或者具有与图1所示不同的配置。
存储器104可用于存储计算机程序，例如，应用软件的软件程序以及模块，如本发明实施例中的网络设备的能耗控制方法对应的计算机程序，处理器102通过运行存储在存储器104内的计算机程序，从而执行各种功能应用以及数据处理，即实现上述的方法。存储器104可包括高速随机存储器，还可包括非易失性存储器，如一个或者多个磁性存储装置、闪存、或者其他非易失性固态存储器。在一些实例中，存储器104可进一步包括相对于处理器102远程设置的存储器，这些远程存储器可以通过网络连接至管理终端。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
传输装置106用于经由一个网络接收或者发送数据。上述的网络具体实例可包括网络设备的通信供应商提供的有线网络。在一个实例中，传输装置106包括一个网络适配器(Network Interface Controller，简称为NIC)，其可通过光纤与其他网络设备相连从而可与互联网进行通讯。在一个实例中，传输装置106可以为光模块，其用于通过有线方式与互联网进行通讯。
在本实施例中提供了一种运行于上述网络设备的网络设备的能耗控制方法，图2是根据本发明实施例的网络设备的能耗控制方法的流程图，如图2所示，该流程包括如下步骤：
步骤S202，根据网络配置信息，采用时间序列预测算法预测得到网络设备的保护带宽；
步骤S204，根据网络设备运行状态信息，采用时间序列预测算法预测网络设备的业务量变化；
步骤S206，根据预测的所述网络设备的保护带宽以及预测的所述网络设备的业务量变化，规划业务处理保护策略，获得对所述网络设备的节能动作决策。
通过上述步骤通过根据网络配置信息，采用时间序列预测算法预测得到网络设备的保护带宽；根据所述网络设备运行状态信息，采用时间序列预测算法预测所述网络设备的业务量变化；根据预测的所述网络设备承载的业务量和最大业务量以及预测的所述网络设备的业务量变化，规划业务处理保护策略，获得对所述网络设备的节能动作决策，解决了相关技术中反应滞后，可靠性差，网络业务容易中断的问题，达到了提高网络业务处理的灵敏度和可靠性的效果。
其中，上述步骤的执行主体可以为基站、终端、承载传输设备等，但不限于此。
其中，在本发明实施例中，采用时间序列预测算法预测得到网络设备的保护带宽的过程为：采用时间序列预测算法预测网络设备端口承载的业务量和端口集最大业务量，以所述端口集最大业务量作为网络设备端口的保护带宽，根据网络设备端口的保护带宽获取网络设备的保护带宽。其中，将网络设备端口的保护带宽汇总相加即可得到网络设备的保护带宽。
本发明实施例中，各个网络设备的网元端口进行归集可以获得对应的端口集。
在一个示例性实施例中，在采用时间序列预测算法预测网络设备承载的业务量和最大业务量之前，还包括：获取网络配置信息；根据网络配置信息构建网络连接拓扑图，并对网元端口进行归集。图3是根据本发明实施例的网络设备的能耗控制方法的流程图，如图3所示，该流程包括如下步骤：
步骤S302，获取网络配置信息；
步骤S304，根据网络配置信息构建网络连接拓扑图，并对网元端口进行归集；
步骤S306，根据网络配置信息，采用时间序列预测算法预测得到网络设备的保护带宽；
步骤S308，根据网络设备运行状态信息，采用时间序列预测算法预测网络设备的业务量变化；
步骤S310，根据预测的所述网络设备的保护带宽以及预测的所述网络设备的业务量变化，规划业务处理保护策略，获得对所述网络设备的节能动作决策。
在一个示例性实施例中，网络设备运行状态信息包括：网络设备性能数据和网络设备器件状态信息；其中，网络设备性能数据包括以下至少之一：端口业务量信息，CPU利用率，内存利用率；网络设备器件状态信息包括以下至少之一：单板工作状态，芯片工作状态，板间通信连接状态，器件的性能指标。
其中，本领域的普通技术人员应该知道，上述网络设备运行状态信息的内容是做举例说明，在具体实施过程中，网络设备运行状态信息的内容可以根据实际情况进行调整，这里不做限制。
在一个示例性实施例中，在采用时间序列预测算法预测网络设备的业务量变化之后，还包括：根据预测的网络设备业务量变化与真实业务量，对时间序列预测算法的权重系数进行动态调整。图4是根据本发明实施例的网络设备的能耗控制方法的流程图，如图4所示，该流程包括如下步骤：
步骤S402，根据网络配置信息，采用时间序列预测算法预测得到网络设备的保护带宽；
步骤S404，根据网络设备运行状态信息，采用时间序列预测算法预测网络设备的业务量变化；
步骤S406，根据预测的网络设备业务量变化与真实业务量，对时间序列预测算法的权重系数进行动态调整；
步骤S408，根据预测的网络设备承载的保护带宽以及预测的网络设备的业务量变化，规划业务处理保护策略，获得对网络设备的节能动作决策。
在一个示例性实施例中，保护策略包括以下至少之一：网络设备主备保护预留；网络设备均衡冗余预留；网络级最大业务负载预留。
在一个示例性实施例中，规划业务处理保护策略，获得节能动作决策，包括：依次计算设备可处理负载的单板数量、模块数量、芯片数量，将计算结果与设备运行资源状态比较获得所述节能动作决策。
在一个示例性实施例中，在规划业务处理保护策略，获得节能动作决策之后，还包括：将节能动作决策转化为通信消息下发至对应设备进行节能操作。图5是根据本发明实施例的网络设备的能耗控制方法的流程图，如图5所示，该流程包括如下步骤：
步骤S502，根据网络配置信息，采用时间序列预测算法预测得到网络设备的保护带宽；
步骤S504，根据网络设备运行状态信息，采用时间序列预测算法预测网络设备的业务量变化；
步骤S506，根据预测的网络设备业务量变化与真实业务量，对时间序列预测算法的权重系数进行动态调整；
步骤S508，根据预测的网络设备承载的保护带宽以及预测的网络设备的业务量变化，规划业务处理保护策略，获得对网络设备的节能动作决策。
步骤S510，将节能动作决策转化为通信消息下发至对应设备进行节能操作。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到根据上述实施例的方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端设备(可以是手机，计算机，服务器，或者网络设备等)执行本发明各个实施例所述的方法。
在本实施例中还提供了一种网络设备的能耗控制装置，该装置用于实现上述实施例及优选实施方式，已经进行过说明的不再赘述。如以下所使用的，术语“模块”可以实现预定功能的软件和/或硬件的组合。尽管以下实施例所描述的装置较佳地以软件来实现，但是硬件，或者软件和硬件的组合的实现也是可能并被构想的。
图6是根据本发明实施例的网络设备的能耗控制装置的结构框图，如图6所示，该能耗控制装置60包括：预测保护带宽模块610，设置为根据网络配置信息，采用时间序列预测算法预测网络设备承载的业务量和最大业务量，以最大业务量作为链路的保护带宽；预测业务量变化模块620，设置为根据网络设备运行状态信息，采用时间序列预测算法预测网络设备的业务量变化；节能决策模块630，设置为根据预测的网络设备承载的业务量和最大业务量以及预测的网络设备的业务量变化，规划业务处理保护策略，获得节能动作决策。
在一个示例性实施例中，图7是根据本发明实施例的网络设备的能耗控制装置的结构框图，如图7所示，该能耗控制装置70除了包括图6中的各个模块外，还包括：配置信息获取模块710，设置为获取网络配置信息；端口归集模块720，设置为根据网络配置信息构建网络连接拓扑图，并对网元端口进行归集。
在一个示例性实施例中，图8是根据本发明实施例的网络设备的能耗控制装置的结构框 图，如图8所示，该能耗控制装置80除了包括图6中的各个模块外，还包括：算法优化模块810，设置为根据预测的网络设备的业务量变化与真实业务量，对时间序列预测算法的权重系数进行动态调整。
在一个示例性实施例中，图9是根据本发明实施例的网络设备的能耗控制装置的结构框图，如图9所示，该能耗控制装置90除了包括图8中的各个模块外，还包括：还包括：决策执行模块910，设置为将节能动作决策转化为通信消息下发至对应网络设备进行节能操作。
需要说明的是，上述各个模块是可以通过软件或硬件来实现的，对于后者，可以通过以下方式实现，但不限于此：上述模块均位于同一处理器中；或者，上述各个模块以任意组合的形式分别位于不同的处理器中。
本发明的实施例还提供了一种计算机可读存储介质，该计算机可读存储介质中存储有计算机程序，其中，该计算机程序被设置为运行时执行上述任一项方法实施例中的步骤。
在一个示例性实施例中，上述计算机可读存储介质可以包括但不限于：U盘、只读存储器(Read-Only Memory，简称为ROM)、随机存取存储器(Random Access Memory，简称为RAM)、移动硬盘、磁碟或者光盘等各种可以存储计算机程序的介质。
本发明的实施例还提供了一种电子装置，包括存储器和处理器，该存储器中存储有计算机程序，该处理器被设置为运行计算机程序以执行上述任一项方法实施例中的步骤。
在一个示例性实施例中，上述电子装置还可以包括传输设备以及输入输出设备，其中，该传输设备和上述处理器连接，该输入输出设备和上述处理器连接。
本实施例中的具体示例可以参考上述实施例及示例性实施方式中所描述的示例，本实施例在此不再赘述。
显然，本领域的技术人员应该明白，上述的本发明的各模块或各步骤可以用通用的计算装置来实现，它们可以集中在单个的计算装置上，或者分布在多个计算装置所组成的网络上，它们可以用计算装置可执行的程序代码来实现，从而，可以将它们存储在存储装置中由计算装置来执行，并且在某些情况下，可以以不同于此处的顺序执行所示出或描述的步骤，或者将它们分别制作成各个集成电路模块，或者将它们中的多个模块或步骤制作成单个集成电路模块来实现。这样，本发明不限制于任何特定的硬件和软件结合。
为了使得本领域的技术人员更好地理解本发明的技术方案，下面结合具体的场景实施例进行阐述。
场景实施例一
本发明实施例提出了一种网络设备的能耗控制方法，以对网络设备运行状态进行预测，通过网络加网元两级业务保护，对设备的功耗进行动态调节的方法。其中，本发明的实施例中的网络设备种类包括路由器、交换机以及PTN/SPN等设备。在通信领域中已存在部分节能技术，比如人工决策法，依赖人工决策出无业务承载的设备资源，将其手动下电，需要配置业务时再手动恢复。该方式依赖人为决策，对运维人员要求高，不适合大规模展开，且可能存在人为误操作情况。且可控制部件少，只能进行运维层面的可配置操作，无法做到精细化管理；比如阈值控制法，根据设备传输的业务量大小，设定阈值，根据一个或多个阈值控制节能动作。该方式需要提前设定好阈值和对应节能行为，灵活性较差，反应滞后，无法及时应对网络中业务变化，在业务量变化剧烈情况下易造成业务丢失；比如外部设备控制法，由管控或其他可与设备通信的控制端，对网络中状态进行预测，下发节能指令至设备，控制其 节能。该方式依赖外部的命令，反应滞后，可靠性差，当网络通信发生故障时无法及时得到响应，从而引发损失的扩大。
为了克服现有技术中存在的能耗控制实时性差，无法应对业务量剧烈变化的问题和缺陷，本发明实施例采用以下技术方案，图10是根据本发明实施例场景实施例的一种网络设备的能耗控制方法实施的网络架构框图，如图10所示，该网络架构包括网络级感知单元1010及设备级控制单元1020，其中，网络级感知单元1010，部署在管控侧，负责对整体组网业务负载信息进行感知和运算，对网络整体情况进行感知，为设备节能提供网络级别补充信息，网络级感知单元1010包括：
网络感知模块10101，设置为获取网络整体信息，包含网元的组网关系，网元配置信息，网络中的承载业务状态；
网络负载预测模块10102，根据网络整体信息构建网络连接拓扑图，根据网元所处的系统位置，及网元与网元的连接关系，自动的对所有网元端口进行归集，以及根据采集的网元历史业务量，利用时间序列预测算法，预测未来一段时间网元可能承载的业务量，进一步的，结合链路保护关系，计算在发生业务切换时，其保护链路上网元未来可能承载的最大业务量，作为该链路的保护带宽。
数据传输模块10103，设置为作为指管控与设备的数据传输通道，负责与网元的数据交互，以网元为单位将预测得到的保护带宽下发至设备侧。
设备级控制单元1020，部署在设备侧，负责设备内部信息的感知，业务情况预测和节能动作的具体决策，包括：
信息感知模块10201，设置为获取设备运行状态，分为两大类，一是设备性能数据，包括端口业务量信息，CPU利用率、内存利用率；二是设备器件状态信息，包括但不仅限于，单板工作状态、芯片工作状态、板间通信连接状态、器件的性能指标。
预测模块10202，通过历史业务量数据，预测下一时间段设备业务量变化情况，该预测单元包含多个时间序列预测算法，包括但不仅限于机器学习时序预测算法、神经网络时序预测算法，算法的模型可以是实时训练得到，也可以是提前预训练好，待使用时直接加载，每种算法都有一定权重系数，预测结果以多个算法加权计算得出。
算法KPI考核模块10203，算法模型在长期运行时，由于业务数据特征的变化，会导致算法模型预测准确度降低，为了防止算法模型劣化导致预测准确度降低，本发明实施例设计了算法KPI考核模块10203，是根据预测数据和真实数据的对比，对信息预测单元的KPI考核和修正。从信息感知模块10201中得到新的业务数据后，与上次预测得到的数据计算，通过比对每一种算法的预测准确性，判断算法模型的劣化程度，对算法的权重系数进行动态调整，降低劣化算法模型的权重系数，同时利用新获取的业务数据，对原有模型参数进行迭代更新，以保证算法可以适应业务数据变化。
决策模块10204，设置为对预测业务信息、设备运行状态信息、网络级感知信息进行综合分析，规划好业务处理保护策略，做出可节能动作的决策。
其中，上述保护策略，是指对设备运行资源的处理能力预留，以防止通信网络因故障引起的保护倒换、业务路径重计算等状况，导致网络业务量短时间内飙升，突破当前设备可承受能力，从而造成业务损伤。图11是根据本发明场景实施例的保护策略原理示意图，如图11所示，保护策略包含设备主备保护预留、设备均衡冗余预留、网络级最大业务负载预留三 级保护。
其中，本发明实施例中的设备运行资源，包含单板级、模块级、芯片级三种颗粒度。其中，单板是指处理业务量分发的交换板，一块单板包含多个器件，模块是指由多个芯片及连接器件组成、共同完成相关功能的电路系统，一个模块包含多个芯片，器件是指电路中独立的集成芯片，包括但不限于处理器、存储器、转发芯片。三种颗粒度可以相互补充，从而完成精细化管理。
其中，主备保护是指对于设备内的资源，采用工作的主用资源，及应急时切换的备用资源，在主用资源发生异常时，可切换至备用资源继续工作，以保证设备正常运行。设备均衡冗余预留是指在设备中可调节的资源是负载均衡的，如同单板中所有端口会共用相同的处理器、存储等资源，该类资源处理能力对于单个端口有较大冗余量，在单个端口业务发生异常升高时，对设备整体性能的冲击较小。网络级最大业务负载预留是指由管控计算得到，通过对网络整体的链路状态和业务情况，分析网元设备未来可能出现的最大负载情况，提前对其进行预留。
动作模块10205，设置为将决策单元得到的节能动作信息转化成对应的通信消息，转发至设备机电管理、驱动管理等执行模块。
执行模块10206，设置为接收节能的消息，控制对应设备资源进行下电/休眠操作等等。
场景实施例二
以PTN产品为例，利用上述场景实施例一提出的网络架构实施。图12是根据本发明场景实施例的网络级信息感知与传输的流程图，如图12所示，该流程包括以下步骤：
步骤S1201，获取网络配置信息；
管控资产数据中应存在网络配置信息，包括网元端口信息、网元所处位置、网元之间的连接关系、网元系统id等，以及网络性能数据，包括各网元端口业务量。
步骤S1202，根据网络配置信息构建网络连接拓扑图，并对网元端口进行归集；
查阅网元资产，获取已开启节能功能的网元，并查找每个开启节能的网元所对应端口信息，根据端口链接，得到网元链路拓扑关系，再根据网络链路拓扑结构，对所有设备端口进行归集。图13是根据本发明场景实施例的网元端口归集原理示意图，如图13所示，网元A-F为处于统一链路环上，对于D网元，其链路D-C-B-A与D-E-F-G-H互为保护关系，该链路上所有端口则可以归为同一集合。
对端口进行归集的基本原则是，同一链路两端端口一定属于同一系统。需对于一个链路两端的网元的层次进行区分，如链路两端网元层次相同，则该两路两个端口属于同一个集合，如链路两端网元层次不同，需要根据本端网元以及对端网元信息进行划分，下面将进行详细说明：
将网络端口分为用户侧网络接口(user network interface，UNI)和网络侧节点接口(Network to Network Interface，NNI)。若当前端口为UNI侧，则其端口归属集ID即为其所在的网元系统ID。若当前端口为NNI侧，且对端网元不属于任何系统，则该端口归属集ID为本网元系统ID。若当前端口为NNI侧，且对端网元只属于一个系统，则该端口归属集ID为对端网元系统ID。若当前端口为NNI侧，且对端网元属于多个系统，若对端网元与本端网元有相同系统ID，则端口归属集ID为该系统ID，否则为对端网元系统ID。
步骤S1203，采用时间序列预测算法预测网络设备端口承载的业务量和端口集最大业务 量，以端口集最大业务量作为设备端口的保护带宽，由设备端口保护带宽汇总得到设备保护带宽；
根据历史负载信息，选择时序预测算法，对每个网元下一段时间业务负载情况分别进行预测。在实际网络业务情况复杂、负载存在不确定的情况下，数据采集过程可能的数据丢失，体现在统计的序列数据中表现即为较多的毛刺阶跃信号和缺失值，为了保证较好的预测效果，需在预测前对原始信号进行预处理。对于缺失信号，可采用插值方法，这里可采用，将附近数据赋值为缺失部分数据。对于毛刺信号，采用滤波处理如中值滤波。对于阶跃信号，可能会正常的流量突增，业务切换等，采用非线性预测算法，识别序列中出现的突变点，使得预测数据更为准确。通过上述预处理及时间序列预测算法，可以得到网元的端口业务承载未来。结合步骤S1202得到的端口归属集，将端口归属集中预测的最大端口业务负载，作为该端口归属集的保护带宽，归属集中所有端口的保护带宽都等于该端口归属集的保护带宽。
步骤S1204：将预测的网络设备承载的业务量和最大业务量下发至网络设备；
由步骤S1203中得到的设备保护带宽，以网元设备为单位，将预测得到的保护带宽信息传输至设备。
场景实施例三
以PTN产品为例，利用上述场景实施例一提出的网络架构实施。图14是根据本发明场景实施例的网络设备侧节能处理流程图，如图14所示，该流程包括以下步骤：
步骤S1401，获取网络设备的设备运行信息；
获取网络设备的设备运行信息，包括CPU利用率、内存利用率及单板负载信息；
步骤S1402，对时序预测算法进行考核修正；
图15是根据本发明场景实施例的时序预测算法考核修正原理示意图，如图15所示，当新一轮数据更新时，若有上次预测数据，则进行算法KPI考核，计算新数据与上次预测数据的均方差。根据均方差大小，识别算法是否已劣化，对于已劣化的算法，有两种处理方式，一是降低其预测结果的权重，二是利用新数据，对算法模型进行增量训练。
步骤S1403：预测网络设备的业务量变化；
根据CPU利用率，决策出空闲CPU核，绑定预测算法模型实例至空闲核中，再根据先前采集的单板负载信息，采用时序预测算法预测下一阶段业务大小，其中，预测时可采用一种或多种时序算法，得到每种算法的预测结果，再结合权重系数，加权平均得到最终的业务量预测结果。
步骤S1404：获取管控下发至设备的网络保护带宽，并对其合理性进行校验；
保护带宽应不小于该端口实际业务量，不大于端口可承受带宽，若超出该范围，需归结到该范围内，在具体实施例中，可以将端口实际业务量作为端口的保护带宽，再根据设备端口保护带宽，汇总得到设备的网络级保护带宽。
步骤S1405：计算未来一段时间内网络设备运行所需承担负载；
运行所需承担负载＝网络级保护带宽+设备预测负载。
步骤S1406：根据计算的网络设备未来负载情况，计算需要开启/关闭节能的资源动作信息。
其中，不同级别设备资源节能计算时，依次计算设备可处理负载的单板数量、模块数量、芯片数量，将计算结果与设备运行资源状态比较获得所述节能动作决策，具体包括：
(1)依次选择N块单板，使得当N＞2时，N块单板可处理负载总和＜＝运行所需承担负载，N+1块单板可处理负载总和＞运行所需承担负载。这N块单板可处理负载总和记为An。
(2)从第N+1块单板依次选择M个模块，使得M个模块可处理负载总和+An＜＝运行所需承担负载，使M+1个模块可处理负载总和+An＞运行所需承担负载。这M个模块可处理负载总和记为Bn。
(3)从第M+1个模块中依次选择0块芯片，使得0-1个芯片可处理负载总和+An+Bn＜＝运行所需承担负载，使0个模块可处理负载总和+An+Bm＞运行所需承担负载。
(4)得到设备处理负载所需资源：N个单板+M个模块+0个芯片，对于剩余的单板、模块、芯片，加入节能标识，即可进行节能操作，将本次计算得到所需节能器件，与已设备资源运行状态进行比较，最终得到需要开启/关闭节能的资源动作信息。
步骤S1407：根据节能动作信息，通知相应的资源管理模块，进行资源节能管理。
综上，本发明实施例提供的网络设备的能耗控制方法，与现有技术相比，在网元内部署感知和决策单元，可以最短时间响应网络中承载的业务变化，同时借助管控对网络整体的感知能力，对未来网络可能出现故障导致流量激增进行预防，保障在网络异常下不会因节能导致损失进一步扩大。网元结合了资源主备保护、设备负载均摊冗余预留、网络级保护带宽三级保护，对突发业务量有充足的吸收空间，保证在业务量到达处理上限之前，开启已节能的部件。
以上所述仅为本发明的优选实施例而已，并不用于限制本发明，对于本领域的技术人员来说，本发明可以有各种更改和变化。凡在本发明的原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。

Claims (14)

  1. 一种网络设备的能耗控制方法，包括：
    根据网络配置信息，采用时间序列预测算法预测得到网络设备的保护带宽；
    根据网络设备运行状态信息，采用时间序列预测算法预测所述网络设备的业务量变化；
    根据预测的所述网络设备的保护带宽以及预测的所述网络设备的业务量变化，规划业务处理保护策略，获得对所述网络设备的节能动作决策。
  2. 根据权利要求1所述的方法，其中，在所述采用时间序列预测算法预测网络设备的保护带宽之前，还包括：
    获取网络配置信息；
    根据网络配置信息构建网络连接拓扑图，并对网元端口进行归集。
  3. 根据权利要求1所述的方法，其中，所述采用时间序列预测算法预测得到网络设备的保护带宽，包括：
    采用时间序列预测算法预测网络设备端口承载的业务量和端口集最大业务量，以所述端口集最大业务量作为网络设备端口的保护带宽，根据网络设备端口的保护带宽获取网络设备的保护带宽。
  4. 根据权利要求1所述的方法，其中，所述网络设备运行状态信息包括：网络设备性能数据和网络设备器件状态信息；
    其中，所述网络设备性能数据包括以下至少之一：端口业务量信息，CPU利用率，内存利用率；所述网络设备器件状态信息包括以下至少之一：单板工作状态，芯片工作状态，板间通信连接状态，器件的性能指标。
  5. 根据权利要求1所述的方法，其中，在所述采用时间序列预测算法预测网络设备的业务量变化之后，还包括：
    根据预测的所述网络设备业务量变化与真实业务量，对所述时间序列预测算法的权重系数进行动态调整。
  6. 根据权利要求1所述的方法，所述保护策略包括以下至少之一：
    网络设备主备保护预留；
    网络设备均衡冗余预留；
    网络级最大业务负载预留。
  7. 根据权利要求6所述的方法，其中，所述规划业务处理保护策略，获得节能动作决策，包括：
    依次计算设备可处理负载的单板数量、模块数量、芯片数量，将计算结果与设备运行资源状态比较获得所述节能动作决策。
  8. 根据权利要求1所述的方法，其中，在所述规划业务处理保护策略，获得节能动作决策之后，还包括：
    将所述节能动作决策转化为通信消息下发至对应设备进行节能操作。
  9. 一种网络设备的能耗控制装置，包括：
    预测保护带宽模块，设置为根据网络配置信息，采用时间序列预测算法预测得到网络设备的保护带宽；
    预测业务量变化模块，设置为根据网络设备运行状态信息，采用时间序列预测算法预测网络设备的业务量变化；
    节能决策模块，设置为根据预测的所述网络设备的保护带宽以及预测的所述网络设备的业务量变化，规划业务处理保护策略，获得节能动作决策。
  10. 根据权利要求9所述的装置，其中，还包括：
    配置信息获取模块，设置为获取网络配置信息；
    端口归集模块，设置为根据网络配置信息构建网络连接拓扑图，并对网元端口进行归集。
  11. 根据权利要求9所述的装置，其中，还包括：
    算法优化模块，设置为根据预测的所述网络设备的业务量变化与真实业务量，对所述时间序列预测算法的权重系数进行动态调整。
  12. 根据权利要求9所述的装置，其中，还包括：
    决策执行模块，设置为将所述节能动作决策转化为通信消息下发至对应网络设备进行节能操作。
  13. 一种计算机可读存储介质，所述计算机可读存储介质中存储有计算机程序，其中，所述计算机程序被处理器执行时实现所述权利要求1至8任一项中所述的方法。
  14. 一种电子装置，包括存储器、处理器以及存储在所述存储器上并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现所述权利要求1至8任一项中所述的方法。
PCT/CN2023/087055 2022-08-24 2023-04-07 网络设备的能耗控制方法及装置 WO2024040976A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211021645.4 2022-08-24
CN202211021645.4A CN117675580A (zh) 2022-08-24 2022-08-24 网络设备的能耗控制方法及装置

Publications (1)

Publication Number Publication Date
WO2024040976A1 true WO2024040976A1 (zh) 2024-02-29

Family

ID=90012324

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/087055 WO2024040976A1 (zh) 2022-08-24 2023-04-07 网络设备的能耗控制方法及装置

Country Status (2)

Country Link
CN (1) CN117675580A (zh)
WO (1) WO2024040976A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102739507A (zh) * 2011-04-14 2012-10-17 中国科学院声学研究所 一种承载状态感知的路由器及其业务流带宽分配方法
WO2018027226A1 (en) * 2016-08-05 2018-02-08 Fractal Industries, Inc. Detection mitigation and remediation of cyberattacks employing an advanced cyber-decision platform
CN109639470A (zh) * 2018-11-30 2019-04-16 四川安迪科技实业有限公司 基于星型组网的vsat卫星通信系统带宽分配方法
CN112367276A (zh) * 2020-12-31 2021-02-12 南京群顶科技有限公司 基于网络流量优先级的网络资源动态自适应方法及系统
CN114118567A (zh) * 2021-11-19 2022-03-01 国网河南省电力公司经济技术研究院 一种基于双通路融合网络的电力业务带宽预测方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102739507A (zh) * 2011-04-14 2012-10-17 中国科学院声学研究所 一种承载状态感知的路由器及其业务流带宽分配方法
WO2018027226A1 (en) * 2016-08-05 2018-02-08 Fractal Industries, Inc. Detection mitigation and remediation of cyberattacks employing an advanced cyber-decision platform
CN109639470A (zh) * 2018-11-30 2019-04-16 四川安迪科技实业有限公司 基于星型组网的vsat卫星通信系统带宽分配方法
CN112367276A (zh) * 2020-12-31 2021-02-12 南京群顶科技有限公司 基于网络流量优先级的网络资源动态自适应方法及系统
CN114118567A (zh) * 2021-11-19 2022-03-01 国网河南省电力公司经济技术研究院 一种基于双通路融合网络的电力业务带宽预测方法

Also Published As

Publication number Publication date
CN117675580A (zh) 2024-03-08

Similar Documents

Publication Publication Date Title
CN108259367B (zh) 一种基于软件定义网络的服务感知的流策略定制方法
US9686125B2 (en) Network reliability assessment
US20180365581A1 (en) Resource-aware call quality evaluation and prediction
US20210281492A1 (en) Determining context and actions for machine learning-detected network issues
CN113741282A (zh) 一种基于边缘计算的设备智能化管理系统及方法
CN103179046A (zh) 基于openflow的数据中心流量控制方法及系统
CN110891283A (zh) 一种基于边缘计算模型的小基站监控装置及方法
US10285108B1 (en) Proactive roaming handshakes based on mobility graphs
US10680919B2 (en) Eliminating bad rankers and dynamically recruiting rankers in a network assurance system
WO2021244247A1 (zh) 转发数据报文的方法、网络节点、系统及存储介质
US9857825B1 (en) Rate based failure detection
CN104601375A (zh) 一种智能变电站通信网络监控系统、整定方法及监控方法
CN105262624A (zh) 一种基于软件定义网络的异构配电通信网络架构
WO2024040976A1 (zh) 网络设备的能耗控制方法及装置
CN114138750B (zh) 一种基于ai咨询数据库集群搭建方法和系统
US20210234795A1 (en) Systems & methods for detecting communication link breaks
Savchenko et al. Computer networks monitoring and management methods
US10498633B2 (en) Traffic activity-based signaling to adjust forwarding behavior of packets
CN117938750B (zh) 调度路由信息的处理方法、装置、设备、存储介质及产品
KR102537023B1 (ko) 인공지능 기반의 트래픽 분석을 사용한 네트워크 트래픽 제어 방법 및 이러한 방법을 수행하는 장치
CN116436806B (zh) 算网资源编排方法、装置、电子设备及计算机存储介质
CN112039685B (zh) 网络管理方法、装置、计算网络和物联网
CN114626742A (zh) 电力业务智能调度方法、服务器及存储介质
CN118316881A (zh) 一种大数据网络通讯协调方法及系统
Khapre et al. Intelligent Adaptive Edge System: Evaluation and Open Issues

Legal Events

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

Ref document number: 23856071

Country of ref document: EP

Kind code of ref document: A1