A tunnel traffic calculation method, device and electronic equipment

Abstract

Embodiments of the present application disclose a tunnel traffic calculation method and device and electronic equipment, the method comprises: obtaining link load between a plurality of second network units; obtaining the total traffic between the plurality of second network units; determining abnormal information according to the link load and the total traffic; screening out the abnormal information to obtain a traffic matrix; obtaining entity channel traffic between the plurality of second network units; determining tunnel traffic between each second network unit according to the traffic matrix and the entity channel traffic. The scheme of the embodiments of the present application can improve the accuracy of tunnel traffic measurement, ensure the monitoring and management quality of tunnel traffic, and improve the user experience.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a method, apparatus, and electronic device for calculating tunnel flow. Background Technology

[0002] With the development and widespread adoption of fourth-generation (4G) and fifth-generation (5G) wireless communication technologies, and the increasing scale of networks, data traffic monitoring plays an increasingly important role in network design and management. To meet the demands of network-side bandwidth deployment and dynamic traffic allocation, it is necessary to monitor, manage, and predict tunnel traffic based on Segment Routing Best Effort (SRBE) and Label Distribution Protocol (LDP), thereby enabling the management, dynamic optimization, and scaling recommendations for the entire 4G and 5G network traffic.

[0003] Since SRBE and LDP tunnels are invisible tunnels between devices within the same Interior Gateway Protocol (IGP) in a network, they are automatically formed within the IGP ring or chain based on the IGP's shortest path algorithm after enabling Segment Routing (SR) and LDP protocols respectively. These tunnels are characterized by path instability and unavailability of traffic. Currently, related technologies calculate information entropy based on network link traffic and use the expectation-maximization algorithm to estimate the traffic demand between the source and destination nodes (OD). However, this estimation method has significant errors in actual measurements, greatly affecting the accuracy of tunnel traffic measurement. Summary of the Invention

[0004] The following is an overview of the subject matter described in detail herein. This overview is not intended to limit the scope of the claims.

[0005] This invention provides a method, apparatus, and electronic device for calculating tunnel flow, which can improve the accuracy of tunnel flow measurement, ensure the quality of tunnel flow monitoring and management, and enhance the user experience.

[0006] In a first aspect, embodiments of the present invention provide a tunnel traffic calculation method, the method comprising: obtaining link load between a plurality of second network units; obtaining the total traffic between the plurality of second network units; determining abnormal information based on the link load and the total traffic; filtering out the abnormal information to obtain a traffic matrix; obtaining physical channel traffic between the plurality of second network units; and determining tunnel traffic between each of the second network units based on the traffic matrix and the physical channel traffic.

[0007] In a second aspect, embodiments of the present invention provide a tunnel traffic calculation device, comprising: an acquisition module for acquiring link load, total traffic, and physical channel traffic between a first network unit and a plurality of second network units; a calculation module for determining abnormal information based on the link load and the total traffic; a filtering module for filtering out the abnormal information to obtain a traffic matrix; and a determination module for determining tunnel traffic between each of the second network units based on the traffic matrix and the physical channel traffic.

[0008] Thirdly, embodiments of the present invention provide an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the tunnel traffic calculation method provided by the embodiments of the present invention.

[0009] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the tunnel flow calculation method provided in embodiments of the present invention.

[0010] This invention provides an embodiment that obtains the link load, total traffic, and physical channel traffic between a first network unit and multiple second network units; determines abnormal information based on the link load and total traffic; filters out the abnormal information to obtain a traffic matrix; and determines the tunnel traffic between the first network unit and each of the second network units based on the traffic matrix and the physical channel traffic. This invention's solution can accurately calculate the tunnel traffic between the first network unit and each of the second network units, facilitating accurate prediction of network traffic trends and reasonable network optimization, thus improving user experience. It is particularly suitable for data transmission scenarios such as backbone networks (BN), IPRAN, and Slicing Packet Networks (SPN).

[0011] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description, claims and drawings. Attached Figure Description

[0012] The accompanying drawings are provided to further understand the technical solutions of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the technical solutions of the present invention, and do not constitute a limitation on the technical solutions of the present invention.

[0013] Figure 1 This is a schematic flowchart of a tunnel flow calculation method provided in an embodiment of the present invention;

[0014] Figure 2 yes Figure 1 A schematic diagram illustrating the specific implementation process of another embodiment of step S2000;

[0015] Figure 3 yes Figure 2 A schematic diagram illustrating the specific implementation process of another embodiment of step S2200;

[0016] Figure 4 yes Figure 1 A schematic diagram illustrating the specific implementation process of another embodiment of step S3000;

[0017] Figure 5 yes Figure 4 A schematic diagram illustrating the specific implementation process of another embodiment of step S3100;

[0018] Figure 6 yes Figure 4 A schematic diagram illustrating the specific implementation process of another embodiment of step S3200;

[0019] Figure 7 yes Figure 4 A schematic diagram illustrating the specific implementation process of another embodiment of step S3300;

[0020] Figure 8 yes Figure 1 A schematic diagram illustrating the specific implementation process of another embodiment of step S4000;

[0021] Figure 9 yes Figure 8 A schematic diagram illustrating the specific implementation process of another embodiment of step S4200;

[0022] Figure 10 yes Figure 8 A schematic diagram illustrating the specific implementation process of another embodiment of step S4200;

[0023] Figure 11 This is a structural diagram of a tunnel flow calculation device provided in an embodiment of the present invention;

[0024] Figure 12 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0026] It should be understood that in the description of the embodiments of the present invention, the use of terms such as "first" and "second" is only for the purpose of distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of technical features indicated, or implicitly indicating the sequential relationship of the technical features indicated. "At least one" refers to one or more, and "more" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, the simultaneous existence of A and B, or the existence of B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, and c can be single or multiple.

[0027] Furthermore, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0028] The tunnel traffic calculation method involved in this invention is based on a Traffic Matrix (TM) to calculate the traffic of virtual tunnels between network units. The Traffic Matrix primarily estimates the traffic between origin and destination (OD) pairs in a communication network. Currently, various manufacturers have proposed different methods for estimating traffic using the Traffic Matrix. The most commonly used method is: given the network topology and network link traffic, calculate the information entropy based on the network link traffic, and then use the expectation-maximization algorithm to estimate the traffic demand between the network source and destination pairs.

[0029] Because the time gaps and timestamps in this method can cause errors in the calculation results due to hardware and software failures during transmission, the estimated values ​​may differ significantly from the measured values ​​in practical applications, affecting the accuracy of tunnel flow calculation.

[0030] Based on the above, embodiments of the present invention provide a tunnel traffic calculation method, apparatus, electronic device, and computer-readable storage medium. This method involves periodically acquiring the link load, total traffic, and physical channel traffic between a first network unit and multiple second network units; determining abnormal information based on the link load and total traffic; filtering out abnormal information to obtain a traffic matrix; and finally determining the tunnel traffic between the first network unit and each of the second network units based on the traffic matrix and physical channel traffic. This achieves the goal of accurately calculating tunnel traffic between network units, improving the accuracy of predicting network traffic trends, and enhancing the user experience.

[0031] Please see Figure 1 , Figure 1 The flowchart of a tunnel flow calculation method provided by an embodiment of the present invention is shown. Figure 1 As shown, the tunnel flow calculation method of this invention includes the following steps:

[0032] S1000, acquires the link load between multiple second network units.

[0033] It is understood that in a real network, there are multiple network elements that are directly or indirectly connected, and any two network elements transmit data through the link between them. For example, this invention considers one designated network element as the first network element and the other network elements as the second network element.

[0034] Understandably, since there is Virtual Private Network (VPN) traffic between the first network unit and the second network unit, the VPN gateway enables remote access by encrypting data packets and converting their destination addresses. For example, the first network unit can collect port traffic, i.e., link load, between the first and second network units through port traffic performance reported via the Simple Network Management Protocol (SNMP) within the network unit.

[0035] In practical applications, the link load between the first network unit and multiple second network units can be periodically collected and transmitted using the existing link load collection module and instructions within the network unit. This is existing technology and will not be elaborated here.

[0036] S2000, obtains the total traffic between multiple second network units.

[0037] It is understood that the total traffic in this invention refers to the sum of Internet Protocol Address (IP) traffic originating from and originating from the first network unit and multiple second network units. For example, the NetFlow module is used to collect IP traffic reported by the user network interface (UNI) of the first network unit.

[0038] Please see Figure 2 , Figure 2 A schematic diagram illustrating the specific implementation process of another embodiment of step S2000 described above is shown. For example... Figure 2 As shown, step S2000 includes at least the following steps:

[0039] S2100, acquires protocol address traffic between multiple second network units.

[0040] Understandably, Netflow's primary function is not to acquire SRBE and LDP tunnel traffic, but rather to monitor private network routing and traffic, as well as traffic between various Autonomous Systems (AS). Therefore, the first network unit can directly acquire protocol address traffic between itself and multiple second network units via Netflow, while also acquiring the IP address of the corresponding second network unit.

[0041] S2200, based on protocol address traffic, obtains the total traffic originating from and originating from multiple second network units.

[0042] Please see Figure 3 , Figure 3 A schematic diagram illustrating the specific implementation process of another embodiment of step S2200 described above is shown. For example... Figure 3 As shown, step S2200 includes at least the following steps:

[0043] S2210, based on the protocol address traffic, obtains the corresponding destination device.

[0044] As is understandable, NetFlow is a data exchange method used to statistically record information such as the OD address and port number of data packets in the network, and is an important means of network traffic statistics and analysis. By deploying NetFlow collection functionality in the first network unit, the NetFlow module can report the traffic between the first network unit and multiple second network units, and use protocols such as IGP, Border Gateway Protocol (BGP), and static routing to find the second network unit to which the IP traffic points.

[0045] S2220, based on the protocol address traffic, obtains the total traffic originating from and originating from multiple second network units.

[0046] Understandably, after finding the destination device of IP traffic based on NetFlow, it is convenient to aggregate and calculate the IP traffic with the same destination device in the first network unit to obtain the total IP traffic between the first network unit and a certain second network unit.

[0047] In practical applications, the first network unit can periodically collect and statistically analyze the IP traffic between the first network unit and multiple second network units by deploying the NetFlow collection function within the network unit. This is existing technology and will not be elaborated here.

[0048] S3000 determines anomaly information based on link load and total traffic.

[0049] Please see Figure 4 , Figure 4 A schematic diagram illustrating the specific implementation process of another embodiment of step S3000 described above is shown. For example... Figure 4 As shown, step S3000 includes at least the following steps:

[0050] S3100 calculates the predicted values ​​of link load and total traffic based on the link load and total traffic.

[0051] It is understood that in this invention, the difference between the predicted and measured values ​​of link load and the difference between the predicted and measured values ​​of total traffic both follow a Gaussian distribution; that is, the data noise of link load and the data noise of total traffic both follow a Gaussian distribution. Therefore, by analyzing and processing the collected measured values ​​of link load and total traffic, it is possible to determine whether the collected measured values ​​belong to abnormal information based on the data noise.

[0052] Please see Figure 5 , Figure 5 A schematic diagram illustrating the specific implementation process of another embodiment of step S3100 described above is shown. For example... Figure 5 As shown, step S3100 includes at least the following steps:

[0053] S3110, construct a first objective function based on link load, first data noise, total traffic, and second data noise; wherein the first data noise is a preset data noise corresponding to the link load, and the second data noise is a preset data noise corresponding to the total traffic.

[0054] It is understandable that the predicted value of link load and the predicted value of total traffic can be expressed by the following formula:

[0055] C = c + εc

[0056] D=d+ε d

[0057] Where C is the predicted value of the total flow, c is the measured value of the total flow, and ε c The second data noise is represented by D, where D is the predicted link load, d is the measured link load, and ε is the measured link load. d This represents the first type of data noise.

[0058] Furthermore, assume F = (f1, f2, ..., f n ) T This represents the total number of bytes collected in the traffic, with a sampling rate of R = (r1, r2, ..., r...). n ), thus obtaining the following formula:

[0059]

[0060] Among them, c i (1≤i≤n) is the predicted value of the total traffic of link i, and n is the total number of links.

[0061] Assume that the link traffic of link i in time interval [t, t+l] is d. i In actual measurements, the interval between two consecutive link performance measurements on link i is [t+Δ, (t+Δ1)+(l+Δ2)]. The actual link load measurement result is denoted as m. i Considering Δ1*Δ2<<1, assuming the traffic ratio on link i is within a short time interval, for example, under [tl,t+2l], we obtain the following formula:

[0062]

[0063] in, d is a control parameter. i The unbiased estimate is Therefore, its mean square error is approximately obtained as follows:

[0064]

[0065] Transforming the above equation, we get: Where λ is a constant obtained by fitting the scattering interval.

[0066] Understandably, by making scientific and reasonable predictions of the data noise of the link load and the data noise of the total traffic, we can accurately obtain the predicted values ​​of the link load and the total traffic by measuring the link load and the total traffic.

[0067] For example, assuming the data noise follows a Gaussian distribution, we get the following formula:

[0068]

[0069]

[0070] Assuming D = AC, substituting into the formula in step S3110, we get: Y = HC + N

[0071] Where A is the transformation matrix between D and C, H = (1; A) is a (m+n)×n matrix, and m is the total number of link loads. The weighted least squares evaluation of C in the above formula can be normalized to the following equivalent formula:

[0072] C=(H T K -1 H) -1 H T K -1 Y

[0073] Where K = E[NN] T ] is the covariance matrix of N, and K is a diagonal matrix when the measurement errors are uncorrelated.

[0074] Furthermore, assuming and The least-squares estimate of C in the above equation is transformed into a quadratic optimization problem, namely, minimizing the weighted error of the measurement data. For example, the first objective function is constructed as follows:

[0075]

[0076] S3120, based on the first objective function, obtains the predicted values ​​of link load and total traffic.

[0077] It is understandable that by solving the first objective function and substituting it into the formula of step S3110, the predicted values ​​of link load and total traffic can be obtained.

[0078] S3200 calculates the weighted error value based on the current value of the link load, the current value of the total traffic, the predicted value of the link load, and the predicted value of the total traffic.

[0079] It is understandable that after obtaining the solution to the first objective function, the following result can be obtained by using d = AC:

[0080]

[0081]

[0082] in, This is a predicted value of the current total flow. This is a measurement of the current total flow rate. The data noise is the sum of the current traffic. This is the predicted value of the current link load. This is a measurement of the current link load. This represents the data noise of the current link load.

[0083] Therefore, And 1≤i≤n, 1≤j≤m.

[0084] Please see Figure 6 , Figure 6 A schematic diagram illustrating the specific implementation process of another embodiment of step S3200 described above is shown. For example... Figure 6 As shown, step S3200 includes at least the following steps:

[0085] S3210, calculate the mean square error of the data noise of the link load to obtain the first mean square error value.

[0086] Understandably, based on the above calculation results, μ i This is the first mean square error value.

[0087] S3220, the absolute value of the difference between the current value of the link load and the predicted value of the link load is divided by the first mean square error value to obtain the first weighted error value.

[0088] Understandably, based on the above calculation results, This is the first weighted error value.

[0089] S3230 calculates the mean square error of the data noise of the total flow rate to obtain the second mean square error value.

[0090] Understandably, based on the above calculation results, σ i This is the second mean square error value.

[0091] S3240, the absolute value of the difference between the current value of the total flow and the predicted value of the total flow is divided by the second mean square error value to obtain the second weighted error value.

[0092] Understandably, based on the above calculation results, This is the second weighted error value.

[0093] S3300 determines abnormal information based on the weighted error value.

[0094] Please see Figure 7 , Figure 7 A schematic diagram illustrating the specific implementation process of another embodiment of step S3300 described above is shown. For example... Figure 7 As shown, step S3300 includes at least the following steps:

[0095] S3310, compare the first weighted error value, the second weighted error value and the first preset threshold.

[0096] Understandably, comparing the first weighted error value, the second weighted error value, and the first preset threshold helps determine the predicted value of the current total flow. Current link load forecast The accuracy. For example, the first preset threshold is set to 5.13.

[0097] S3320, if the maximum value of the first weighted error value and the second weighted error value is greater than the first preset threshold, mark the current value of the link load and the current value of the total traffic as abnormal information.

[0098] Understandably, when and When the maximum value is greater than the first preset threshold, the predicted value of the current total traffic can be determined. Current link load forecast At least one of the measured values ​​has a significant error compared to the predicted value of the current total flow. Current link load forecast Mark as abnormal information to facilitate the filtering and removal of abnormal information.

[0099] S4000 filters out abnormal information to obtain the traffic matrix.

[0100] Please see Figure 8 , Figure 8 A schematic diagram illustrating the specific implementation process of another embodiment of step S4000 described above is shown. For example... Figure 8 As shown, step S4000 includes at least the following steps:

[0101] S4100 sets the predicted value of link load in the anomaly information as the measured value of link load, and sets the predicted value of total traffic in the anomaly information as the measured value of total traffic.

[0102] Understandably, when and If the maximum value exceeds the first preset threshold, the predicted value of the current total traffic, which is considered an anomaly, needs to be changed. Current link load forecast Perform a screening process. For example, [the process involves] filtering out [the unwanted items]. Set as Bundle Set as This aims to eliminate the impact of abnormal information on the accuracy of the traffic matrix. The specific operation is shown in the following formula:

[0103]

[0104]

[0105] S4200 performs correction processing on the measured value of the total flow rate to obtain the flow rate matrix.

[0106] Understandably, after filtering out abnormal information, further optimization and correction are needed to prevent errors in the total flow rate from affecting the accuracy of the flow matrix and to improve its rationality. For example, when measurement noise and missing values ​​exist in the flow matrix, additional correction processing is required to eliminate the impact of these outliers on the flow matrix.

[0107] Please see Figure 9 and Figure 10 , Figure 9 and Figure 10 A schematic diagram illustrating the specific implementation process of another embodiment of step S4200 described above is shown. For example... Figure 9 and Figure 10 As shown, step S4200 includes at least the following steps:

[0108] S4210 performs linear modeling of the flow matrix and configures the parameters of the linear model to obtain the predicted value of the flow matrix.

[0109] Understandably, when performing flow matrix prediction, a linear system is used to model flow changes, as shown in the following equation:

[0110] C i+1 =FC i +ω T

[0111] y t =AC i +m t

[0112] Where F is the state transition matrix of the decision components in the process of obtaining the total flow, ω T For noise term, m t To measure noise. Assume C i|t Let C be at time t. i The predicted value is based on information prior to time t-1, C t|i C is at time t i The estimated value is:

[0113] C i|t =CC t|i

[0114] P i|t =CP t|i +Q

[0115] Where P is the covariance matrix of the error, Q is the covariance matrix of the noise term, and the predicted value C is used. i|t By updating the state and transitions using the measured value y, we obtain the following formula:

[0116] C i|t+1 =C i|t +G t+1 [y t+1 -AC t|i ]

[0117] P i|t+1 =(IG) t+1 A)P i|t +(IG t+1 A)+G t+1 RG t+1

[0118] Where R is the covariance matrix of the measurement noise, I is the identity matrix, and G is the Kalman gain matrix. The linear minimum variance estimation method for the flow matrix is ​​obtained from the above two equations. Where C, R, Q, and their associated initial values ​​C... 00 and P 00 Obtained through direct measurement. It is understandable that using the minimum variance estimation method for matrix calculation is existing technology, and will not be elaborated upon here.

[0119] S4220, If the difference between the predicted value of the flow matrix and the flow matrix is ​​greater than the second preset threshold, calibrate the parameters of the linear model.

[0120] Understandably, when the deviation between the predicted value of the flow matrix and the flow matrix exceeds the second preset threshold, C, R, Q and their associated initial values ​​need to be recalibrated to optimize and correct the linear model and ensure its accuracy.

[0121] S4230, in the event of data loss in the total flow, uses the gravity model to obtain prior information and adds the corresponding prior information to the flow matrix.

[0122] Understandably, in practical applications, NetFlow deployment is limited by device hardware and software, which can lead to errors when obtaining large traffic data from NetFlow. In such cases, a gravity model is used to obtain prior information C0 and fill in the corresponding elements of the traffic matrix to ensure the data integrity of the traffic matrix.

[0123] S4240 uses an iterative proportional fitting algorithm to construct a second objective function based on the flow matrix and prior information.

[0124] Understandably, by solving a quadratic programming problem in L2 normal form for vectors, the formula for the second objective function is obtained as follows:

[0125] min||(C-C0) / ω||

[0126] To satisfy ||AC-y||=0, where ω is the weight vector.

[0127] For example, when using the least squares method to solve the link traffic condition subspace, the least squares method may produce negative values ​​due to inaccurate prior information C0. In this case, an Iterative Proportional Fitting Algorithm (IPFA) is used to ensure that non-negative values ​​are produced. The calculation of the Iterative Proportional Fitting Algorithm is existing technology and will not be described in detail here.

[0128] S4250 updates the flow matrix based on the second objective function.

[0129] Understandably, by solving the second objective function, the corrected flow matrix is ​​obtained to ensure the integrity and reference value of the flow matrix.

[0130] Understandably, when the data obtained by NetFlow is incomplete or lost during transmission due to network unit configuration issues, and when the prior information is generated from elements of the traffic matrix directly derived from NetFlow data, further adjustments to the NetFlow data are needed, and the traffic matrix should be calibrated using link load. The relevant calibration process is consistent with step S4200 above and will not be repeated here.

[0131] S5000 acquires physical channel traffic between multiple second network units.

[0132] It is understood that the physical channel traffic in this embodiment of the invention includes, but is not limited to, data traffic under various connection-oriented, point-to-point service-bearing network protocols. Examples include Secure Real-time Transport Protocol (SRTP), Multi-Protocol Label Switching Transport Profile (MPLS-TP), and Segment Routing Traffic Management Protocol (SR-TE). This physical channel traffic can be periodically acquired and monitored in real-time by existing monitoring modules in the network unit; the methods of acquisition and transmission are prior art and will not be elaborated here.

[0133] S6000 determines the tunnel traffic between each second network unit based on the traffic matrix and the physical channel traffic.

[0134] Understandably, in practical applications, the tunnel traffic between the first and second network units is the difference between the traffic matrix and the physical channel traffic. Therefore, provided the traffic matrix and physical channel traffic are accurate, the tunnel traffic between the first and second network units obtained through these data has practical reference value. It enables traffic monitoring, management, and prediction, thereby allowing for overall network traffic management, dynamic optimization, and scaling recommendations to avoid network congestion and traffic overload.

[0135] The tunnel traffic calculation method provided in this invention is applicable to 4G, 5G technologies and their mixed network scenarios. It can effectively improve the accuracy of tunnel traffic measurement, ensure the quality of tunnel traffic monitoring and management, and improve the user experience.

[0136] See Figure 11 , Figure 11 This is a schematic diagram of the structure of the tunnel flow calculation device 700 provided in the embodiment of the present invention. The tunnel flow calculation method provided in the embodiment of the present invention involves the following modules in the tunnel flow calculation device: acquisition module 710, calculation module 720, screening module 730 and determination module 740.

[0137] The acquisition module 710 is used to acquire the link load, total traffic, and physical channel traffic between the first network unit and multiple second network units; the calculation module 720 is used to determine abnormal information based on the link load and total traffic; the filtering module 730 is used to filter out abnormal information and obtain a traffic matrix; and the determination module 740 is used to determine the tunnel traffic between each second network unit based on the traffic matrix and the physical channel traffic.

[0138] It should be noted that the information interaction and execution process between the modules of the above-mentioned device are based on the same concept as the method embodiment of the present invention. For details on their specific functions and technical effects, please refer to the method embodiment section, which will not be repeated here.

[0139] Figure 12 An electronic device 800 according to an embodiment of the present invention is shown. The electronic device 800 includes, but is not limited to:

[0140] Memory 801 is used to store programs;

[0141] The processor 802 is used to execute the program stored in the memory 801. When the processor 802 executes the program stored in the memory 801, the processor 802 is used to execute the tunnel flow calculation method described above.

[0142] The processor 802 and the memory 801 can be connected via a bus or other means.

[0143] The memory 801, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs, such as the tunnel flow calculation method described in any embodiment of the present invention. The processor 802 implements the above-described tunnel flow calculation method by running the non-transitory software program and instructions stored in the memory 801.

[0144] The memory 801 may include a program storage area and a data storage area. The program storage area may store the operating system and application programs required for at least one function; the data storage area may store the tunnel traffic calculation method described above. Furthermore, the memory 801 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 801 may optionally include memory remotely located relative to the processor 802, and these remote memories can be connected to the processor 802 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0145] The non-transient software program and instructions required to implement the above-described tunnel flow calculation method are stored in memory 801. When executed by one or more processors 802, the tunnel flow calculation method provided in any embodiment of the present invention is executed.

[0146] This invention also provides a storage medium storing computer-executable instructions for executing the tunnel flow calculation method described above.

[0147] In one embodiment, the storage medium stores computer-executable instructions that are executed by one or more control processors 802, such as one of the processors 802 in the electronic device 800, to enable the one or more processors 802 to execute the tunnel flow calculation method provided in any embodiment of the present invention.

[0148] The embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0149] It will be understood by those skilled in the art that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically include computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0150] The foregoing detailed description of preferred embodiments of the present invention is not limited to the above-described embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of the present invention.

Claims

1. A tunnel traffic calculation method, applied to a first network unit, the method comprising: Acquire link load between multiple second network units; Obtain the total traffic between the network units and the plurality of second network units; Anomaly information is determined based on the link load and the total traffic. The anomaly information is determined by a weighted error value obtained from the current value of the link load, the current value of the total traffic, the predicted value of the link load, and the predicted value of the total traffic. After filtering out the abnormal information, a traffic matrix is ​​obtained; Acquire physical channel traffic between multiple second network units; Based on the traffic matrix and the entity channel traffic, the tunnel traffic between each of the second network units is determined.

2. The method according to claim 1, characterized in that, The acquisition of the total traffic between the network and the plurality of second network units includes: Obtain protocol address traffic between multiple second network units; Based on the protocol address traffic, the total traffic originating from and originating from the same source and destination as the multiple second network units is obtained.

3. The method according to claim 2, characterized in that, The step of obtaining the total traffic originating from and destined for the second network unit based on the protocol address traffic includes: Based on the protocol address traffic, the corresponding second network unit is obtained; The traffic of the protocol address in the same second network unit is aggregated and calculated to obtain the total traffic.

4. The method according to claim 1, characterized in that, The step of determining the abnormal information based on the link load and the total traffic includes: Calculate the predicted value of the link load and the predicted value of the total traffic based on the link load and the total traffic. The weighted error value is obtained based on the current value of the link load, the current value of the total traffic, the predicted value of the link load, and the predicted value of the total traffic. Based on the weighted error value, anomaly information is determined.

5. The method according to claim 4, characterized in that, The step of calculating the predicted value of the link load and the predicted value of the total traffic based on the link load and the total traffic includes: A first objective function is constructed based on the link load, the first data noise, the total traffic, and the second data noise; wherein the first data noise is a preset data noise corresponding to the link load, and the second data noise is a preset data noise corresponding to the total traffic. Based on the first objective function, the predicted value of the link load and the predicted value of the total traffic are obtained.

6. The method according to claim 4, characterized in that, The step of obtaining a weighted error value based on the current value of the link load, the current value of the total traffic, the predicted value of the link load, and the predicted value of the total traffic includes: Calculate the mean square error of the data noise of the link load to obtain the first mean square error value; The first weighted error value is obtained by dividing the absolute value of the difference between the current value of the link load and the predicted value of the link load by the first mean square error value. The mean square error of the data noise of the total flow rate is calculated to obtain the second mean square error value; The absolute value of the difference between the current value of the total flow and the predicted value of the total flow is divided by the second mean square error value to obtain the second weighted error value.

7. The method according to claim 6, characterized in that, The step of determining the abnormal information based on the weighted error value includes: If the maximum value of the first mean square error value and the second mean square error value is greater than the first preset threshold, the current value of the link load and the current value of the total traffic are marked as the abnormal information.

8. The method according to claim 7, characterized in that, The process of filtering out the abnormal information to obtain the traffic matrix includes: The predicted value of the link load in the anomaly information is set as the measured value of the link load, and the predicted value of the total traffic in the anomaly information is set as the measured value of the total traffic. The measured values ​​of the total flow rate are corrected to obtain the flow rate matrix.

9. A tunnel flow rate calculation device, characterized in that, include: The acquisition module is used to acquire the link load, total traffic, and physical channel traffic between the first network unit and multiple second network units; The calculation module is used to determine abnormal information based on the link load and the total traffic. The anomaly information is determined by a weighted error value obtained from the current value of the link load, the current value of the total traffic, the predicted value of the link load, and the predicted value of the total traffic. A filtering module is used to filter out the abnormal information to obtain a flow matrix; The determination module is used to determine the tunnel traffic between each of the second network units based on the traffic matrix and the entity channel traffic.

10. An electronic device, characterized in that, include: The processor includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the tunnel flow calculation method as described in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The device contains a computer program that, when executed by a processor, implements the tunnel flow calculation method as described in any one of claims 1 to 8.

Images