Efficient flow control method, device and system for edge cloud gateway

Abstract

The invention relates to an efficient flow control method, device and system for an edge cloud gateway. The method mainly comprises the following steps: generating an AI algorithm according to administrative region traffic characteristics, training actual administrative region traffic through the AI algorithm, and then generating an administrative region model library; an original strategy combining the administrative region, the time period and the App is converted into an execution rule Rule [administrative region ID] [AppID]; after the edge cloud gateway receives the user network traffic, the AppID is identified according to the traffic characteristics, and the administrative region ID is identified in the administrative region model library; and according to the identified administrative region ID and AppID, judging whether Rule [administrative region ID] [AppID] accords with a control condition, if so, executing flow control and then transferring out of the edge cloud gateway, otherwise, not executing flow control. According to the invention, the accuracy of position identification can be improved, and misjudgment of subsequent control is avoided; and the traffic forwarding time delay of the edge cloud gateway can be greatly reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of edge cloud gateways, and in particular to an efficient flow control method, device and system for edge cloud gateways. Background Art

[0002] The edge cloud gateway is an efficient network data forwarding device that not only meets the needs of broadband users for network access, but also provides access capabilities for different terminal devices in the user network and achieves precise control. For example, regulatory authorities can supervise and control different App applications according to different administrative regions and at different time periods.

[0003] refer to Figure 1 As shown in the figure, the time combined with application-level traffic control inside the edge cloud gateway is implemented by two main components, the App traffic identification unit and the App traffic control unit. The entire control process is as follows: ① The edge cloud gateway receives the user network traffic and transfers it to the App traffic identification unit. ② The App traffic identification unit completes App identification based on the traffic characteristics and marks the AppID of the traffic. Assume that the identified AppID is 50. ③ The traffic is transferred to the App traffic control unit, and the traffic control unit matches the App control policy; first, the current time and date T1 is obtained; secondly, according to satellite positioning, the current user access administrative area L1 is obtained; finally, the AppID marked by the traffic is obtained as 50. ④ The traffic control unit uses the current time and date T1, the current user access administrative area L1, and the AppID marked by the traffic as input conditions, queries the matching strategy, and executes the action to complete the traffic control. ⑤ After traffic control, the traffic is transferred out of the edge cloud gateway.

[0004] The above prior art has the following disadvantages:

[0005] (1) For the location acquisition of administrative regions, the edge cloud gateway location location is the location location, not the user's actual access point location. There is a situation where it cannot truly reflect the administrative region where the user accesses the network, resulting in control misjudgment.

[0006] (2) During the traffic control process, the control strategy is executed in a circular matching manner. Assuming that the number of policy administrative areas is n, the number of time periods is n, and the number of apps that need to be controlled is n; then the algorithm complexity of the final policy matching is O(n)*O(n)*O(n)=O(n 3); Under this level of complexity, it takes a long time to match policies during packet forwarding, which can easily lead to increased forwarding delays. In particular, as the device traffic throughput increases, the device forwarding delay will be greatly increased; for example, in the case of 70Gbps and 15MPPS traffic throughput in the laboratory, the average forwarding delay of this solution exceeds 1ms, and the maximum forwarding delay exceeds 2ms.

[0007] In view of this, how to overcome the defects of the existing technology and how to solve at least one of the problems of control misjudgment caused by inaccurate administrative area positioning and forwarding delay caused by high complexity has become an important technical problem that the industry needs to solve urgently. Summary of the invention

[0008] In view of the defects or improvement needs in the prior art: how to solve the problem of control misjudgment caused by inaccurate administrative area positioning and the problem of forwarding delay caused by high complexity. The present invention provides an efficient traffic control method, device and system for an edge cloud gateway. According to the traffic characteristics of the administrative area, an administrative area model library after AI training is established. In the traffic analysis stage, the AppID is identified according to the traffic characteristics, and then the administrative area ID is identified through the administrative area model library, which can improve the accuracy of location identification and avoid subsequent control misjudgment. In addition, the effective time judgment of the policy is separated from the traffic forwarding process for parallel calculation, and the original control policy is converted into an execution rule Rule[administrative area ID][AppID]. In this way, in the forwarding process, it is only necessary to judge whether control is required based on the AppID and the administrative area ID. Therefore, the time complexity of the entire forwarding control algorithm is actually O(1), which greatly reduces the delay of edge cloud gateway traffic forwarding.

[0009] The present invention adopts the following technical solution:

[0010] In a first aspect, the present invention provides an efficient flow control method for an edge cloud gateway, comprising:

[0011] Generate an AI algorithm based on the traffic characteristics of the administrative region, and generate an administrative region model library after training the actual administrative region traffic through the AI ​​algorithm;

[0012] Convert the original policy combining administrative regions, time periods, and apps into execution rules: Rule[administrative region ID][AppID];

[0013] After receiving the user's network traffic, the edge cloud gateway identifies the AppID based on the traffic characteristics and identifies the administrative region ID in the administrative region model library;

[0014] According to the identified administrative area ID and AppID, it is determined whether Rule [administrative area ID] [AppID] meets the control conditions. If so, flow control is performed and then transferred out of the edge cloud gateway. Otherwise, flow control is not performed.

[0015] In some embodiments, generating an AI algorithm based on administrative area traffic characteristics specifically includes:

[0016] Based on the characteristics of administrative area traffic: MAC, source IP location, requested DNS location, and App traffic curve, a weighted regression model is formed, and finally an AI algorithm for traffic fingerprint recognition of administrative areas is generated.

[0017] In some embodiments, the generating of the administrative region model library after training the actual administrative region traffic through the AI ​​algorithm specifically includes: generating the administrative region model library after performing AI training on the administrative region traffic through the administrative region access network.

[0018] In some embodiments, the conversion of the original policy combining the administrative region, time period, and App into the execution rule Rule[administrative region ID][App ID] specifically includes:

[0019] Get the current time and date, and match the original policy in effect in a cyclic manner according to the current time and date;

[0020] Get the administrative region ID and AppID in the original policy and set them to the Rule[Administrative Region ID][AppID] array.

[0021] In some embodiments, in the process of converting the original policy combining the administrative region, time period and App into the execution rule Rule[administrative region ID][AppID], the task is executed periodically.

[0022] In some embodiments, judging whether Rule [administrative region ID] [AppID] meets the control condition according to the identified administrative region ID and AppID specifically includes:

[0023] Substitute the identified administrative region ID and AppID into Rule[administrative region ID][AppID]. If the substituted Rule[administrative region ID][AppID] is greater than 0, the control condition is met and flow control is performed.

[0024] Substitute the identified administrative region ID and AppID into Rule[administrative region ID][AppID]. If the substituted Rule[administrative region ID][AppID] is not greater than 0, the control condition is not met and traffic control is not performed.

[0025] In some embodiments, the receiving and sending of the traffic are both implemented through physical ports.

[0026] In a second aspect, the present invention further provides an efficient flow control device for an edge cloud gateway, the device comprising:

[0027] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the processor to execute the efficient flow control method for the edge cloud gateway described in the first aspect.

[0028] In a third aspect, the present invention further provides an efficient traffic control system for an edge cloud gateway, applying the efficient traffic control method for the edge cloud gateway as described in the first aspect, the system includes a traffic identification unit, a traffic control unit, an AI training unit, and a policy decomposition unit, the traffic identification unit includes an administrative area traffic fingerprint model matching module and an App analysis and identification module, wherein:

[0029] The AI ​​training unit is used to generate an AI algorithm according to the traffic characteristics of the administrative area, generate an administrative area model library after AI training, and load the administrative area model library into the administrative area traffic fingerprint model matching module of the traffic identification unit;

[0030] The policy decomposition unit is used to convert the original policy combining the administrative area, time period and App into an execution rule Rule[administrative area ID][AppID];

[0031] The App analysis and identification module of the traffic identification unit is used to identify the AppID according to the traffic characteristics after the edge cloud gateway receives the user network traffic, and at the same time transfer it to the administrative area traffic fingerprint model matching module to perform administrative area identification, and identify the administrative area ID in the administrative area model library according to the AppID;

[0032] The traffic control unit is used to determine whether Rule[administrative area ID][AppID] meets the control conditions based on the identified administrative area ID and AppID. If so, traffic control is executed and then transferred out of the edge cloud gateway. Otherwise, traffic control is not executed.

[0033] In a fourth aspect, the present invention further provides a non-volatile computer storage medium, wherein the computer storage medium stores computer executable instructions, which are executed by one or more processors to complete the efficient traffic control method for the edge cloud gateway described in the first aspect.

[0034] Compared with the prior art, the present invention provides an efficient flow control method, device and system for edge cloud gateway, and its beneficial effect is: according to the flow characteristics of administrative regions, an administrative region model library after AI training is established, AppID is identified according to the flow characteristics in the flow analysis stage, and then the administrative region ID is identified through the administrative region model library, which can improve the accuracy of location identification and avoid misjudgment of subsequent control. In addition, the effective time judgment of the policy is separated from the flow forwarding process for parallel calculation, and the original control policy is converted into an execution rule Rule [administrative region ID] [AppID], so that in the forwarding process, it is only necessary to judge whether control is required based on AppID and administrative region ID. Therefore, the time complexity of the entire forwarding control algorithm is actually O (1), which greatly reduces the delay of edge cloud gateway flow forwarding. After experiments, under the flow throughput of 70Gbps and 15MPPS in the laboratory, the scheme of the embodiment of the present invention can make the average forwarding delay less than 150us and the maximum forwarding delay of 250us, which is 85% higher than the forwarding performance of the original scheme. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 A schematic diagram of a control and processing scheme for flow data by an edge cloud gateway provided by the background technology of the present invention;

[0037] Figure 2 A flowchart of an efficient flow control method for an edge cloud gateway provided in Example 1 of the present invention;

[0038] Figure 3 An expanded flow chart of step 200 provided in Embodiment 1 of the present invention;

[0039] Figure 4 A specific flow chart for determining whether a control condition is met provided in Example 1 of the present invention;

[0040] Figure 5 A schematic diagram of a unit module of an efficient traffic control system of an edge cloud gateway provided in Example 2 of the present invention;

[0041] Figure 6 A schematic diagram of the structure of an efficient flow control device for an edge cloud gateway provided in Example 3 of the present invention. DETAILED DESCRIPTION

[0042] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but do not limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, without departing from the concept of the present invention, several variations and improvements can also be made. These all belong to the protection scope of the present invention. It should be noted that, if there is no conflict, the various features in the embodiments of the present invention can be combined with each other, all within the protection scope of the present invention. In addition, although the functional module division may be performed in the device schematic diagram, and the logical order may be shown in the flow chart, in some cases, the steps shown or described may be performed in a sequence different from the module division in the device, or in the flow chart.

[0043] Unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in this specification and in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. In addition, the technical features involved in each embodiment of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0044] Unless the context requires otherwise, throughout the specification and claims, the term "including" is to be interpreted as open inclusion, that is, "including, but not limited to". In the description of the specification, the terms "one embodiment", "some embodiments", "exemplary embodiments", "examples", "specific examples" or "some examples" and the like are intended to indicate that specific features, structures, materials or characteristics associated with the embodiment or example are included in at least one embodiment or example of the present disclosure. The schematic representation of the above terms does not necessarily refer to the same embodiment or example. In addition, the specific features, structures, materials or characteristics may be included in any one or more embodiments or examples in any appropriate manner, that is, although they may be carried in the embodiments or examples of the above terms due to reasons such as the order and position of appearance, it is not limited to that they can be carried in combination by one embodiment or example.

[0045] The abbreviations and key terms in the present invention are defined as follows:

[0046] ECGW: Edge Cloud GateWay, edge cloud gateway, is a gateway that moves the three-layer functions of home gateway devices to edge cloud deployment. The hardware uses a general server to deploy gateway function modules through virtualization and containerization to achieve cloud-based gateways. The cloud gateway mainly identifies user traffic and then dials up to the Internet or directs traffic. The system can provide value-added services such as cloud security, cloud entertainment, cloud games, cloud storage, and cloud monitoring through directed traffic diversion;

[0047] App: Application, usually refers to applications developed for mobile devices such as smartphones, such as WeChat and TikTok on mobile phones;

[0048] AppID: Application Identity, the unique code and exclusive number of the application inside the edge cloud gateway; the edge cloud gateway supports a total of 8,000 apps, and the AppID range is: [1, 8000];

[0049] AI: Artificial Intelligence; Artificial Intelligence is a very broad science, including robotics, speech recognition, image recognition, natural language processing, expert systems, machine learning, computer vision, etc.

[0050] MAC: Media Access Control; used to define the location of a network device, usually represented by 6 hexadecimal digits. Each network device has a unique MAC address, which ensures that data packets can be correctly delivered to the destination;

[0051] IP: refers to Internet Protocol, the abbreviation of Internet Protocol, which is the network layer protocol in the TCP / IP system. The purpose of designing IP is to improve the scalability of the network: first, to solve Internet problems and realize the interconnection of large-scale, heterogeneous networks; second, to separate the coupling relationship between top-level network applications and bottom-level network technologies to facilitate the independent development of the two. According to the end-to-end design principle, IP only provides a connectionless, unreliable, best-effort data packet transmission service for the host.

[0052] DNS: Domain Name System, domain name system; is a service of the Internet. As a distributed database that maps domain names and IP addresses to each other, it enables people to access the Internet more conveniently.

[0053] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0054] Embodiment 1:

[0055] like Figure 2 As shown, an embodiment of the present invention provides an efficient flow control method for an edge cloud gateway, and the method includes the following steps.

[0056] Step 100: Generate an AI algorithm based on the administrative area traffic characteristics, and generate an administrative area model library after training the actual administrative area traffic through the AI ​​algorithm. In this step, the characteristics carried by the traffic include MAC, source IP location, requested DNS location, and App traffic curve graph.

[0057] Step 200: Convert the original policy combining administrative regions, time periods and apps into execution rules: Rule[Administrative Region ID][AppID]. It should be noted that Rule[Administrative Region ID][AppID] is a two-dimensional array of execution rules. The first dimension is the ID of the administrative region, and the second dimension is the ID of the App application. If there are 100 administrative regions and 8,000 applications, then the definition of this array is Rule

[100]

[8000] . For example, the ID of Wuchang District is 32, and the ID of the QQ application is 1,000. Then the rule for the QQ application in Wuchang District is Rule

[32]

[1000] . It should be noted that the meaning of the two-dimensional array in the program is that the so-called execution rule here is a numerical identifier of a two-dimensional array in the program. Taking the QQ application in Wuchang District as an example, if Rule

[32]

[1000] >0 indicates that control is required, and if Rule

[32]

[1000] =0, control is not required. In this step, the task is executed according to a 1ms cycle, and the original policy combining the administrative area, time period and App is converted into the execution rule Rule[Administrative Area ID][AppID]. The cycle setting is similar to a clock, and the time granularity is 1ms. This means that every 1ms, the policy judgment and analysis will be converted into Rule[Administrative Area ID][AppID]. For example: A user in Wuchang District sets a rule as follows (the area is Wuchang District, and QQ is controlled from 10:00:00 to 11:59:59 every day). Then this task will judge whether the current policy is effective every 1ms and convert it into Rule[Administrative Area ID][AppID]. Assuming that the current time is 10:00:01, Rule

[32]

[1000] +=1 will be set; (here, it is assumed that the Wuchang District ID is 32 and the QQ AppID is 1000). Rule

[32]

[1000] +=1 is a cumulative representation in the software. This is equivalent to Rule

[32]

[1000] =Rule

[32]

[1000] +1. Assuming that Rule

[32]

[1000] =0, after executing the above formula, Rule

[32]

[1000] =1. If the current time is 12:00:00, then Rule

[32]

[1000] -=1 will be set; that is, Rule

[32]

[1000] =Rule

[32]

[1000] -1; according to this logic, it is ensured that Rule

[32]

[1000] is greater than 0 during the effective time and equal to 0 during the invalid time.

[0058] Step 300: After receiving the user network traffic, the edge cloud gateway identifies the AppID based on the traffic characteristics and identifies the administrative region ID in the administrative region model library. This step is a traffic analysis step, which is used to identify the AppID and administrative region ID of the traffic. Because the administrative region model library after AI training is established based on the traffic characteristics of the administrative region, in the traffic analysis stage of this step, the AppID can be identified based on the traffic characteristics, and then the administrative region ID can be identified through the administrative region model library, which can improve the accuracy of location identification and avoid misjudgment of subsequent control. It should be noted that AppID is just a number representation of the application, such as 1000 for QQ and 1001 for WeChat. The application characteristics of different regions are different. For example, the traffic characteristics of the same QQ application will be different when I use it in Wuchang District and Hankou District. Assuming that Wuchang District is numbered 32, Hankou District is numbered 33, and the QQ application is numbered 1000, then the QQ execution rule in Wuchang District is expressed as Rule

[32]

[1000] , and the execution rule in Hankou District is expressed as Rule

[33]

[1000] . This mapping allows the program to identify and control more accurately. For example, if a user in Wuchang sets QQ control from 10:00:00 to 11:00:00, then during the actual traffic processing, the AI ​​model is used to identify that the traffic is from users in Wuchang District and is a QQ application, and the traffic is marked as 32, 1000. Then the program only needs to determine whether Rule

[32]

[1000] >0 is true. If so, this part of the traffic is controlled. If not, this part of the traffic is not controlled.

[0059] Step 400: Determine whether Rule [administrative region ID] [AppID] meets the control conditions based on the identified administrative region ID and AppID. If so, execute flow control and transfer out of the edge cloud gateway. Otherwise, do not execute flow control. This step is a flow control step, which is used to determine whether to perform flow control. Because the original policy combining administrative region, time period and App is converted into the execution rule Rule [administrative region ID] [AppID] in advance, in this step, the identified administrative region ID and AppID can be substituted into the execution rule Rule [administrative region ID] [AppID] to see whether the corresponding value meets the conditions for executing flow control. If it meets the conditions, flow control is executed, otherwise it is not executed. With such a design, compared with the prior art, the time complexity of the entire forwarding control algorithm is reduced from O(n) to O(n). 3 ) is reduced to O(1), greatly reducing the latency of edge cloud gateway traffic forwarding.

[0060] Through the above steps, the embodiment of the present invention establishes an AI model library according to the traffic characteristics of the administrative area, identifies the administrative area characteristics in the traffic analysis stage, improves the accuracy of the location, and avoids misjudgment of subsequent control; the policy effectiveness time judgment is separated from the traffic forwarding process for parallel calculation, so that in the forwarding process, only the AppID and administrative area ID need to be controlled. Therefore, the time complexity of the entire forwarding control algorithm is actually O(1), which greatly reduces the delay of edge cloud gateway traffic forwarding.

[0061] In some embodiments, for step 100, the generation of an AI algorithm based on administrative area traffic characteristics specifically includes: forming a weighted regression model based on the characteristics carried by the administrative area traffic: MAC, source IP location, requested DNS location, and App traffic curve graph, and finally generating an AI algorithm for traffic fingerprint recognition of the administrative area. The technology for generating the algorithm can adopt existing DPI technology, which will not be elaborated here.

[0062] In some embodiments, for step 100, generating an administrative area model library after AI training specifically includes: generating an administrative area model library after AI training is performed through an administrative area access network. The administrative area model library can be expressed as: [Administrative District 1--001, Administrative District 2--002...Administrative District n--00n]. It should be noted that the administrative area model library is the numbering of administrative district features, and the administrative districts after AI training are numbered for ease of expression. For example, after Wuchang District training, there will be a model of Wuchang District, and it will be numbered 32. When the traffic in Wuchang District is accessed, the traffic will be numbered 32 after identification.

[0063] refer to Figure 3 As shown, in some embodiments, for step 200, converting the original policy combining the administrative region, time period, and App into the execution rule Rule [administrative region ID] [App ID] specifically includes the following steps:

[0064] Step 201: Get the current date and time, and match the original policy in effect in a cyclic manner according to the current date and time. For example, get the current date and time T1, and match the original policy. Assuming that the current time is Saturday 10:00, use a cyclic method to match the effective policy, and then match Figure 1 Chinese policy2000.

[0065] Step 202: Get the administrative region ID and AppID in the original policy, and set them to the Rule[Administrative Region ID][AppID] array. For example, get the administrative region ID and AppID in policy2000 and set them to the Rule[Region ID][AppID] array; for example, if policy2000 has administrative region 200, and AppIDs are 1, 23, 24, 50, 60, 8000, then Rule

[200] [1]+=1; Rule

[200]

[23] +=1; Rule

[200]

[24] +=1; Rule

[200]

[50] +=1; Rule

[200]

[60] +=1; Rule

[200]

[8000] +=1. It should be noted that these expressions are expressions that accumulate 1 in the program. Taking Rule

[200]

[50] +=1 as an example, it is equivalent to Rule

[200]

[50] =Rule

[200]

[50] +1.

[0066] refer to Figure 4 As shown, in some embodiments, for step 400, judging whether Rule [administrative region ID] [AppID] meets the control condition according to the identified administrative region ID and AppID specifically includes:

[0067] Step 401: Substitute the identified administrative region ID and AppID into Rule[Administrative region ID][AppID]. If the substituted Rule[Administrative region ID][AppID] is greater than 0, the control condition is met and flow control is performed. For example, in step 300, assuming that the identified AppID is 50 and the administrative region ID is 200, then it is determined whether Rule

[200]

[50] is greater than 0. Since the value of Rule

[200]

[50] is 1 and greater than 0, the control condition is met and flow control is performed.

[0068] Step 402: Substitute the identified administrative region ID and AppID into Rule[administrative region ID][AppID]. If the substituted Rule[administrative region ID][AppID] is not greater than 0, the control condition is not met and traffic control is not performed.

[0069] In the above steps, the receiving and sending of the traffic are both realized through a physical port (eg, a network card port).

[0070] In summary, this embodiment proposes to provide an efficient traffic control method for an edge cloud gateway, including the following advantages: an administrative region model library after AI training is established according to the traffic characteristics of the administrative region, AppID is identified according to the traffic characteristics in the traffic analysis stage, and then the administrative region ID is identified through the administrative region model library, which can improve the accuracy of location identification and avoid misjudgment of subsequent control. In addition, the effective time judgment of the policy is separated from the traffic forwarding process for parallel calculation, and the original control policy is converted into an execution rule Rule[administrative region ID][AppID]. In this way, in the forwarding process, it is only necessary to judge whether control is required based on AppID and administrative region ID. Therefore, the time complexity of the entire forwarding control algorithm is actually O(1), which greatly reduces the delay of edge cloud gateway traffic forwarding. For example, for the QQ application in Wuchang District, it only needs to be judged: if(Rule

[32]

[1000] >0) is executed else is not executed, and the algorithm complexity of this statement is only O(1). After experiments, under the conditions of 70 Gbps and 15 MPPS flow throughput in the laboratory, the solution of the embodiment of the present invention can reduce the average forwarding delay to less than 150 us and the maximum forwarding delay to 250 us, which improves the forwarding performance by 85% compared with the original solution.

[0071] Embodiment 2:

[0072] Based on the efficient flow control method of the edge cloud gateway provided in the above-mentioned embodiment 1, embodiment 2 of the present invention provides an efficient flow control system of the edge cloud gateway to explain the solution of the present invention in more detail.

[0073] refer to Figure 5As shown, the efficient traffic control system of the edge cloud gateway of this embodiment includes a traffic identification unit, a traffic control unit, an AI training unit and a policy decomposition unit, wherein the traffic identification unit includes an administrative area traffic fingerprint model matching module and an App analysis and identification module, wherein: the AI ​​training unit is used to generate an AI algorithm according to the traffic characteristics of the administrative area, generate an administrative area model library after AI training, and load the administrative area model library into the administrative area traffic fingerprint model matching module of the traffic identification unit; the policy decomposition unit is used to convert the original policy combining the administrative area, time period and App into an execution rule Rule[administrative area ID][AppID]; the App analysis and identification module of the traffic identification unit is used to identify the AppID according to the traffic characteristics after the edge cloud gateway receives the user network traffic, and at the same time transfer it to the administrative area traffic fingerprint model matching module for administrative area identification, and identify the administrative area ID in the administrative area model library according to the AppID; the traffic control unit is used to determine whether Rule[administrative area ID][AppID] meets the control conditions according to the identified administrative area ID and AppID, and if so, execute traffic control and then transfer out of the edge cloud gateway, otherwise do not execute traffic control.

[0074] Specifically, refer to Figure 5 As shown, this embodiment adds an administrative area traffic fingerprint model matching module in the traffic identification unit. A weighted regression model is formed according to the characteristics carried by the traffic: MAC, source IP location, requested DNS location and App traffic curve graph, and finally an AI algorithm for fingerprint identification of administrative areas is generated; after AI training through the administrative area access network, a regional model library [Administrative Area 1--001, Administrative Area 2--002...Administrative Area n--00n] is generated, and finally loaded into the administrative area traffic fingerprint model matching module.

[0075] Add a policy decomposition unit, execute tasks according to a 1ms cycle, and convert the original administrative area + time period + App control policy into the execution rule Rule[Administrative Area ID][AppID]; the conversion process is as follows: Get the current time date T1 and match the original policy. Assuming that the current time is Saturday 10:00, use a loop to match the effective policy, and then match policy2000 in the figure. Get the administrative area ID and AppID in policy2000 and set them to the Rule[Administrative Area ID][AppID] array; for example, if policy2000 has administrative area 200 and AppIDs such as 1, 23, 24, 50, 60, and 8000, then Rule

[200] [1]+=1; Rule

[200]

[23] +=1; Rule

[200]

[24] +=1; Rule

[200]

[50] +=1; Rule

[200]

[60] +=1; Rule

[200]

[8000] +=1.

[0076] The conditional logic for the traffic control unit to determine whether to control the traffic message is: if (Rule [area ID] [AppID]> 0) {control} else {not control}.

[0077] The edge cloud gateway receives user network traffic and processes it as follows:

[0078] ①The edge cloud gateway receives user network traffic and transfers it to the traffic identification unit.

[0079] ② The App identification and analysis module in the traffic identification unit completes App identification based on traffic characteristics and marks the AppID of the message. Assume that the identified AppID is 50. At the same time, it is transferred to the administrative area traffic fingerprint model matching module for area identification. Assume that the final administrative area ID is 200.

[0080] ③ The traffic is transferred to the traffic control unit, which matches the App control policy.

[0081] ④ Based on the identified AppID of 50 and the area ID of 200, judge: if(Rule

[200]

[50] >0){execute control}.

[0082] ⑤The traffic is transferred out of the edge cloud gateway after traffic control.

[0083] In summary, the embodiment of the present invention establishes an AI model library based on the traffic characteristics of administrative regions, identifies the characteristics of administrative regions in the traffic analysis stage, improves the accuracy of the location, and avoids misjudgment of subsequent control; the embodiment of the present invention separates the effective time judgment of the policy from the traffic forwarding process for parallel calculation, so that in the forwarding process, only the AppID needs to be controlled. Therefore, the time complexity of the entire forwarding control algorithm is actually O(1), which greatly reduces the delay of edge cloud gateway traffic forwarding. Under the laboratory's 70Gbps and 15MPPS traffic throughput, the average forwarding delay is less than 150us, and the maximum forwarding delay is 250us, which improves the forwarding performance by 85% compared with the original solution.

[0084] Embodiment 3:

[0085] Based on the efficient flow control method of the edge cloud gateway provided in the above embodiment 1, the present invention also provides an efficient flow control device of the edge cloud gateway that can be used to implement the above method and system, such as Figure 6 , which is a schematic diagram of the device architecture of an embodiment of the present invention. The efficient flow control device of the edge cloud gateway of this embodiment includes one or more processors 21 and a memory 22. Among them, Figure 6 A processor 21 is taken as an example.

[0086] The processor 21 and the memory 22 may be connected via a bus or other means. Figure 6 The example of connecting through bus is taken in the following.

[0087] The memory 22, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as the efficient flow control method of the edge cloud gateway in Example 1. The processor 21 executes various functional applications and data processing of the efficient flow control device of the edge cloud gateway by running the non-volatile software programs, instructions and modules stored in the memory 22, that is, the efficient flow control method of the edge cloud gateway of Example 1 is implemented.

[0088] The memory 22 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 22 may optionally include a memory remotely arranged relative to the processor 21, and these remote memories may be connected to the processor 21 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0089] The program instructions / modules are stored in the memory 22, and when executed by one or more processors 21, the efficient flow control method of the edge cloud gateway in the above embodiment 1 is executed, for example, the above described Figure 2-Figure 4 The steps shown.

[0090] The above product can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the method provided by the embodiment of the present invention.

[0091] It should be noted that the device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0092] Through the description of the above implementation methods, ordinary technicians in this field can clearly understand that each implementation method can be implemented by means of software plus a general hardware platform, and of course, it can also be implemented by hardware. Ordinary technicians in this field can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0093] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may also be combined, the steps may be implemented in any order, and there are many other changes in different aspects of the present invention as described above, which are not provided in detail for the sake of simplicity. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An efficient flow control method for an edge cloud gateway, characterized in that: include: Generate an AI algorithm based on the traffic characteristics of the administrative region, and generate an administrative region model library after training the actual administrative region traffic through the AI ​​algorithm; Convert the original policy combining administrative regions, time periods, and apps into execution rules: Rule[administrative region ID][AppID]; After receiving the user's network traffic, the edge cloud gateway identifies the AppID based on the traffic characteristics and identifies the administrative region ID in the administrative region model library; According to the identified administrative area ID and AppID, it is determined whether Rule [administrative area ID] [AppID] meets the control conditions. If so, flow control is performed and then transferred out of the edge cloud gateway. Otherwise, flow control is not performed.

2. The efficient flow control method of the edge cloud gateway according to claim 1 is characterized in that: The AI ​​algorithm generated according to the traffic characteristics of the administrative area specifically includes: Based on the characteristics of administrative area traffic: MAC, source IP location, requested DNS location, and App traffic curve, a weighted regression model is formed, and finally an AI algorithm for traffic fingerprint recognition of administrative areas is generated.

3. The efficient flow control method of the edge cloud gateway according to claim 2 is characterized in that: The generating of the administrative area model library after training the actual administrative area traffic through the AI ​​algorithm specifically includes: generating the administrative area model library after performing AI training on the administrative area traffic through the administrative area access network.

4. The efficient flow control method of the edge cloud gateway according to claim 1 is characterized in that: The conversion of the original policy combining administrative regions, time periods and apps into execution rules Rule[administrative region ID][AppID] specifically includes: Get the current time and date, and match the original policy in effect in a cyclic manner according to the current time and date; Get the administrative region ID and AppID in the original policy and set them to the Rule[Administrative Region ID][AppID] array.

5. The efficient flow control method of the edge cloud gateway according to claim 4 is characterized in that: In the process of converting the original policy combining the administrative area, time period and App into the execution rule Rule[administrative area ID][AppID], the task is executed according to the cycle.

6. The efficient flow control method for edge cloud gateway according to any one of claims 1 to 5, characterized in that: The determining whether Rule [administrative region ID] [AppID] meets the control condition according to the identified administrative region ID and AppID specifically includes: Substitute the identified administrative region ID and AppID into Rule[administrative region ID][AppID]. If the substituted Rule[administrative region ID][AppID] is greater than 0, the control condition is met and flow control is performed. Substitute the identified administrative region ID and AppID into Rule[administrative region ID][AppID]. If the substituted Rule[administrative region ID][AppID] is not greater than 0, the control condition is not met and traffic control is not performed.

7. The efficient flow control method for edge cloud gateway according to any one of claims 1 to 5, characterized in that: The reception and transmission of the traffic are both achieved through physical ports.

8. An efficient flow control device for an edge cloud gateway, characterized in that: The device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the processor to execute the efficient traffic control method for the edge cloud gateway described in any one of claims 1-7.

9. An efficient traffic control system for an edge cloud gateway, using the efficient traffic control method for an edge cloud gateway as described in any one of claims 1 to 7, characterized in that: The system includes a traffic identification unit, a traffic control unit, an AI training unit, and a strategy decomposition unit. The traffic identification unit includes an administrative area traffic fingerprint model matching module and an App analysis and identification module, wherein: The AI ​​training unit is used to generate an AI algorithm according to the traffic characteristics of the administrative area, generate an administrative area model library after AI training, and load the administrative area model library into the administrative area traffic fingerprint model matching module of the traffic identification unit; The policy decomposition unit is used to convert the original policy combining the administrative area, time period and App into an execution rule Rule[administrative area ID][AppID]; The App analysis and identification module of the traffic identification unit is used to identify the AppID according to the traffic characteristics after the edge cloud gateway receives the user network traffic, and at the same time transfer it to the administrative area traffic fingerprint model matching module to perform administrative area identification, and identify the administrative area ID in the administrative area model library according to the AppID; The traffic control unit is used to determine whether Rule[administrative area ID][AppID] meets the control conditions based on the identified administrative area ID and AppID. If so, traffic control is executed and then transferred out of the edge cloud gateway. Otherwise, traffic control is not executed.

10. A non-volatile computer storage medium, characterized in that: The computer storage medium stores computer executable instructions, which are executed by one or more processors to complete the efficient traffic control method for the edge cloud gateway as described in any one of claims 1-7.

Images