SD-WAN management traffic transmission optimization method and device, medium and program product

Abstract

The embodiment of the invention provides an SD-WAN management traffic transmission optimization method and device, a medium and a program product, and relates to the technical field of data transmission. The method comprises the following steps: acquiring locally collected original management flow data; processing the original management flow data according to different processing strategies to obtain various types of data to be transmitted; distributing the to-be-transmitted data of each category to a plurality of preset transmission queues; wherein the plurality of transmission queues at least comprise a first priority queue and a second priority queue; and sending the data to be transmitted to an SD-WAN controller based on the plurality of transmission queues. According to the embodiment of the invention, the data is classified and processed, and the queues with different priorities are adopted to transmit the data, so that the data transmission efficiency can be improved on the premise of ensuring the monitoring real-time performance.

Description

Technical Field

[0001] This application relates to the field of data transmission technology, and more specifically, to an SD-WAN management traffic transmission optimization method, device, medium, and program product. Background Technology

[0002] As SD-WAN (Software-Defined Wide Area Network) networks expand in scale (tens of thousands of CPEs) and real-time requirements increase (such as millisecond-level fault detection), the amount of management and monitoring data that needs to be exchanged between SD-WAN controllers and CPEs (Customer Premise Equipment) increases dramatically.

[0003] Currently, in order to save bandwidth, the polling interval is generally extended for monitoring data collection and transmission. However, this method sacrifices the real-time performance and granularity of monitoring, making it difficult to meet the needs of rapid fault detection and accurate optimization. Alternatively, the raw management traffic data can be compressed. However, the data compression and decompression process itself requires a lot of computing resources and time, resulting in low data transmission efficiency.

[0004] Therefore, a solution is needed that can improve data transmission efficiency while ensuring real-time monitoring. Summary of the Invention

[0005] The purpose of this application is to provide an SD-WAN management traffic transmission optimization method, device, medium, and program product to improve data transmission efficiency while ensuring real-time monitoring.

[0006] In a first aspect, embodiments of this application provide an SD-WAN management traffic transmission optimization method, applied to a user-end device, comprising: Obtain locally collected raw management traffic data; The original management traffic data is processed according to different processing strategies to obtain various types of data to be transmitted; The data to be transmitted for each category is allocated to a plurality of preset transmission queues; wherein the plurality of transmission queues include at least a first priority queue and a second priority queue; The data to be transmitted is sent to the SD-WAN controller based on the multiple transmission queues.

[0007] In this embodiment of the application, by classifying and processing data and transmitting data using queues with different priorities, data transmission efficiency can be improved while ensuring real-time monitoring.

[0008] In some embodiments, the raw management traffic data is processed according to different processing strategies to obtain multiple categories of data to be transmitted, including: The original management traffic data is subjected to state detection according to a preset state detection strategy, and incremental data to be transmitted is generated based on the state detection results. The original management traffic data is aggregated according to a preset data aggregation strategy to generate corresponding summary data to be transmitted.

[0009] In this embodiment, incremental data is generated through status detection, and summary data is generated through data aggregation, which can improve data transmission efficiency while ensuring the real-time and comprehensive nature of monitoring.

[0010] In some embodiments, the first priority queue has a higher priority than the second priority queue; The step of allocating data to be transmitted according to different categories to multiple preset transmission queues includes: The incremental data to be transmitted is allocated to the first priority queue; The summary data to be transmitted is assigned to the second priority queue.

[0011] In this embodiment, by using a higher priority queue for incremental data and a lower priority queue for summary data, it is possible to better balance the real-time nature of event monitoring and the comprehensiveness of status monitoring.

[0012] In some embodiments, performing state detection on the original management traffic data according to a preset state detection strategy and generating incremental data to be transmitted based on the state detection results includes: The original management traffic data is subjected to state detection according to a preset state detection strategy to obtain the state detection result. If the state detection result indicates that a preset state change event has occurred, corresponding first incremental data to be transmitted is generated. If the state detection result indicates that a preset key abnormal event has occurred, a corresponding second incremental data to be transmitted is generated. The step of allocating the incremental data to be transmitted to the first priority queue includes: The first incremental data to be transmitted is allocated to the first priority queue; The second incremental data to be transmitted is allocated to a preset third priority queue; wherein the priority of the third priority queue is higher than that of the first priority queue.

[0013] In this embodiment of the application, by further monitoring key abnormal events based on the status detection results and using the highest priority queue to transmit key abnormal event data, the real-time performance of monitoring is further improved.

[0014] In some embodiments, sending the data to be transmitted to the SD-WAN controller based on the plurality of transmission queues includes: Based on the preset queue buffering strategy, the data to be transmitted in the first priority queue is sent to the SD-WAN controller; Based on the preset bandwidth shaping strategy, the data to be transmitted in the second priority queue is sent to the SD-WAN controller; The data to be transmitted in the third priority queue is sent to the SD-WAN controller based on the preset instant transmission strategy.

[0015] In the embodiments of this application, by employing queue buffering, bandwidth shaping, and instant transmission strategies to transmit data for queues of different priorities, the flexibility of traffic transmission management is further improved.

[0016] In some embodiments, the step of aggregating the original management traffic data according to a preset data aggregation strategy to generate corresponding summary data to be transmitted includes: Based on preset data classification rules, the log event data in the original management traffic data is filtered and classified, and corresponding summary data to be transmitted is generated based on the filtered and classified data. The time-series data in the original management traffic data is segmented and aggregated based on a preset data aggregation algorithm, and corresponding summary data to be transmitted is generated based on the segmented and aggregated data.

[0017] In this embodiment of the application, the flexibility of data simplification is further improved by simplifying log event data and time series data to generate summary data respectively.

[0018] In some embodiments, sending the data to be transmitted to the SD-WAN controller based on the plurality of transmission queues includes: The key information of the data to be transmitted is extracted according to a preset data template, and key data to be transmitted corresponding to the data to be transmitted is generated based on the key information; wherein, the data template is issued by the SD-WAN controller to the user terminal device; The critical data to be transmitted is sent to the SD-WAN controller.

[0019] In this embodiment of the application, by extracting key information according to a preset data template to generate key data for transmission, the data to be transmitted can be further simplified, and the data transmission efficiency can be further improved.

[0020] In some embodiments, sending the critical data to be transmitted to the SD-WAN controller includes: The key data to be transmitted is encoded according to a preset encoding rule to obtain encoded data to be transmitted; The encoded data to be transmitted is sent to the SD-WAN controller.

[0021] In this embodiment of the application, the efficiency of data transmission is further improved by encoding the data to be transmitted before transmission.

[0022] In some embodiments, the SD-WAN management traffic transmission optimization method further includes: In response to a detailed data fetch request sent by the SD-WAN controller; Based on the original management traffic data, filter the target detailed data corresponding to the detailed data retrieval request; The target detailed data is assigned to the highest priority target priority queue, and the target detailed data is sent to the SD-WAN controller based on the target priority queue.

[0023] In this embodiment of the application, by setting up the function of actively pulling detailed data, the flexibility of traffic monitoring can be further improved.

[0024] Secondly, embodiments of this application provide an SD-WAN management traffic transmission optimization device, applied to a user-end device, comprising: The data acquisition module is used to acquire raw management traffic data collected locally. The classification processing module is used to process the original management traffic data according to different processing strategies to obtain multiple categories of data to be transmitted; The queue allocation module is used to allocate data of various categories to multiple preset transmission queues; wherein, the multiple transmission queues include at least a first priority queue and a second priority queue. The data transmission module is used to send the data to be transmitted to the SD-WAN controller based on the multiple transmission queues.

[0025] Thirdly, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, can implement the method described in any embodiment of the first aspect.

[0026] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, can implement the method described in any embodiment of the first aspect.

[0027] Fifthly, embodiments of this application provide a computer program product, the computer program product including a computer program, wherein when the computer program is executed by a processor, it can implement the method described in any embodiment of the first aspect. Attached Figure Description

[0028] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 A flowchart illustrating an SD-WAN management traffic transmission optimization method provided in this application embodiment; Figure 2 A flowchart illustrating the traffic transmission process for managing existing SD-WAN technologies; Figure 3 This is a diagram illustrating the architecture of an SD-WAN management traffic transmission system provided in an embodiment of this application. Figure 4 A flowchart of CPE-side processing and transmission management provided for embodiments of this application; Figure 5 A schematic diagram of the differentiated transmission scheduling mechanism provided in the embodiments of this application; Figure 6 A schematic diagram of the structure of an SD-WAN management traffic transmission optimization device provided in an embodiment of this application; Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0030] The technical solutions in the embodiments of this application will now be described with reference to the accompanying drawings.

[0031] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0032] It should be noted that an SD-WAN controller is a centralized management software (usually provided as a virtualized device or cloud service) responsible for unified policy distribution, configuration management, monitoring, and optimization of devices throughout the SD-WAN network. CPE (Customer Premise Equipment) refers to network devices provided by service providers and installed at the user's end (e.g., home, enterprise). It is the final interface device connecting the user's internal network to the operator's wide area network.

[0033] Data between the CPE and the SD-WAN controller is typically transmitted through the SD-WAN service tunnel itself (overlay tunnel). For example... Figure 2 As shown, the SD-WAN controller is connected to multiple CPEs through an overlay tunnel. The SD-WAN controller periodically sends polling requests to each CPE. Based on the polling requests, the CPE devices continuously report complete status data, resulting in high-bandwidth management traffic.

[0034] As SD-WAN networks scale up (tens of thousands of CPEs) and real-time requirements increase (such as millisecond-level fault detection), the management traffic between the SD-WAN controller and the CPEs increases dramatically, leading to the following problems: 1. Management traffic and business traffic compete for limited bandwidth resources in the transmission tunnel. Management traffic can account for up to 30% on average. Especially on low-bandwidth links (such as Internet VPN), management traffic may significantly crowd out the available bandwidth of business applications and affect user experience. 2. The queuing and transmission of large amounts of management data may increase the delay of critical management instructions (such as rapid failover strategies); 3. The CPE and controller need to handle high-frequency, large-volume polling and reporting messages, which severely consumes CPU and memory resources; 4. For traffic-based billing links (such as 4G / 5G), unnecessary management data transmission will increase operating costs.

[0035] To save bandwidth, existing solutions generally employ the following technical measures: adjusting the polling interval to reduce the amount of data; providing raw data compression to compress data packets before transmission; or using data sampling to report only a portion of the sample data.

[0036] However, the existing solutions mentioned above have the following shortcomings: 1. Although extending the polling interval can reduce the amount of data transmitted, it sacrifices the real-time performance and granularity of monitoring, making it difficult to meet the needs of rapid fault detection and accurate optimization; 2. Compressing data packets before transmission can reduce bandwidth usage to some extent, but compression / decompression itself consumes computing resources, and the compression rate is limited for highly dynamic data. 3. If a dedicated management channel is used to establish a physical or logical channel (Underlay management channel) independent of the business tunnel for management data, additional network configuration and costs (such as dedicated MPLS lines or cellular links) are required, increasing the complexity of the architecture and deployment costs, and it is not suitable for all scenarios; 4. Using data sampling and transmission methods, only reporting a portion of the sample data, will result in the loss of important details and affect the accuracy of centralized analysis by the controller.

[0037] To address the problems existing in the prior art, this application provides an SD-WAN management traffic transmission optimization method. This method performs intelligent local data processing at the CPE end (such as change detection, event triggering, incremental reporting, and data aggregation summary generation), and performs differentiated transmission of categorized data (such as data layering, priority scheduling, and bandwidth limiting). It also supports the controller in pulling detailed data on demand and uses optimized lightweight protocols for data transmission. This significantly reduces the bandwidth consumption of management traffic on service traffic, effectively improving the user's service experience. Furthermore, an event-driven mechanism ensures millisecond-level reporting of critical faults, improving network reliability and real-time monitoring, reducing equipment and link resource consumption and operating costs, thereby enhancing the system's large-scale deployment capabilities.

[0038] like Figure 1 As shown in the figure, this application provides an SD-WAN management traffic transmission optimization method, applied to user terminal equipment, which may include the following steps: S1. Obtain locally collected raw management traffic data; First, obtain the raw data (management traffic data) collected locally by the CPE, including raw performance metrics, status information, log events, etc.

[0039] S2. Process the raw management traffic data according to different processing strategies to obtain various types of data to be transmitted; Then, the raw management traffic data is processed according to different processing strategies (at least two). For example, on the one hand, the raw management traffic data is subjected to status detection through a preset detection strategy to obtain the corresponding detection data; on the other hand, the raw management traffic data is subjected to statistical analysis based on a preset summary strategy to obtain the corresponding summary data.

[0040] It is understandable that different processing strategies can result in multiple categories of data to be transmitted, with one processing strategy corresponding to one category of data to be transmitted.

[0041] S3. Distribute the data to be transmitted for each category to multiple preset transmission queues; wherein the multiple transmission queues include at least a first priority queue and a second priority queue; For example, at least two transmission queues can be set up, each responsible for managing the transmission of different categories of data to be transmitted. It should be noted that the first priority queue and the second priority queue are two transmission queues with different priorities, and each of the first priority queue and / or the second priority queue can contain one or more transmission queues.

[0042] For example, the mapping relationship between the data to be transmitted and the transmission queue can be pre-set for each category. In this way, after the data to be transmitted for the corresponding category is obtained through classification processing, it will be automatically assigned to the corresponding transmission queue.

[0043] For example, data of the same category to be transmitted can only be assigned to transmission queues of the same priority, while data of different categories to be transmitted can be assigned to transmission queues of the same priority.

[0044] S4. Send the data to be transmitted to the SD-WAN controller based on multiple transmission queues.

[0045] Finally, multiple transmission queues are used to send the corresponding data to be transmitted to the SD-WAN controller. Understandably, different transmission queues employ different transmission strategies, including different bandwidth limits, transmission intervals, and transmission protocols.

[0046] In this embodiment, by classifying the data collected locally by the CPE and using queues with different priorities for differentiated transmission of the classified data, data transmission efficiency can be improved while ensuring real-time monitoring.

[0047] In some embodiments, step S2, processing the original management traffic data according to different processing strategies to obtain various types of data to be transmitted, may include: S201. Perform status detection on the original management traffic data according to the preset status detection strategy, and generate incremental data to be transmitted based on the status detection results. S202. Aggregate the original management traffic data according to the preset data aggregation strategy to generate corresponding summary data to be transmitted.

[0048] It should be noted that the classification process is mainly divided into two parts. One part is to perform status detection on the original management traffic data according to the preset status detection strategy, and the other part is to perform aggregation processing on the original management traffic data according to the preset data aggregation strategy.

[0049] For example, according to a preset state detection strategy, state detection can be performed on raw management traffic data such as raw performance indicators and state information. For example, based on a set threshold, when the value or rate of change of a relevant indicator is detected to exceed the preset threshold, a major state change event is triggered, and corresponding incremental data (incremental data to be transmitted) is generated.

[0050] For example, according to a preset data aggregation strategy, high-frequency collected time-series data or event data can be aggregated. Through data segmentation statistics, high-frequency data sequences can be simplified into low-frequency summary data (summary data to be transmitted).

[0051] Based on this, incremental data is generated through status detection, and summary data is generated through data aggregation, which can improve data transmission efficiency while ensuring the real-time and comprehensive nature of monitoring.

[0052] In some embodiments, the first priority queue has a higher priority than the second priority queue; Step S3, which involves allocating the data to be transmitted for each category to multiple preset transmission queues, may include: S301. Allocate the incremental data to be transmitted to the first priority queue; S302. Assign the summary data to be transmitted to the second priority queue.

[0053] It is understandable that, since the incremental data to be transmitted reflects significant state changes in the original monitoring traffic, the incremental data to be transmitted generated based on state detection can be assigned to the first priority queue, which has a relatively high priority.

[0054] For summary data to be transmitted, since it reflects the long-term stable characteristics of monitoring traffic, it can be assigned to the second priority queue with a relatively lower priority.

[0055] Based on this, by using a higher priority queue for incremental data and a lower priority queue for summary data, we can ensure the high real-time monitoring requirements for state change events, while also taking into account the comprehensiveness of long-term state indicators, effectively reducing the bandwidth consumption of overall monitoring traffic data, and improving data transmission efficiency.

[0056] In some embodiments, step S201, which involves performing state detection on the original management traffic data according to a preset state detection strategy and generating incremental data to be transmitted based on the state detection results, may include: S201-1. Perform status detection on the original management traffic data according to the preset status detection strategy, and obtain the status detection results; S201-2. If the state detection result indicates that a preset state change event has occurred, generate the corresponding first incremental data to be transmitted. S201-3. If the state detection result indicates that a preset key abnormal event has occurred, generate the corresponding second incremental data to be transmitted. Step S301, allocating incremental data to be transmitted to the first priority queue, may include: S301-1. Allocate the first incremental data to be transmitted to the first priority queue; S301-2. Allocate the second incremental data to be transmitted to a preset third priority queue; wherein the priority of the third priority queue is higher than that of the first priority queue.

[0057] It should be noted that the process of performing state detection according to the preset state detection strategy can be further divided into two parts: 1. If a preset state change event is determined to have occurred based on the state detection result, such as the indicator value or the rate of change of the indicator exceeding the threshold, then it is determined that a state change event has occurred. At this time, the corresponding first incremental data to be transmitted is generated based on the specific indicator value or the amount of change. 2. If a preset critical abnormal event is determined to have occurred based on the status detection results, such as the indicator value or rate of change exceeding the preset abnormal threshold, or a specific fault or emergency event is detected, it is determined to be a critical abnormal event. At this time, the corresponding second incremental data to be transmitted can be generated based on the event type, occurrence time, current indicator value, and other information.

[0058] Understandably, critical anomaly events reflect more urgent and pressing issues than state change events. Therefore, it is necessary to prioritize transmitting the second incremental data to be transmitted for critical anomaly events to the SD-WAN controller. Specifically, the first incremental data to be transmitted can be assigned to the first priority queue, and the second incremental data to be transmitted can be assigned to a preset third priority queue (which has a higher priority than the first priority queue).

[0059] Based on this, by further monitoring key abnormal events based on the status detection results and using the highest priority queue to transmit key abnormal event data, the real-time performance of monitoring is further improved.

[0060] In some embodiments, step S4, sending the data to be transmitted to the SD-WAN controller based on multiple transmission queues, may include: S401. Based on the preset queue buffering strategy, the data to be transmitted in the first priority queue is sent to the SD-WAN controller; S402. Based on the preset bandwidth shaping strategy, the data to be transmitted in the second priority queue is sent to the SD-WAN controller; S403. Based on the preset instant transmission strategy, the data to be transmitted in the third priority queue is sent to the SD-WAN controller.

[0061] It is understandable that different transmission strategies are needed for queues with different priorities. For example, for a transmission queue with a higher priority, more bandwidth resources are allocated for its transmission and the transmission frequency is higher.

[0062] For example, the third priority queue, the first priority queue, and the second priority queue correspond to the high priority queue, the medium priority queue, and the low priority queue, respectively.

[0063] For example, a high-priority queue is used to store critical event data, a medium-priority queue is used to store important status change data, and a low-priority queue is used to store periodic summary data.

[0064] For example, the scheduler prioritizes sending data to the high-priority queue, then the medium-priority queue, and finally the low-priority queue.

[0065] For the first priority queue, a normal queue buffering strategy can be used to transmit the data in sequence according to the preset transmission frequency and preset data packet size.

[0066] For the second priority queue, which is the lowest priority transmission queue, a bandwidth shaper can be used, such as a token bucket algorithm, to shape the bandwidth for data transmission. It can be understood that the transmission strategy for the second priority queue acts as a "traffic policeman," its core function being to strictly limit the rate of low-priority periodic management data, preventing it from unrestrainedly consuming valuable bandwidth resources of the SD-WAN service tunnel.

[0067] For the third priority queue, which is the highest priority transmission queue, an immediate transmission strategy is mainly adopted. This means maximizing bandwidth resources for data transmission to ensure that important changes or abnormal event information can be transmitted to the controller as soon as possible. For example, when a large amount of data to be transmitted is generated simultaneously in the third priority queue, a multi-channel parallel transmission method can be further used for data transmission to ensure efficient data delivery.

[0068] Based on this, by employing queue buffering, bandwidth shaping, and instant transmission strategies to transmit data for queues of different priorities, the flexibility of traffic transmission management is further improved.

[0069] In some embodiments, step S202, which involves aggregating the original management traffic data according to a preset data aggregation strategy to generate corresponding summary data to be transmitted, may include: S202-1. Based on preset data classification rules, the log event data in the original management traffic data is filtered and classified, and corresponding summary data to be transmitted is generated based on the filtered and classified data. S202-2. Based on the preset data aggregation algorithm, the time series data in the original management traffic data is segmented and aggregated, and the corresponding summary data to be transmitted is generated based on the segmented and aggregated data.

[0070] Specifically, the summary data to be transmitted obtained by aggregating the original management traffic data can be divided into two parts: 1. For log event data, the raw collected data can be filtered and classified. For example, redundant data such as non-critical events can be filtered out, and the same events within a certain period can be classified and simplified as summary data to be transmitted. 2. For time-series data in the original management traffic data, such as indicator curve data (discrete data) collected at a preset frequency, it can be segmented and aggregated based on a preset data aggregation algorithm to serve as a summary for transmission. For example, a preset data simplification algorithm can be applied, such as the Piecewise Aggregate Approximation algorithm that preserves key inflection points. This is an algorithm widely used in time series analysis, data mining, and signal processing; it is a method for dimensionality reduction and feature representation of time series. Segmentation refers to dividing the entire time series into continuous subsequences (segments) of equal length; aggregation refers to performing aggregation calculations on the data points of each subsequence (usually calculating the average value of the data within that segment); and approximation refers to using the aggregated value of each segment to represent all the original data points of that segment, thus approximating the entire original sequence with fewer data points.

[0071] It should be noted that for high-frequency collected raw data (such as traffic count per second), a segmented aggregation processing strategy can be adopted. Specifically, based on a specified time window (such as 1 minute or 5 minutes), the average, maximum, minimum, sum, standard deviation and other statistical measures of the raw data segments are calculated to generate corresponding summary data packets. Thus, the original, high-frequency and highly redundant detailed data can be replaced by these aggregated or simplified summary information.

[0072] In this embodiment of the application, the flexibility of data simplification is further improved by simplifying log event data and time series data to generate summary data respectively.

[0073] In some embodiments, step S4, sending the data to be transmitted to the SD-WAN controller based on multiple transmission queues, may include: S411. Extract key information of the data to be transmitted according to the preset data template, and generate key data to be transmitted corresponding to the data to be transmitted based on the key information; wherein, the data template is sent from the SD-WAN controller to the user terminal device; S412. Send the critical data to be transmitted to the SD-WAN controller.

[0074] It should be noted that for data from each transmission queue, an optimized lightweight protocol can be used to encapsulate the data before transmission.

[0075] Specifically, the corresponding data template can be obtained according to the transmission policy sent by the SD-WAN controller, and then the key information of the data to be transmitted can be extracted based on the data template. In this way, there is no need to transmit the original data to be transmitted, which further simplifies the data that needs to be transmitted.

[0076] Based on this, by extracting key information according to a preset data template to generate key data for transmission, the data to be transmitted can be further simplified, thus further improving data transmission efficiency.

[0077] In some embodiments, step S412, sending the key data to be transmitted to the SD-WAN controller, may include: S412-1. Encode the key data to be transmitted according to the preset encoding rules to obtain the encoded data to be transmitted; S412-2, Send the encoded data to be transmitted to the SD-WAN controller.

[0078] It should be noted that by adopting the Optimized Lightweight Management Protocol (OLMP), the pain points of traditional network management protocols (such as SNMP polling and the redundancy of NETCONF / XML) in SD-WAN scenarios can be completely resolved. 1. Extremely low bandwidth usage: Minimizes the number of bytes per data packet.

[0079] 2. Extremely low processing overhead: On resource-constrained CPE devices, the computational cost of encoding and decoding is lower.

[0080] 3. High real-time performance: Supports event-driven asynchronous notifications and efficient bidirectional communication.

[0081] 4. Strong scalability: It can flexibly add new monitoring metrics or management functions.

[0082] The embodiments of this application can achieve the above-mentioned optimization objectives from multiple levels: First: Binary encoding is used instead of the original text format.

[0083] It should be noted that the traditional method uses text formats such as JSON or XML for data serialization. For example, data based on the serial number in the traditional method is: {"device_id": "cpe-branch-01", "metric": "cpu_util", "value": 65, "timestamp": 1692081234}.

[0084] As can be seen, the above method suffers from the problem of repeated transmission of field names (such as device_id, metric) and excessive space consumption. Since characters such as parentheses, quotation marks, and colons are all additional overhead, a simple status report can easily exceed 100 bytes.

[0085] This application embodiment can employ a binary encoding protocol, such as Protocol Buffers (Protobuf) or MessagePack. By pre-defining a data structure schema between the sender and receiver, only the data value and a very short field tag can be transmitted during data transmission, instead of the complete field name, for example: 1. First, define the pattern file (.proto) and the data structure for the reported messages: message TelemetryReport { string device_id = 1; field label = 1 string metric = 2; field label = 2 float value = 3; field label = 3 int64 timestamp = 4; field label = 4 } 2. The encoded binary data used for transmission is as follows: Device ID cpe-branch-01 is encoded as: 0A 0D 63 70 65 2D 62 72 61 6E 63 68 2D30 31 (Tag=1, Type=2, Length=13, Value="cpe-branch-01") CPU utilization of 65 is encoded as: 15 00 00 82 42 (Tag=3, Type=5, Value=65.0) Based on this, the size of the entire data packet after the above encoding may only be 20-30 bytes, which is 60%-80% smaller than the JSON format.

[0086] second: Key information can be extracted and transmitted based on TLV (Tag-Length-Value) format data templates, where each data unit consists of three parts: Tag: 1-2 bytes, uniquely identifying a field (corresponding to the tag number in Protobuf); Length: 1-2 bytes, indicating the length of the Value field; Value: the actual data content.

[0087] For example, for incremental data, such as a change in CPU utilization from 65% to 66%, traditional transmission methods would require repeatedly reporting all fields (device ID, metric name, timestamp, etc.). However, using the optimized lightweight management protocol of this application embodiment, only the changed value needs to be reported: [Tag: 3][Length: 4][Value: 66.0] (assuming Tag=3 represents the value field). Since this update packet may only be 6-7 bytes, it significantly saves bandwidth resources.

[0088] third: By designing a minimalist message header and supporting batch reporting, message type simplification and aggregation can be achieved.

[0089] In traditional solutions, each request / response has complex HTTP headers (such as POST / api / telemetryHTTP / 1.1...) or SOAP envelopes.

[0090] The lightweight protocol adopted in this application embodiment can fix the message header to 2-4 bytes, which includes: version number (2 bits); message type (4 bits): such as 0x1=heartbeat, 0x2=event, 0x3=aggregated data, 0x4=configuration response; sequence number (1 byte): used for request-response matching.

[0091] Then, aggregated reporting is achieved by packaging multiple monitoring metrics into a single message. For example, a message can contain TLV structures for multiple metrics such as CPU, memory, and link latency, which can share the same device ID and timestamp (or carry their own), avoiding the overhead of establishing independent connections and transmitting messages for each metric.

[0092] fourth: Lightweight compression of specific, highly redundant data.

[0093] For example, a dictionary can be created for commonly used strings (such as device ID, metric name cpu_util) using dictionary encoding, and only the dictionary index (such as a 1-2 byte integer) is transmitted instead of the complete string.

[0094] Alternatively, lossy compression can be used to preprocess the time-series monitoring data at the CPE end, transmitting only the maximum, minimum, average, and sampling points within a certain period, instead of all the original data.

[0095] The following specific examples illustrate the optimization process of the lightweight protocol described above: Let's assume the current scenario is: CPE reports a sudden change in CPU utilization.

[0096] The workflow is as follows: 1. Event Trigger: The CPE locally detected that the CPU utilization jumped from 40% to 70% (Δ>15% threshold).

[0097] 2. Construct the message: The protocol stack uses a predefined Protobuf schema (data template) to construct an event message.

[0098] Extract key data and populate the event message with device_id, metric="cpu_util", value=70.0, and timestamp.

[0099] The padded event message is serialized into a binary format, approximately 25 bytes in size.

[0100] 3. Encapsulation: Add a minimal OLMP header (2 bytes, indicating that this is a 0x2 event type).

[0101] 4. Transmission: Place the approximately 27-byte data packet into a high-priority queue and immediately send it to the SS-WAN controller.

[0102] 5. Parsing: After receiving the data, the controller quickly deserializes it according to the same schema (data template) to obtain clear structured data, triggering alarms or policy adjustments.

[0103] In this embodiment of the application, the efficiency of data transmission is further improved by encoding the data to be transmitted before transmission.

[0104] It should be noted that, compared with traditional protocols (SNMP / HTTP+JSON), the optimized lightweight protocol (OLMP) has the following advantages: 1. Using binary encoding (such as Protobuf) can reduce bandwidth by 60-80%; 2. Supports predefined data templates (only Tag+Value is transmitted), enabling extremely simple messages and high parsing efficiency; 3. Supports event-driven and incremental update methods, which can effectively save bandwidth and improve monitoring real-time performance; 4. The message definition is simple, with a minimal custom header (2-4 bytes) and minimal overhead; 5. Direct parsing of binary streams is extremely low-cost and suitable for resource-constrained CPEs.

[0105] Therefore, the lightweight management protocol of this application adopts a combination of binary encoding format and predefined patterns. The two communicating parties serialize and deserialize the data to be transmitted based on the pre-issued pattern. During transmission, only field tags and corresponding values ​​are transmitted, eliminating redundant field names and format characters in traditional text protocols. At the same time, the protocol supports incremental encoding mechanism, only reporting data fields whose status has changed, and further compresses highly repetitive metadata using lossless compression techniques such as dictionary encoding, thereby reducing the volume of transmitted data by more than 70%. The protocol header is simplified to less than 4 bytes, containing core information such as version, message type, and sequence number, which greatly reduces protocol overhead.

[0106] In some embodiments, the SD-WAN management traffic transmission optimization method further includes: In response to a detailed data fetch request sent by the SD-WAN controller; Filter the target detailed data corresponding to the detailed data retrieval request based on the original management traffic data; Target detail data is assigned to the highest priority target priority queue, and then the target detail data is sent to the SD-WAN controller based on the target priority queue.

[0107] It should be noted that the SD-WAN controller can choose to actively retrieve detailed data based on the data uploaded by the CPE, according to its needs. Specifically, the SD-WAN controller can send a detailed data retrieval request to the CPE, specifying the device ID, data type, time range, specific metrics, and specific log level to be viewed.

[0108] Upon receiving a detailed data retrieval request, the CPE can filter the corresponding target detailed data from the raw management traffic data collected locally and send it to the SD-WAN controller through the highest priority transmission queue.

[0109] Therefore, by setting up the function of actively pulling detailed data, the flexibility of traffic monitoring can be further improved.

[0110] The following examples illustrate the scheme of this application: like Figure 3 As shown, the solution in this application mainly includes an SD-WAN controller and a CPE (customer premises equipment).

[0111] The SD-WAN controller mainly includes a policy engine (for configuring and issuing optimization policies), an intelligent request module (for requesting data on demand), and a data reconstruction engine (for processing received data).

[0112] The strategy engine is used to define and distribute data classification rules, detection thresholds, aggregation strategies, reporting strategies, priority strategies, and other information to various CPEs.

[0113] The intelligent request module handles requests to retrieve detailed data on demand. The controller only proactively requests more detailed data (such as raw logs, complete configuration snapshots, real-time packet captures, etc.) from the CPE when needed (e.g., when a user views specific device details or performs troubleshooting). The data requests issued by the controller can precisely specify the required data range (time range, specific metrics, specific log levels, etc.), avoiding the CPE returning a large amount of unnecessary data.

[0114] The data reconstruction engine receives data transmitted from the CPE and restores and reconstructs it according to a preset transmission protocol, forming comprehensive monitoring data for analysis. It can process data based on priority, for example, prioritizing incremental data triggered by critical events before processing aggregated data. Based on the reconstructed data, it processes event alarms and stores aggregated data.

[0115] CPE devices include a local data collector (collecting raw data), an edge processing engine (including change detectors, aggregators, etc.), and a differentiated transmission module (including classifiers, priority queues, and bandwidth shapers).

[0116] The local data collector is used to collect raw performance metrics, status information, log events, and other data.

[0117] The edge processing engine is used to implement state change detection, data aggregation (calculating statistics and generating summaries), data simplification, event detection, log filtering, etc.

[0118] The differentiated transmission module includes the following functions: 1. Maintain data queues with different priorities.

[0119] 2. Implement event-driven and incremental reporting logic.

[0120] 3. Classify the data and map it to a priority queue.

[0121] 4. Implement bandwidth shaping for low-priority data. A bandwidth shaper, especially when combined with the token bucket algorithm, can be used to act as a "traffic policeman." Its core function is to impose strict rate limits on low-priority periodic management data, ensuring that it never consumes valuable bandwidth resources of the SD-WAN service tunnel without restraint.

[0122] 5. Use an optimized, lightweight protocol to encapsulate and transmit data.

[0123] 6. Respond to the controller's on-demand data retrieval requests.

[0124] The policy executor is used to receive and execute optimized policy configurations from the controller.

[0125] The following will provide a detailed explanation with specific examples: 1. Administrators can configure policies through the management interface of the SD-WAN controller: (1) Define a list of critical events (e.g., Link Down, CPU>90%, Memory>90%, critical App packet loss rate>5%) and the corresponding trigger thresholds.

[0126] (2) Define the threshold for important state changes (e.g., CPU change ±15%, latency change ±20ms).

[0127] (3) Set the type of periodic summary data (e.g., average bandwidth, average latency, packet loss rate) and the reporting interval (e.g., 5 minutes).

[0128] (4) Set the heart rate interval (e.g., 60 seconds).

[0129] (5) Configure the transmission priority of each data category: Critical > Significant > Periodic.

[0130] (6) Set a bandwidth limit for periodic data (e.g., no more than 1% of the available tunnel bandwidth).

[0131] The configured policies are distributed to each CPE device.

[0132] 2. For example Figure 4 As shown, the data processing flow at the CPE end is as follows: (1) Continuously collect raw data (raw management traffic data); (2) Local processing: Calculate the instantaneous CPU / Memory value every 1 second; calculate the instantaneous application traffic / packet loss value every 5 seconds.

[0133] (3) Change detection: Compare the current CPU value with the previous reported value (or the average value over 5 minutes). If the absolute value of the change exceeds 15%, it is marked as a significant state change. If the Link status changes to Down, it is marked as a Critical Event. If an instantaneous packet loss rate >5% is detected, it will be immediately marked as a critical event.

[0134] (4) Data aggregation: Every 5 minutes, the average CPU utilization, average memory utilization, total traffic, average latency, maximum latency, packet loss rate, etc. are calculated and periodic aggregate data (summary data) is generated.

[0135] (5) Event Trigger: Once an emergency event or significant state change is detected, an incremental data packet containing the event type, occurrence time, and changes / current values ​​of relevant indicators is immediately generated and placed in a high-priority queue (emergency event data) or a medium-priority queue (significant state change event data).

[0136] (6) Timed triggering: Every 5 minutes, the generated periodic aggregate summary data packet is placed into the low priority queue. A simplified heartbeat packet (device ID + timestamp) is sent every 60 seconds.

[0137] (7) For example Figure 5 As shown, the scheduling and transmission process is as follows: Data in the high-priority queue is sent first (immediately), followed by data in the medium-priority queue (queue buffer sending).

[0138] Implement token bucket shaping on data in low-priority queues to ensure that their sending rate does not exceed the configured limit (e.g., 1% bandwidth).

[0139] All transmitted data is encapsulated using an optimized binary protocol (such as Protobuf), and the header is simplified.

[0140] 3. SD-WAN controller operation: (1) Receive and prioritize event packets from high-priority queues, immediately update device status, and trigger alarms or automatic repair actions.

[0141] (2) Receive and process aggregated data packets from low-priority queues for performance trend analysis and reporting.

[0142] (3) Receive heartbeat packets to confirm that the device is online.

[0143] (4) When an administrator clicks "View Real-time CPU" on a device on the interface, the controller sends an on-demand fetch request to the corresponding CPE. After receiving the request, the CPE packages the current detailed CPU, Memory and other indicators (which may include brief high-frequency sampling) and sends them once through a high-priority queue.

[0144] Based on the above example, assuming a CPE experiences no critical events or significant state changes within 5 minutes, the traditional method might require reporting detailed data 300 times (1 sample per second * 300 seconds). However, the proposed solution only requires reporting 5 heartbeat packets + 1 5-minute aggregated summary packet. While achieving at least the same monitoring effect, the proposed solution reduces the amount of data transmitted by more than 99% compared to traditional methods. Furthermore, when critical events such as link failures occur, the failure events can be reported in milliseconds, far faster than traditional polling-based monitoring transmission schemes (e.g., once every 5 seconds).

[0145] Compared with the prior art, the embodiments of this application have the following beneficial effects: 1. Significantly reduce bandwidth usage: By reducing unnecessary data transmission (no reporting when there are no changes, incremental reporting when there are changes, using summaries instead of details for periodic data, and supporting on-demand data retrieval), optimizing protocol efficiency and priority scheduling, the bandwidth usage of management and monitoring data on SD-WAN tunnels can be reduced by 50%-90%, freeing up more bandwidth for business traffic.

[0146] 2. Improve business experience: Reduce the interference of management traffic on business traffic and ensure bandwidth and low latency for critical applications.

[0147] 3. Maintain or even improve real-time monitoring: The event-driven mechanism ensures that critical faults and state changes can be detected and reported extremely quickly (milliseconds), which is faster than the traditional polling method (seconds), thus improving network reliability and response speed.

[0148] 4. Reduced resource consumption: The CPE and controller consume less CPU, memory, and storage resources due to the reduced amount of data processed, transmitted, and stored.

[0149] 5. Reduce operating costs: Reduce data transmission volume and save costs on traffic-based billing links.

[0150] 6. Enhanced scalability: Enables SD-WAN controllers to efficiently manage larger groups of CPE devices.

[0151] Please refer to Figure 6 , Figure 6 This document illustrates a block diagram of an SD-WAN management traffic transmission optimization apparatus provided in some embodiments of this application. It should be understood that this SD-WAN management traffic transmission optimization apparatus is similar to the one described above. Figure 1 Corresponding to the method embodiments, it is able to perform each step involved in the above method embodiments. The specific functions of the SD-WAN management traffic transmission optimization device can be found in the description above. To avoid repetition, detailed descriptions are appropriately omitted here.

[0152] Figure 6 The SD-WAN management traffic optimization device includes at least one software function module that can be stored in memory or embedded in the SD-WAN management traffic optimization device in the form of software or firmware. This SD-WAN management traffic optimization device is applied to user-end equipment and includes: The data acquisition module 610 is used to acquire raw management traffic data collected locally; The classification processing module 620 is used to process the raw management traffic data according to different processing strategies to obtain multiple categories of data to be transmitted. The queue allocation module 630 is used to allocate data of various categories to multiple preset transmission queues; wherein the multiple transmission queues include at least a first priority queue and a second priority queue. The data transmission module 640 is used to send data to be transmitted to the SD-WAN controller based on multiple transmission queues.

[0153] It is understood that the above-described device embodiments correspond to the method embodiments of the present invention. The SD-WAN management traffic transmission optimization device provided by the embodiments of the present invention can implement the SD-WAN management traffic transmission optimization method provided by any one of the method embodiments of the present invention.

[0154] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the device described above can be referred to the corresponding process in the aforementioned method, and will not be elaborated further here.

[0155] like Figure 7 As shown, some embodiments of this application provide an electronic device 700, which includes: a memory 710, a processor 720, and a computer program stored in the memory 710 and executable on the processor 720. When the processor 720 reads the program from the memory 710 via a bus 730 and executes the program, it can implement any of the methods included in the above-described SD-WAN management traffic transmission optimization method.

[0156] Processor 720 can process digital signals and can include various computing architectures. For example, it can be a complex instruction set computer architecture, a reduced instruction set computer architecture, or an architecture that implements multiple instruction set combinations. In some examples, processor 720 can be a microprocessor.

[0157] The memory 710 can be used to store instructions executed by the processor 720 or data related to the execution of instructions. These instructions and / or data may include code for implementing some or all of the functions of one or more modules described in the embodiments of this application. The processor 720 of this disclosure embodiment can be used to execute the instructions in the memory 710 to implement the methods shown above. The memory 710 includes dynamic random access memory, static random access memory, flash memory, optical memory, or other memories well known to those skilled in the art.

[0158] Some embodiments of this application also provide a computer-readable storage medium storing a computer program that, when executed by a processor, describes the method described in the method embodiments.

[0159] Some embodiments of this application also provide a computer program product that, when run on a computer, causes the computer to perform the methods described in the method embodiments.

[0160] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For apparatus embodiments, since they are basically similar to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0161] It should be understood, in the several embodiments provided in this application, that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative; for example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0162] In addition, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0163] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0164] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application. It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0165] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0166] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. A method for SD-WAN managed traffic transmission optimization, the method comprising: Applied to user terminal devices, including: Obtain locally collected raw management traffic data; The original management traffic data is processed according to different processing strategies to obtain various types of data to be transmitted; The data to be transmitted for each category is allocated to a plurality of preset transmission queues; wherein the plurality of transmission queues include at least a first priority queue and a second priority queue; The data to be transmitted is sent to the SD-WAN controller based on the multiple transmission queues.

2. The SD-WAN managed traffic transmission optimization method of claim 1, wherein, The original management traffic data is processed according to different processing strategies to obtain various types of data to be transmitted, including: The original management traffic data is subjected to state detection according to a preset state detection strategy, and incremental data to be transmitted is generated based on the state detection results. The original management traffic data is aggregated according to a preset data aggregation strategy to generate corresponding summary data to be transmitted.

3. The SD-WAN managed traffic transmission optimization method of claim 2, wherein, The first priority queue has a higher priority than the second priority queue; The step of allocating data to be transmitted according to different categories to multiple preset transmission queues includes: The incremental data to be transmitted is allocated to the first priority queue; The summary data to be transmitted is assigned to the second priority queue.

4. The SD-WAN managed traffic transmission optimization method of claim 3, wherein, The step of performing state detection on the original management traffic data according to a preset state detection strategy, and generating incremental data to be transmitted based on the state detection results, includes: The original management traffic data is subjected to state detection according to a preset state detection strategy to obtain the state detection result. If the state detection result indicates that a preset state change event has occurred, corresponding first incremental data to be transmitted is generated. If the state detection result indicates that a preset key abnormal event has occurred, a corresponding second incremental data to be transmitted is generated. The step of allocating the incremental data to be transmitted to the first priority queue includes: The first incremental data to be transmitted is allocated to the first priority queue; The second incremental data to be transmitted is allocated to a preset third priority queue; wherein the priority of the third priority queue is higher than that of the first priority queue.

5. The SD-WAN managed traffic transmission optimization method of claim 4, wherein, The step of sending the data to be transmitted to the SD-WAN controller based on the multiple transmission queues includes: Based on the preset queue buffering strategy, the data to be transmitted in the first priority queue is sent to the SD-WAN controller; Based on the preset bandwidth shaping strategy, the data to be transmitted in the second priority queue is sent to the SD-WAN controller; The data to be transmitted in the third priority queue is sent to the SD-WAN controller based on the preset instant transmission strategy.

6. The SD-WAN managed traffic transmission optimization method of claim 2, wherein, The process of aggregating the original management traffic data according to a preset data aggregation strategy to generate corresponding summary data to be transmitted includes: Based on preset data classification rules, the log event data in the original management traffic data is filtered and classified, and corresponding summary data to be transmitted is generated based on the filtered and classified data. The time-series data in the original management traffic data is segmented and aggregated based on a preset data aggregation algorithm, and corresponding summary data to be transmitted is generated based on the segmented and aggregated data.

7. The SD-WAN managed traffic transmission optimization method of claim 1, wherein, The step of sending the data to be transmitted to the SD-WAN controller based on the multiple transmission queues includes: The key information of the data to be transmitted is extracted according to a preset data template, and key data to be transmitted corresponding to the data to be transmitted is generated based on the key information; wherein, the data template is issued by the SD-WAN controller to the user terminal device; The critical data to be transmitted is sent to the SD-WAN controller.

8. The SD-WAN managed traffic transmission optimization method of claim 7, wherein, Sending the critical data to be transmitted to the SD-WAN controller includes: The key data to be transmitted is encoded according to a preset encoding rule to obtain encoded data to be transmitted; The encoded data to be transmitted is sent to the SD-WAN controller. 9.The SD-WAN management traffic transmission optimization method of any of claims 1 to 8, wherein, Also includes: In response to a detailed data fetch request sent by the SD-WAN controller; Based on the original management traffic data, filter the target detailed data corresponding to the detailed data retrieval request; The target detailed data is assigned to the highest priority target priority queue, and the target detailed data is sent to the SD-WAN controller based on the target priority queue.

10. An electronic device, comprising: It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, it can implement the SD-WAN management traffic transmission optimization method according to any one of claims 1-9.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the SD-WAN management traffic transmission optimization method as described in any one of claims 1-9.

12. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the SD-WAN management traffic transmission optimization method according to any one of claims 1-9.

Images