SD-WAN-based intelligent traffic optimization method and system

Abstract

The invention discloses an intelligent traffic optimization method and system based on an SD-WAN, and belongs to the field of information transmission, the intelligent traffic optimization method based on the SD-WAN comprises the following steps: identifying a traffic application type, setting a priority strategy, and performing priority grading on traffic; according to different priorities of the traffic, distributing different traffic transmission paths; the method comprises the following steps: setting a bandwidth minimum proportion for the bandwidth of high-priority traffic, and reducing the bandwidth proportion of low-priority traffic and increasing the bandwidth proportion of high-priority traffic in a traffic peak period, and has the beneficial effects that different traffic transmission paths are selected based on different traffic application types, so that the demand of an MPLS traffic transmission path is reduced, and the cost is reduced; the lowest bandwidth and the lowest proportion of an MPLS flow transmission path are set, and normal transmission of high-priority flow is ensured; and the use information of the non-service traffic is detected, so that the employees are prevented from occupying excessive bandwidth due to non-working requirements, and the work is prevented from being influenced.

Description

Technical Field

[0001] The present invention belongs to the field of information transmission, and particularly relates to an intelligent traffic optimization method and system based on SD-WAN. Background Art

[0002] SD-WAN (Software Defined Wide Area Network) is a technology that simplifies the management and operation of wide area networks through software-defined network technology. It uses software to control network traffic, optimize connections, improve performance, and reduce costs.

[0003] Traditional WAN refers to a computer network that covers a wide geographical area and is used to connect different local area networks (LANs) or subnets across cities, countries, and even continents. In this architecture, MPLS dedicated lines are often used as the main transmission means. MPLS dedicated line is a traffic transmission technology based on the MPLS protocol.

[0004] MPLS dedicated lines are dedicated and not shared by other users. Although they ensure the security and reliability of data transmission, they also increase costs and need to be improved. Summary of the Invention

[0005] Based on this, it is necessary to provide an intelligent traffic optimization method and system based on SD-WAN for the above problems.

[0006] An embodiment of the present invention is implemented as follows. An intelligent traffic optimization method based on SD-WAN includes the following steps:

[0007] Identify the traffic application type, set a priority policy, and classify the traffic by priority.

[0008] Allocate different traffic transmission paths according to the different priorities of the traffic.

[0009] Set the minimum bandwidth ratio for high-priority traffic, and reduce the bandwidth ratio of low-priority traffic and increase the bandwidth ratio of high-priority traffic during traffic peaks.

[0010] In one embodiment, the present invention provides an intelligent traffic optimization method based on SD-WAN. In the step of identifying the traffic application type, setting a priority policy, and classifying the traffic by priority, it specifically includes:

[0011] Capture traffic data packets flowing through the network, extract key information by parsing the traffic data packets, and the key information includes protocol type, port number, source address, and destination address.

[0012] According to the extracted key information, using traffic recognition technology, classify the traffic into different application types, and mark different types of traffic. The application types include real-time service traffic, critical service traffic, general service traffic, background traffic, and non-service traffic; the traffic recognition technologies include DPI, DSCP, and 802.1p.

[0013] Set a priority policy, classify the priority of traffic according to the marks on the traffic, classify real-time service traffic as the first priority, critical service traffic as the second priority, general service traffic as the third priority, background traffic as the fourth priority, and non-service traffic as the fifth priority.

[0014] In one embodiment, the present invention provides an intelligent traffic optimization method based on SD-WAN. In the step of allocating different traffic transmission paths according to different priorities of traffic, it specifically includes:

[0015] Allocate traffic with the first priority and the second priority to the MPLS traffic transmission path, and allocate traffic with the third priority, the fourth priority, and the fifth priority to the broadband traffic transmission path;

[0016] Under the same traffic transmission path, determine the forwarding order of different traffic according to scheduling algorithms (such as FIFO, WFQ, SPQ), and traffic with a higher priority classification enters the forwarding queue first.

[0017] In one embodiment, the present invention provides an intelligent traffic optimization method based on SD-WAN. After the step of determining the forwarding order of different traffic according to the scheduling algorithm under the same traffic transmission path, and traffic with a higher priority classification enters the forwarding queue first, it further includes:

[0018] According to the actual situation of the enterprise and the work needs of employees, set the usage threshold of traffic with the fifth priority, regularly conduct statistical analysis on historical traffic data with the fifth priority, and adjust the usage threshold of traffic with the fifth priority;

[0019] Real-time monitor all traffic with the fifth priority. When there is traffic with the fifth priority that exceeds the set usage threshold, determine the abnormal device;

[0020] Feed back the information of the abnormal device to the cloud platform.

[0021] In one embodiment, the present invention provides an intelligent traffic optimization method based on SD-WAN. In the step of setting the minimum bandwidth occupancy ratio for high-priority traffic, reducing the bandwidth occupancy ratio of low-priority traffic and increasing the bandwidth occupancy ratio of high-priority traffic during the traffic peak period, it specifically includes:

[0022] Evaluate the business requirements of the enterprise, determine the minimum bandwidth ratio required for the business corresponding to high-priority traffic (the first priority), log in to the SD-WAN management platform, and set the minimum bandwidth ratio in the SD-WAN policy for high-priority traffic;

[0023] Determine the traffic peak period through the combination of historical data analysis and real-time monitoring;

[0024] During the traffic peak period, reduce the bandwidth ratio of low-priority traffic and increase the bandwidth ratio of high-priority traffic.

[0025] In one embodiment, the present invention provides an intelligent traffic optimization system based on SD-WAN, including:

[0026] A type classification module for identifying the traffic application type, setting a priority policy, and classifying the traffic by priority;

[0027] A path allocation module for allocating different traffic transmission paths according to the different priorities of the traffic;

[0028] A bandwidth correction module for setting the minimum bandwidth ratio for high-priority traffic, reducing the bandwidth ratio of low-priority traffic during the traffic peak period, and increasing the bandwidth ratio of high-priority traffic.

[0029] In one embodiment, the present invention provides an intelligent traffic optimization system based on SD-WAN, and the type classification module includes:

[0030] An information extraction unit for capturing traffic data packets flowing through the network, extracting key information by parsing the traffic data packets, and the key information includes protocol type, port number, source address, and destination address;

[0031] A type identification and marking unit for classifying the traffic into different application types according to the extracted key information by using traffic identification technology and marking different types of traffic. The application types include real-time service traffic, critical service traffic, general service traffic, background traffic, and non-service traffic; the traffic identification technologies include DPI, DSCP, and 802.1p;

[0032] A priority classification unit for setting a priority policy and classifying the traffic by priority according to the marks on the traffic, classifying real-time service traffic as the first priority, critical service traffic as the second priority, general service traffic as the third priority, background traffic as the fourth priority, and non-service traffic as the fifth priority.

[0033] In one embodiment, the present invention provides an intelligent traffic optimization system based on SD-WAN, and the path allocation module includes:

[0034] a path determination unit, configured to allocate traffic of a first priority and a second priority to an MPLS traffic transmission path, and allocate traffic of a third priority, a fourth priority, and a fifth priority to a broadband traffic transmission path;

[0035] The priority forwarding unit is used to determine the forwarding order of different flows according to the scheduling algorithm (such as FIFO, WFQ, SPQ) under the same flow transmission path. The flow with higher priority level enters the forwarding queue first.

[0036] In one embodiment, the present invention provides an intelligent traffic optimization system based on SD-WAN, and the path allocation module further includes:

[0037] A usage threshold setting adjustment unit is used to set a usage threshold of the fifth priority traffic according to the actual situation of the enterprise and the work needs of employees, regularly perform statistical analysis on historical fifth priority traffic data, and adjust the usage threshold of the fifth priority traffic;

[0038] An abnormal device determination unit is used to monitor all traffic of the fifth priority in real time, and to determine an abnormal device when the usage of the fifth priority traffic exceeds a set usage threshold;

[0039] The abnormal feedback unit is used to feed back the information of abnormal devices to the cloud platform.

[0040] In one embodiment, the present invention provides an intelligent traffic optimization system based on SD-WAN, and the bandwidth correction module includes:

[0041] The minimum bandwidth share setting unit is used to evaluate the business needs of the enterprise, determine the minimum bandwidth share required for the high-priority traffic corresponding to the business (first priority), log in to the SD-WAN management platform, and set the minimum bandwidth share in the SD-WAN policy for high-priority traffic;

[0042] The traffic peak period judgment unit is used to determine the traffic peak period through a combination of historical data analysis and real-time monitoring;

[0043] The bandwidth share adjustment unit is used to reduce the bandwidth share of low-priority traffic and increase the bandwidth share of high-priority traffic during traffic peak hours.

[0044] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention selects different traffic transmission paths based on different traffic application types to reduce the demand for MPLS traffic transmission paths and reduce costs; sets the minimum bandwidth and minimum share of the MPLS traffic transmission path to ensure the normal transmission of high-priority traffic; and detects the usage information of non-business traffic (fifth priority) to prevent employees from occupying too much bandwidth due to non-work needs and affecting their work. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a schematic flowchart of an intelligent traffic optimization method based on SD-WAN provided by an embodiment of the present invention.

[0046] Figure 2 This is a schematic flowchart of traffic priority classification provided by an embodiment of the present invention.

[0047] Figure 3 This is a schematic flowchart of allocating traffic transmission paths provided by an embodiment of the present invention.

[0048] Figure 4 This is a schematic flowchart of detecting abnormal usage conditions provided by an embodiment of the present invention.

[0049] Figure 5 This is a schematic flowchart of adjusting bandwidth provided by an embodiment of the present invention.

[0050] Figure 6 This is a schematic diagram of an intelligent traffic optimization system based on SD-WAN provided by an embodiment of the present invention.

[0051] Figure 7 This is a schematic diagram of the type classification module provided by an embodiment of the present invention.

[0052] Figure 8 This is a first partial schematic diagram of the path allocation module provided by an embodiment of the present invention.

[0053] Figure 9 This is a second partial schematic diagram of the path allocation module provided by an embodiment of the present invention.

[0054] Figure 10 This is a schematic diagram of the bandwidth correction module provided by an embodiment of the present invention. Detailed implementation manners

[0055] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0056] It can be understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish one element from another. For example, without departing from the scope of this application, the first xx script may be referred to as the second xx script, and similarly, the second xx script may be referred to as the first xx script.

[0057] In one embodiment, as Figure 1As shown in the figure, an intelligent traffic optimization method based on SD-WAN is applied to an intelligent traffic optimization system based on SD-WAN, and includes the following steps:

[0058] Step S1, identify the traffic application type, set the priority policy, and classify the traffic by priority;

[0059] Step S2, allocate different traffic transmission paths according to the different priorities of the traffic;

[0060] Step S3, set the minimum bandwidth ratio for high-priority traffic. During the traffic peak period, reduce the bandwidth ratio of low-priority traffic and increase the bandwidth ratio of high-priority traffic.

[0061] In daily work, since the traffic application types are different, for some non-essential traffic usage, the MPLS traffic transmission path can be not used to reduce costs. At this time, ordinary broadband can be used. For example, if the traffic application type is identified as real-time service traffic and is regarded as the first priority, it is allocated to the MPLS traffic transmission path to ensure the transmission speed and security; if the traffic application type is identified as background traffic and is regarded as the fourth priority, it is allocated to the broadband traffic transmission path. If the general service traffic is identified as the third priority, it is allocated to the broadband traffic transmission path. When allocating paths, since both background traffic and general service traffic are transmitted through the broadband traffic transmission path, and the general service traffic has a higher priority than the background traffic, the general service traffic will enter the forwarding queue first to ensure the priority of normal work.

[0062] High-priority traffic (such as video conferencing, VOIP, etc.) is sensitive to latency and jitter, and its transmission stability needs to be ensured. By setting the minimum bandwidth ratio (such as 20%), even when the network is congested, basic resources can still be reserved for critical traffic to avoid service interruption. For example, a real-time video call requires continuous bandwidth support. If the minimum ratio is not set, it may be stuck or disconnected due to other traffic occupying the bandwidth. In traditional networks, low-priority traffic (such as file downloads) may preempt the bandwidth, resulting in a decline in the performance of critical services. SD-WAN restricts the excessive occupation of resources by non-critical traffic by forcibly allocating the minimum bandwidth.

[0063] During the traffic peak period, reduce the bandwidth ratio of low-priority traffic and increase the bandwidth ratio of high-priority traffic. SD-WAN dynamically allocates bandwidth based on the real-time network status (such as link quality, latency). After setting the minimum ratio, the remaining bandwidth can be flexibly allocated to other traffic, which not only guarantees critical services but also improves the overall bandwidth utilization rate.

[0064] In one embodiment, such as Figure 2As shown in the figure, an intelligent traffic optimization method based on SD-WAN. In step S1, for the step of identifying traffic application types, setting priority policies, and classifying traffic priorities, it specifically includes:

[0065] Step S11, capture traffic data packets flowing through the network. By parsing the traffic data packets, extract key information, where the key information includes protocol type, port number, source address, and destination address;

[0066] Step S12, according to the extracted key information, use traffic identification technology to classify the traffic into different application types and mark different types of traffic. The application types include real-time service traffic, critical service traffic, general service traffic, background traffic, and non-service traffic; the traffic identification technologies include DPI, DSCP, and 802.1p;

[0067] Step S13, set priority policies, classify the traffic priorities according to the marks on the traffic, classify real-time service traffic as the first priority, critical service traffic as the second priority, general service traffic as the third priority, background traffic as the fourth priority, and non-service traffic as the fifth priority.

[0068] In SA-WAN, classifying different traffic priorities is a complex but crucial process. This process usually needs to consider multiple factors such as traffic type, business requirements, network conditions, and security policies. The following is a general classification framework, for example:

[0069] I. Real-time service traffic (first priority)

[0070] VoIP (Voice over Internet Protocol) / video call: This type of traffic is very sensitive to latency and jitter and requires ensuring high-quality transmission to avoid call interruption or quality degradation.

[0071] Online games: Game traffic also requires low latency and high reliability to ensure real-time interaction between players.

[0072] II. Critical service traffic (second priority)

[0073] Enterprise applications: Such as key business applications like enterprise resources and customer relationships. The smooth operation of these applications is crucial for the daily operation of the enterprise.

[0074] Video conferencing: Although not as real-time as VoIP calls, video conferencing also requires a stable network connection and sufficient bandwidth.

[0075] III. General service traffic (third priority)

[0076] Web browsing: Employees often need to access web pages in their daily work. This type of traffic has low requirements for latency but needs to ensure sufficient bandwidth to provide a smooth browsing experience.

[0077] File transfer: Such as email attachments, internal file sharing, etc. This type of traffic is usually not real-time but needs to be completed within a reasonable time.

[0078] IV. Background traffic (fourth priority)

[0079] System updates: Such as automatic updates of operating systems and application software. This type of traffic can be carried out when the network is idle.

[0080] Other non-critical traffic: Such as advertisements, pop-ups, etc. This type of traffic has little impact on the business and can be placed at a lower priority.

[0081] V. Non-business traffic (fifth priority)

[0082] Social media: Employees may access social media during work. This type of traffic has the least impact on the business and can be given the lowest priority.

[0083] Video streaming: Such as watching videos online. This type of traffic usually consumes a large amount of bandwidth, but if it is not work-related content, it can be placed at the lowest priority.

[0084] In one embodiment, as Figure 3 shown, in the intelligent traffic optimization method based on SD-WAN, in step S2 of allocating different traffic transmission paths according to different priorities of traffic, it specifically includes:

[0085] Step S21: Allocate traffic with the first priority and the second priority to the MPLS traffic transmission path, and allocate traffic with the third priority, the fourth priority, and the fifth priority to the broadband traffic transmission path;

[0086] Step S22: Under the same traffic transmission path, determine the forwarding order of different traffic according to the scheduling algorithm (such as FIFO, WFQ, SPQ), and traffic with a higher priority level enters the forwarding queue first.

[0087] Table 1 - Comparison table of scheduling algorithms:

[0088]

[0089] In actual deployment, SPQ and WFQ are often used in combination (such as in the CBQ model). The high-priority queue uses SPQ, and other queues use WFQ. An example of the SPQ scenario is as follows: Packets in queue 7 (third-priority traffic) on the broadband traffic transmission path are always sent first. Even if there is a large backlog of data in queue 4 (fifth-priority traffic), the forwarding of queue 7 is still guaranteed first.

[0090] In one embodiment, as Figure 4 shown, an intelligent traffic optimization method based on SD-WAN. After step S22, which determines the forwarding order of different traffic according to the scheduling algorithm under the same traffic transmission path and gives priority to traffic with a higher priority level to enter the forwarding queue, it further includes:

[0091] Step S23, set the usage threshold for the fifth-priority traffic according to the actual situation of the enterprise and the work needs of employees, regularly conduct statistical analysis on the historical fifth-priority traffic data, and adjust the usage threshold for the fifth-priority traffic;

[0092] Step S24, monitor all the fifth-priority traffic in real time. When there is fifth-priority traffic usage exceeding the set usage threshold, determine the abnormal device;

[0093] Step S25, feedback the information of the abnormal device to the cloud platform.

[0094] The fifth-priority traffic is not traffic required for work. To ensure the daily use of employees and prevent employees from occupying too much bandwidth (such as downloading large games, etc.), the usage threshold for the fifth-priority traffic is set. When the traffic used by employees exceeds the usage threshold, the abnormal device is determined and feedback to the cloud platform. Relevant personnel can know which employee is working abnormally through the corresponding code of the abnormal device, so as to timely understand the work status of employees and have timely conversations to solve problems.

[0095] In one embodiment, as Figure 5 shown, an intelligent traffic optimization method based on SD-WAN. In step S3, which sets the minimum bandwidth ratio for high-priority traffic and reduces the bandwidth ratio of low-priority traffic and increases the bandwidth ratio of high-priority traffic during the traffic peak period, it specifically includes:

[0096] Step S31, evaluate the business needs of the enterprise, determine the minimum bandwidth ratio (first priority) required for the business corresponding to the high-priority traffic, log in to the SD-WAN management platform, and set the minimum bandwidth ratio in the SD-WAN policy for the high-priority traffic;

[0097] Step S32, determine the traffic peak period through the combination of historical data analysis and real-time monitoring;

[0098] Step S33: During peak traffic periods, reduce the bandwidth ratio of low-priority traffic and increase the bandwidth ratio of high-priority traffic.

[0099] Real-time monitoring data collection: Real-time capture of key metrics such as bandwidth utilization, latency, packet loss rate, and jitter through SD-WAN edge devices, report them to the management platform, and count the inbound / outbound bandwidth and packet volume.

[0100] Obtain historical monitoring data (supporting the past 7 days to a custom period), archive it by time (such as every 10 seconds, every 5 minutes), and form a traffic baseline template. Extract the traffic trends in daily / weekly / monthly dimensions and identify repetitive peak periods (such as 10:00 - 12:00 on weekdays).

[0101] In one embodiment, as Figure 6 shown, an intelligent traffic optimization system based on SD-WAN includes:

[0102] Type classification module 1, used to identify traffic application types, set priority policies, and classify traffic by priority;

[0103] Path allocation module 2, used to allocate different traffic transmission paths according to different priorities of traffic;

[0104] Bandwidth correction module 3, used to set the minimum bandwidth ratio for high-priority traffic, reduce the bandwidth ratio of low-priority traffic during peak traffic periods, and increase the bandwidth ratio of high-priority traffic.

[0105] Select different traffic transmission paths based on different traffic application types to reduce the demand for MPLS traffic transmission paths, which not only ensures the security and timeliness of important content transmission but also reduces costs.

[0106] In one embodiment, as Figure 7 shown, the type classification module 1 of an intelligent traffic optimization system based on SD-WAN includes:

[0107] Information extraction unit 11, used to capture traffic data packets flowing through the network, extract key information by parsing the traffic data packets, and the key information includes protocol type, port number, source address, and destination address;

[0108] Type identification and marking unit 12, used to classify traffic into different application types according to the extracted key information by using traffic identification technology, and mark different types of traffic. The application types include real-time service traffic, critical service traffic, ordinary service traffic, background traffic, and non-service traffic; the traffic identification technologies include DPI, DSCP, and 802.1p;

[0109] The priority classification unit 13 is used to set the priority policy, classify the traffic according to the markings on the traffic, classify the real-time service traffic into the first priority, the critical service traffic into the second priority, the general service traffic into the third priority, the background traffic into the fourth priority, and the non-service traffic into the fifth priority.

[0110] DPI (Deep Packet Inspection), a deep parsing technology based on the content of application-layer data packets, can identify protocol types, user behaviors, and traffic patterns; DSCP (Differentiated Services Code Point), an IP-layer Quality of Service (QoS) marking technology, realizes traffic priority division by modifying the DS field (original TOS field) in the IP packet header; 802.1p (IEEE 802.1Q Priority Marking), a data link layer (layer 2) QoS protocol, realizes traffic classification based on the 3-bit priority field in the TCI field of the VLAN tag.

[0111] Table 2 - Technical Comparison Table:

[0112]

[0113] In actual deployment, the three are often used in combination. For example, after identifying the critical traffic through DPI, the DSCP / 802.1p value is marked, and then the network device performs priority scheduling.

[0114] In one embodiment, as Figure 8 shown, an intelligent traffic optimization system based on SD-WAN, the path allocation module 2 includes:

[0115] The path determination unit 21 is used to allocate the traffic of the first priority and the second priority to the MPLS traffic transmission path, and allocate the traffic of the third priority, the fourth priority, and the fifth priority to the broadband traffic transmission path;

[0116] The priority forwarding unit 22 is used to determine the forwarding order of different traffic according to the scheduling algorithm (such as FIFO, WFQ, SPQ) under the same traffic transmission path, and the traffic with a higher priority classification enters the forwarding queue first.

[0117] The traffic of the first priority and the second priority is regarded as important content, and it is necessary to ensure the transmission speed and transmission security. The MPLS protocol replaces the traditional IP hop-by-hop routing through pre-allocated labels (LSP), reduces the packet processing delay, and has high forwarding efficiency; through the predefined path (LSP) and fast rerouting (FRR) technologies, it automatically switches to the backup path in case of network failure, ensuring service continuity, and has high reliability and security. The traffic of the third priority, the fourth priority, and the fifth priority is regarded as relatively unimportant content, and the commonly used broadband is selected to reduce costs.

[0118] In one embodiment, as Figure 9 shown, an intelligent traffic optimization system based on SD-WAN, the path allocation module 2 further includes:

[0119] A usage threshold setting and adjustment unit 23, configured to set the usage threshold of the traffic of the fifth priority according to the actual situation of the enterprise and the work requirements of employees, regularly perform statistical analysis on the historical traffic data of the fifth priority, and adjust the usage threshold of the traffic of the fifth priority;

[0120] An abnormal device determination unit 24, configured to monitor all the traffic of the fifth priority in real time, and determine an abnormal device when there is traffic of the fifth priority exceeding the set usage threshold;

[0121] An abnormal feedback unit 25, configured to feedback the information of the abnormal device to the cloud platform.

[0122] In actual use, it is also possible to detect the usage status of the traffic of other priorities, and when a large amount of traffic is used on a certain device, an abnormality can be detected in time.

[0123] In one embodiment, as Figure 10 shown, an intelligent traffic optimization system based on SD-WAN, the bandwidth correction module 3 includes:

[0124] A minimum bandwidth occupancy ratio setting unit 31, configured to evaluate the business requirements of the enterprise, determine the minimum bandwidth occupancy ratio (first priority) required for the services corresponding to the high-priority traffic, log in to the SD-WAN management platform, and set the minimum bandwidth occupancy ratio in the SD-WAN policy for the high-priority traffic;

[0125] A traffic peak period determination unit 32, configured to determine the traffic peak period through the combination of historical data analysis and real-time monitoring;

[0126] A bandwidth occupancy ratio adjustment unit 33, configured to reduce the bandwidth occupancy ratio of the low-priority traffic and increase the bandwidth occupancy ratio of the high-priority traffic during the traffic peak period.

[0127] Obtain historical traffic data, select a model, such as LSTM / GRU, Prophet, XGBoost / LightGBM, etc., divide the data set, and divide it into a training set, a validation set, and a test set according to 7:2:1. Hyperparameter optimization, use Bayesian optimization or grid search to adjust the model parameters. Transfer learning, aiming at the differences in traffic patterns of different branches, reuse the pre-trained model and fine-tune the parameters of the last layer. Deploy the model to the SD-WAN controller, output the predicted bandwidth requirements for the next 1 hour at a minute-level frequency, compare the prediction results with the real-time monitoring data, and trigger the retraining of the model. Train a prediction model based on historical traffic data, dynamically correct it in combination with real-time data, and predict the future traffic peak time and bandwidth requirements.

[0128] It should be understood that although the steps in the flowcharts of the embodiments of the present invention are shown in sequence according to the indications of the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in each embodiment may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or sub-steps or stages of other steps.

[0129] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0130] The technical features of the above-described embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the various technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0131] The embodiments described above merely represent several implementation manners of the present invention. Their descriptions are relatively specific and detailed, but should not be construed as limiting the scope of the patent for the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all fall within the protection scope of the present invention. Therefore, the protection scope of the patent for the present invention shall be subject to the appended claims.

[0132] The above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

[0133] In addition, it should be understood that although this specification is described in terms of embodiments, not every embodiment only includes an independent technical solution. This narrative manner of the specification is only for clarity. Those skilled in the art should regard the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. An intelligent traffic optimization method based on SD-WAN, characterized in that, The intelligent traffic optimization method based on SD-WAN includes the following steps: Identify the traffic application types, set the priority policy, and classify the traffic by priority; Allocate different traffic transmission paths according to the different priorities of the traffic; Set the minimum bandwidth ratio for high-priority traffic. During the traffic peak period, reduce the bandwidth ratio of low-priority traffic and increase the bandwidth ratio of high-priority traffic.

2. The intelligent traffic optimization method based on SD-WAN according to claim 1, wherein In the step of identifying the traffic application types, setting the priority policy, and classifying the traffic by priority, it specifically includes: Capture the traffic data packets flowing through the network. By parsing the traffic data packets, extract the key information, which includes protocol type, port number, source address, and destination address; According to the extracted key information, use traffic identification technology to classify the traffic into different application types and mark different types of traffic. The application types include real-time service traffic, critical service traffic, ordinary service traffic, background traffic, and non-service traffic; The traffic identification technology includes DPI, DSCP, and 802.1p; Set the priority policy, classify the traffic by priority according to the marks on the traffic, classify the real-time service traffic as the first priority, the critical service traffic as the second priority, the ordinary service traffic as the third priority, the background traffic as the fourth priority, and the non-service traffic as the fifth priority.

3. The intelligent traffic optimization method based on SD-WAN according to claim 1, wherein In the step of allocating different traffic transmission paths according to the different priorities of the traffic, it specifically includes: Allocate the traffic of the first priority and the second priority to the MPLS traffic transmission path, and allocate the traffic of the third priority, the fourth priority, and the fifth priority to the broadband traffic transmission path; Under the same traffic transmission path, determine the forwarding order of different traffic according to the scheduling algorithm, and the traffic with a higher priority classification enters the forwarding queue first.

4. The intelligent traffic optimization method based on SD-WAN according to claim 3, characterized in that After the step of determining the forwarding order of different traffic according to the scheduling algorithm under the same traffic transmission path and the traffic with a higher priority classification enters the forwarding queue first, it further includes: According to the actual situation of the enterprise and the work requirements of employees, set the usage threshold for the traffic of the fifth priority, regularly conduct statistical analysis on the historical traffic data of the fifth priority, and adjust the usage threshold for the traffic of the fifth priority; Real-time monitor all the traffic of the fifth priority. When there is traffic of the fifth priority that exceeds the set usage threshold, determine the abnormal device; Feed back the information of the abnormal device to the cloud platform.

5. The intelligent traffic optimization method based on SD-WAN according to any one of claims 1 to 4, characterized in that, In the step of setting the minimum bandwidth ratio for high-priority traffic, reducing the bandwidth ratio of low-priority traffic and increasing the bandwidth ratio of high-priority traffic during the traffic peak period, it specifically includes: Evaluate the business requirements of the enterprise, determine the minimum bandwidth ratio required for the business corresponding to the high-priority traffic, log in to the SD-WAN management platform, and set the minimum bandwidth ratio in the SD-WAN policy for the high-priority traffic; Determine the traffic peak period through the combination of historical data analysis and real-time monitoring; During the traffic peak period, reduce the bandwidth ratio of low-priority traffic and increase the bandwidth ratio of high-priority traffic.

6. An intelligent traffic optimization system based on SD-WAN, characterized in that, The intelligent traffic optimization system based on SD-WAN includes: A type classification module, which is used to identify the types of traffic applications, set priority policies, and classify the traffic by priority; A path allocation module, which is used to allocate different traffic transmission paths according to the different priorities of the traffic; A bandwidth correction module, which is used to set the minimum bandwidth ratio for high-priority traffic, and during the traffic peak period, reduce the bandwidth ratio of low-priority traffic and increase the bandwidth ratio of high-priority traffic.

7. The intelligent traffic optimization system based on SD-WAN according to claim 6, wherein The type classification module includes: An information extraction unit, which is used to capture traffic data packets flowing through the network, extract key information by parsing the traffic data packets, and the key information includes protocol type, port number, source address, and destination address; A type identification and marking unit, which is used to classify the traffic into different application types according to the extracted key information by using traffic identification technology, and mark different types of traffic. The application types include real-time service traffic, critical service traffic, general service traffic, background traffic, and non-service traffic; the traffic identification technologies include DPI, DSCP, and 802.1p; A priority classification unit, which is used to set priority policies, classify the traffic by priority according to the marks on the traffic, classify real-time service traffic as the first priority, critical service traffic as the second priority, general service traffic as the third priority, background traffic as the fourth priority, and non-service traffic as the fifth priority.

8. The intelligent traffic optimization system based on SD-WAN according to claim 6, characterized in that, The path allocation module includes: A path determination unit, which is used to allocate the traffic with the first priority and the second priority to the MPLS traffic transmission path, and allocate the traffic with the third priority, the fourth priority, and the fifth priority to the broadband traffic transmission path; A priority forwarding unit, which is used to determine the forwarding order of different traffic according to the scheduling algorithm under the same traffic transmission path, and the traffic with a higher priority classification enters the forwarding queue first.

9. The intelligent traffic optimization system based on SD-WAN according to claim 8, characterized in that, The path allocation module also includes: A usage threshold setting and adjustment unit, which is used to set the usage threshold of the traffic with the fifth priority according to the actual situation of the enterprise and the work needs of employees, regularly statistically analyze the historical traffic data of the fifth priority, and adjust the usage threshold of the traffic with the fifth priority; An abnormal device determination unit, which is used to monitor all the traffic with the fifth priority in real time, and determine the abnormal device when there is traffic with the fifth priority exceeding the set usage threshold; An abnormal feedback unit, which is used to feedback the information of the abnormal device to the cloud platform.

10. The intelligent traffic optimization system based on SD-WAN according to any one of claims 6 to 9, characterized in that The bandwidth correction module includes: A minimum bandwidth ratio setting unit, which is used to evaluate the business needs of the enterprise, determine the minimum bandwidth ratio required for the services corresponding to high-priority traffic, log in to the SD-WAN management platform, and set the minimum bandwidth ratio in the SD-WAN policy for high-priority traffic; A traffic peak period determination unit, which is used to determine the traffic peak period by combining historical data analysis and real-time monitoring; A bandwidth ratio adjustment unit, which is used to reduce the bandwidth ratio of low-priority traffic and increase the bandwidth ratio of high-priority traffic during the traffic peak period.

Images