Implementation method, device and equipment for bearing two-layer giant frame through Internet based on intelligent routing technology

Abstract

The invention relates to an implementation method, device and equipment for bearing a two-layer giant frame through the Internet based on an intelligent routing technology, and the method comprises the steps: obtaining all link quality data and generating a quality weight calculation result based on a user data packet protocol according to a detection packet transmitted between a transmitting end and a receiving end, and obtaining a quality weight calculation result according to a cutting reference value of a link; the method comprises the following steps of: cutting and blocking an input two-layer network giant frame to obtain the quality of the giant frame, selecting an optimal link, generating an error correction data packet for transmission, checking according to the received error correction data packet, adding the error correction data packet into a splicing list, checking the integrity of the data packet in the list, and carrying out assembly or validity judgment. According to the invention, through intelligent routing, a larger huge frame can use a link with higher quality, and idling or overload caused by a fixed path is avoided while the transmission reliability of the huge frame is improved. And by checking the error correction information of the data packets, only the data packets which fail to be checked are retransmitted, so that the reduction of invalid transportation is realized.

Description

Technical Field

[0001] This application relates to the technical field of SD-WAN, and particularly to an implementation method, device, and equipment for carrying Layer 2 jumbo frames over the Internet based on intelligent routing technology. Background Art

[0002] In recent years, as an extension of SD-WAN technology, the solution of carrying Layer 2 networks based on Layer 3 tunnels has gradually become popular. Its core principle is to encapsulate Layer 2 network frames (such as Ethernet frames), carry them through TCP or UDP protocols, and finally transmit them to remote devices via the IP network (especially the Internet). From the perspective of the protocol stack structure, the encapsulated "Layer 2 link data packet" is embedded inside the Layer 3 tunnel, and this tunnel itself can also be used to transmit other data (such as ordinary IP traffic), and the entire tunnel relies on the Internet for transmission. This technology enables traditional Layer 2 communications (such as ARP, MAC address learning) that are limited to local area networks to span wide area networks and is applicable to scenarios such as enterprise remote networking and cloud migration.

[0003] However, when carrying Layer 2 networks over the Internet, there are two key problems: MTU limitation and transmission quality adaptation. First, the standard MTU of the Internet usually does not exceed 1500 bytes, while Layer 2 networks (especially data centers or storage networks) often use jumbo frames, and the typical length can reach 9000 bytes, far exceeding the carrying capacity of the Internet. When the length of the L2 data packet exceeds the effective MTU of the tunnel, the original frame cannot be directly transmitted and must be fragmented or discarded, resulting in a decrease in efficiency or even communication failure. Second, different Layer 2 data frames have different requirements for network quality. For example, jumbo frame transmission requires higher bandwidth and more stable connections because once a packet is lost, the cost of retransmitting the entire jumbo frame is extremely high; while ordinary small frames (such as VoIP data) are more sensitive to latency. Traditional solutions lack dynamic adaptation capabilities and cannot intelligently adjust transmission strategies for different services.

[0004] To address the above problems, an overall solution is urgently needed, and its core should include: a dynamic MTU negotiation mechanism (such as path MTU discovery + fragmentation optimization), intelligent routing technology (selecting the optimal link according to frame size and service requirements), and error correction and retransmission optimization (such as block-level verification + selective retransmission). For example, in the SD-WAN architecture, edge devices can detect the MTU and quality (packet loss rate, latency) of each link in real time, reasonably divide jumbo frames into blocks, and transmit them through high-stability links (such as dedicated lines), while small frames are flexibly scheduled to low-latency paths (such as 5G). This solution can not only solve the MTU mismatch problem but also improve the transmission efficiency of Layer 2 traffic in complex network environments and provide reliable support for modern distributed services. Summary of the Invention

[0005] The present application provides an implementation method for carrying layer-2 jumbo frames over the Internet based on intelligent routing technology, which is characterized by including: Based on the User Datagram Protocol, all link quality data is obtained according to the detection packets transmitted between the sender and the receiver, and a quality weight calculation result is generated; According to the cutting reference value of the link, the input layer-2 network jumbo frame is cut and divided into blocks to obtain the jumbo frame quality, the optimal link is selected, and an error correction data packet is generated for transmission; According to the received error correction data packet, verification is performed and it is added to the splicing list, the integrity of the data packets in the list is checked, and assembly or validity judgment is carried out.

[0006] Optionally, the step of obtaining all link quality data according to the detection packets transmitted between the sender and the receiver based on the User Datagram Protocol and generating a quality weight calculation result includes: The detection packet is a lightweight data packet containing a timestamp and a sequence number, which is sent at a fixed frequency across all links for measuring the packet loss rate and delay data of each link; The generation of the quality weight calculation result is the overall link quality score generated by weighted calculation based on the packet loss rate and delay data of each link, which is sorted and a link quality label is added.

[0007] Optionally, the step of cutting and dividing the input layer-2 network jumbo frame according to the cutting reference value of the link to obtain the jumbo frame quality, selecting the optimal link, and generating an error correction data packet for transmission includes: The cutting reference value of the link is the minimum value among the maximum transmission unit values of all links; The step of cutting and dividing the input layer-2 network jumbo frame to obtain the jumbo frame quality is to cut the jumbo frame to obtain the number of blocks by deducting the reserved bytes of the data packet with the cutting reference value of the link, and judge the quality of the current jumbo frame and add a jumbo frame quality label by setting a jumbo frame block number threshold; The selection of the optimal link is to find the corresponding link quality label according to the jumbo frame quality label; The error correction data packet is the secondary network jumbo frame split data packet generated after cutting, with a hash check value and a sequence number added additionally for detecting transmission errors and identifying the block order.

[0008] Optionally, the step of performing verification on the received error correction data packet, adding it to the splicing list, checking the integrity of the data packets in the list, and performing assembly or validity judgment includes: Perform verification and add to the splicing list. For the receiving end, obtain the sequence number and hash verification value in the error correction data packet for judgment. If the sequence number is the expected or missing sequence number and the hash verification value is verified correctly, the data packet is added to the splicing list; otherwise, this data packet is discarded and a retransmission is requested. Check the integrity of the data packets in the list. After the splicing list receives a data packet, judge the reception status of the data packet. If all are collected, start assembling the data packets; otherwise, judge the validity of this round of assembly.

[0009] Optionally, for the selection of the optimal link, by setting the threshold of the number of jumbo frame blocks, judge the link that the transmitted jumbo frame should use, including: Set the threshold of the number of jumbo frame blocks to n1 and n2, and n1 < n2. Set the number of cut blocks of the layer 2 network jumbo frame to N. When N ≤ n1, add a low-quality jumbo frame label to this jumbo frame, and select the link with the worst link quality label for data packet transmission. When n1 < N ≤ n2, add a medium-quality jumbo frame label to this jumbo frame, and select the link with the medium link quality label for data packet transmission. When N > n2, add a high-quality jumbo frame label to this jumbo frame, and select the link with the optimal link quality label for data packet transmission.

[0010] Optionally, for checking the integrity of the data packets in the list, after the splicing list receives a data packet, judge the reception status of the data packet. If all are collected, start assembling the data packets; otherwise, judge the validity of this round of assembly, including: Judging the validity of this round of assembly is to judge the sequence gap between the missing data packet and the latest received data packet. When the gap is less than the set threshold, this round of assembly is valid, and wait for the missing data packet to be resent. When the gap is greater than the set threshold, this round of assembly is invalid, clear the received data packets and start over.

[0011] Optionally, the above-mentioned implementation method of carrying layer 2 jumbo frames over the Internet based on intelligent routing technology is characterized in that when there is only one physical link between the sending end and the receiving end, it further includes: Create virtual links with different forward error correction coding strengths in the physical link, and distinguish the quality of the virtual links through detection packets to perform jumbo frame transmission.

[0012] This application also provides an implementation device for carrying layer 2 jumbo frames over the Internet based on intelligent routing technology, which is characterized in that the device includes: A link test module, configured to obtain all link quality data according to the detection packets transmitted between the sending end and the receiving end, and generate a quality weight calculation result. An intelligent routing selection module, which is used to cut and divide the input layer-2 network jumbo frame according to the cut reference value of the link to obtain the jumbo frame quality, select the optimal link, and generate an error correction data packet for transmission; A receiving and assembling module, which is used to perform verification on the received error correction data packet, add it to the splicing list, check the integrity of the data packets in the list, and perform assembly or validity judgment.

[0013] Optionally, the intelligent routing selection module includes: A link selection module, which is used to sort the quality of all links and select the link with the corresponding quality according to the size of the cut of the layer-2 network jumbo frame; A jumbo frame cutting module, which is used to cut the layer-2 network jumbo frame and generate a data packet containing error correction information.

[0014] The present application also provides an electronic device, which is characterized in that it is used to implement any of the above-mentioned implementation methods for carrying layer-2 jumbo frames over the Internet based on intelligent routing selection technology, including: A first SD-WAN edge device, which is deployed at the sending end and is used to send link detection packets, perform dynamic maximum transmission unit value cutting, encapsulate data packets, and perform intelligent routing selection; A second SD-WAN edge device, which is deployed at the receiving end and is used for link quality feedback, verifying the received data packets, and performing data packet error correction and recombination.

[0015] The beneficial effects of the present application are as follows: Through intelligent routing selection, larger jumbo frames use higher-quality links, which improves the reliability of jumbo frame transmission while avoiding idleness or overload caused by fixed paths. And by verifying the error correction information of the data packets and only retransmitting the data packets that fail the verification, the reduction of ineffective transportation is achieved. Description of the Drawings

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following briefly introduces the drawings required in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0017] Figure 1 Shows a flowchart of an implementation method for carrying layer-2 jumbo frames over the Internet based on intelligent routing selection technology disclosed in the present application; Figure 2 Shows a schematic diagram of the protocol stack structure of an implementation method for carrying layer-2 jumbo frames over the Internet based on intelligent routing selection technology disclosed in the present application; Figure 3Schematic diagram of a layer-2 network jumbo frame showing an implementation method for carrying layer-2 jumbo frames over the Internet based on intelligent routing technology; Figure 4 Schematic diagram of the maximum transmission unit value cutting for an implementation method for carrying layer-2 jumbo frames over the Internet based on intelligent routing technology; Figure 5 Schematic diagram of the layer-2 link data packet cutting mechanism for an implementation method for carrying layer-2 jumbo frames over the Internet based on intelligent routing technology; Figure 6 Schematic diagram of the logic for the receiving end to receive the cut L2 data packet sequence and assemble it for an implementation method for carrying layer-2 jumbo frames over the Internet based on intelligent routing technology Figure 7 Schematic diagram of the link for an implementation method for carrying layer-2 jumbo frames over the Internet based on intelligent routing technology; Figure 8 Schematic diagram of the device for an implementation method for carrying layer-2 jumbo frames over the Internet based on intelligent routing technology. Detailed implementation manners

[0018] Various exemplary embodiments, features, and aspects of the present application will be described in detail below with reference to the accompanying drawings. Identical reference numerals in the drawings denote elements having the same or similar functions. Although various aspects of the embodiments are shown in the drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0019] Among them, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present application, "a plurality of" means two or more unless otherwise specifically defined.

[0020] The special term "exemplary" here means "serving as an example, embodiment, or illustration". Any embodiment described as "exemplary" here does not have to be construed as superior to or better than other embodiments.

[0021] In addition, for better illustration of the present application, numerous specific details are given in the following detailed implementation manners. Those skilled in the art should understand that the present application can also be implemented without some specific details. In some instances, methods, means, elements, and circuits well known to those skilled in the art are not described in detail so as to highlight the gist of the present application.

[0022] This application is an implementation method for carrying layer-2 jumbo frames over the Internet based on intelligent routing technology. In this application, through dynamic link quality assessment, the packet loss rate and latency of each path are detected in real time, the links are sorted by quality and quality tags are added; based on the minimum maximum transmission unit value (MTU) in the entire link, the jumbo frame is segmented, and according to the set threshold of the number of jumbo frame blocks, the quality level of the current jumbo frame is obtained and a quality tag is added, and the optimal link is selected for transmission by matching the quality tags of the link and the jumbo frame; finally, through error correction verification and ordered recombination, the hash value and sequence number are used to ensure data integrity and order.

[0023] Embodiment 1 As Figure 1 shown, it is a flowchart of an implementation method for carrying layer-2 jumbo frames over the Internet based on intelligent routing technology according to an embodiment of the present application, which specifically includes the following contents: S100, based on the User Datagram Protocol, according to the detection packets transmitted between the sender and the receiver, obtain all link quality data and generate a quality weight calculation result.

[0024] Specifically, in this embodiment, the sender sends detection packets to all links at a fixed frequency, the receiver receives the detection packets, and according to the number of lost packets and the timestamp of the detection packets, obtains the packet loss rate and latency of each link. Then, weighted calculation is performed through preset weights to generate the quality scores of each link, and the links are sorted according to the scores to distinguish links of different qualities, and corresponding link quality tags are added to the links of different qualities.

[0025] S200, according to the cutting reference value of the link, segment and block the input layer-2 network jumbo frame to obtain the jumbo frame quality, select the optimal link, and generate error correction data packets for transmission.

[0026] Specifically, in this embodiment, a threshold of the number of jumbo frame blocks regarding the size of the layer-2 network jumbo frame is set. Through this threshold, it is judged what quality of link should be used for the jumbo frame to be transmitted, and then the MTU of the used link is obtained in real time for jumbo frame segmentation, and data packets with an error correction mechanism are generated for sending.

[0027] S300, according to the received error correction data packets, perform verification and add them to the splicing list, check the integrity of the data packets in the list, and perform assembly or validity judgment.

[0028] Specifically, in this embodiment, after receiving the data packets, the receiver judges whether the data packet is the currently expected or missing data packet through the sequence number, and judges the data integrity of the data packet through the hash check value. Then, the data packet is added to the splicing list, and the collection status of the data packets in the list is judged. If all are collected, assembly is immediately performed; if not all are collected, the validity judgment of this round of assembly is performed.

[0029] In the current technical background, as an extension of SD-WAN technology, the 2-layer network carried by a 3-layer tunnel has been recently popularized. Its core principle is to encapsulate the 2-layer network frames and carry them in the form of TCP or UDP, and transmit them to remote devices through an IP network (especially the Internet). The protocol stack structure is as Figure 2 shown. The "2-layer link data packet" is the encapsulated 2-layer network frame, carried on (inside) the tunnel. This tunnel can also be used to carry "other data", and the tunnel is carried on the Internet. Currently, there are the following two problems in carrying the 2-layer network based on the Internet: First, as Figure 3 shown, for the Internet, the general MTU does not exceed 1500 bytes, while there are relatively large frames in the 2-layer network, especially jumbo frames. The typical packet length is about 9000 bytes, which is much larger than the MTU that the Internet can carry. When the length of the L2 link data packet far exceeds the packet length of the Internet-based tunnel, small packets cannot accommodate large packets. Second, for different 2-layer network data frames, different transmission qualities are required to meet different needs. Especially for jumbo frames, a larger bandwidth and more stable network quality are needed (the cost of retransmission is too high), while the traditional solution does not have this technology.

[0030] In summary, the present application significantly optimizes the transmission efficiency and reliability of 2-layer jumbo frames based on the Internet through intelligent routing and dynamic MTU cutting technology. First, use UDP detection packets to evaluate the link quality in real time and add tags to dynamically distinguish the link quality, ensuring that the routing is accurately adapted to the network state. Second, use the jumbo frame block number threshold as the judgment benchmark, add quality tags according to the jumbo frame block number, and perform intelligent routing through the label matching of the jumbo frame and the link. Finally, achieve block-level error correction through sequence numbers and hash checks. The receiving end can quickly identify out-of-order or damaged data packets and only need to retransmit invalid blocks, reducing the redundant overhead.

[0031] As an optional implementation of the present application, optionally, in step S100, based on the User Datagram Protocol, all link quality data is obtained according to the detection packets transmitted between the sending end and the receiving end, and a quality weight calculation result is generated, including: S101, the detection packet is a lightweight data packet containing a timestamp and a sequence number, and is sent at a fixed frequency in all links for measuring the packet loss rate and delay data of each link.

[0032] Specifically, since the present application is designed based on the User Datagram Protocol (UDP), the detection packet used is a lightweight UDP packet. As Figure 7 shown, the detection packet is configured to be transmitted from the sending end to the receiving end, and a 64-byte detection packet is sent at a frequency of once every 1 second to test the link quality.

[0033] Among them, the test link quality is to obtain the delay and packet loss rate of the link used by the detection packet by detecting the timestamp and sequence number in the packet respectively, and the jitter (Jitter) parameter can also be obtained through the round-trip time (RTT), so as to obtain the link quality index.

[0034] S102. The generation of the quality weight calculation result is to sort all the link quality scores generated by weighted calculation according to the packet loss rate and delay data of each link, and add link quality labels.

[0035] Specifically, the link quality score is obtained by multiplying the quality index data (packet loss rate and delay) by the corresponding weight, and the weight can be set to different values according to different usage scenarios. In this embodiment, in the scenario of jumbo frame transmission in the layer 2 network, the packet loss rate weight is set to 80%, the delay weight is set to 20%, and the weighted calculation formula for the link quality score is: 。

[0036] Among them, the sorting of all link scores is to arrange the links of each quality from low to high according to the scores.

[0037] For example, the packet loss rate of link A is 0.5% and the delay is 40 ms; the packet loss rate of link B is 2% and the delay is 30 ms; the packet loss rate of link C is 1% and the delay is 60 ms. The quality scores of the three links are: ; ; 。

[0038] Since the lower the score, the higher the link quality, the sorting result is link A > link C > link B. The link quality label of link A is the optimal quality link, the link quality label of link C is the intermediate quality link, and the link quality label of link B is the worst quality link.

[0039] In addition, the essence of a link is a virtual communication channel established through tunneling technology. Each tunnel consists of logical interfaces at both ends, and these interfaces are all assigned independent IP addresses for data routing. The system continuously monitors the transmission quality parameters of each link (including packet loss rate, delay, jitter, etc.), and uses an intelligent algorithm to dynamically assign quality level labels to each logical interface. For example, label 1 represents the optimal quality link, label 2 represents the intermediate quality link, and so on. During data transmission, the sending end will automatically match the corresponding quality level label according to the number of blocks of the layer 2 jumbo frame, so as to select the most suitable transmission channel.

[0040] As an alternative implementation of the present application, optionally, in step S200, according to the cutting reference value of the link, the input layer-2 network jumbo frame is cut and divided into blocks to obtain the jumbo frame quality, and the optimal link is selected to generate an error correction data packet for transmission, including: S201, the cutting reference value of the link is the minimum value among the maximum transmission unit values of all links.

[0041] Specifically, as Figure 3 shown, the MTU of the layer-2 network jumbo frame is much larger than the MTU of the tunnel packet based on the Internet. Therefore, it is necessary to cut it and generate split data packets for Figure 4 transmission in the form shown, and then assembled by the receiving end. The cutting reference value of the link in the present application is to select the minimum value of the MTUs of all links between the sending end and the receiving end and subtract the reserved bytes to form the cutting reference value.

[0042] S202, the cutting and dividing the input layer-2 network jumbo frame to obtain the jumbo frame quality is to deduct the reserved bytes of the data packet through the cutting reference value of the link, cut the jumbo frame to obtain the number of blocks, and judge the quality of the current jumbo frame and add a jumbo frame quality label by setting a jumbo frame block number threshold.

[0043] Specifically, detect the MTUs of all links and select the minimum value as the cutting reference value, and perform a cutting process on the jumbo frame according to this MTU value. When cutting, the reserved bytes should be reserved before cutting. The specific form is: .

[0044] The cutting process is as Figure 5 shown. First, receive the complete layer-2 network jumbo frame, then obtain the cutting reference value, and judge whether the MTU value can accommodate the current remaining jumbo frame data. If it can, encapsulate it with UDP and send it through the tunnel. If not, first cut part of the jumbo frame according to the MTU value, encapsulate the cut part with UDP and send it through the tunnel. Then judge whether there is still jumbo frame data remaining after sending. If not, this round of packaging ends and waits for new layer-2 network data. If there is still data remaining, the link should be re-judged whether the MTU value can accommodate the current remaining jumbo frame data until there is no data remaining and the loop ends.

[0045] Among them, the jumbo frame block number thresholds are set to n1 and n2, and n1 < n2. If the number of cut blocks of the layer-2 network jumbo frame is less than n1, the jumbo frame quality label of this jumbo frame is a low-quality jumbo frame. If the number of cut blocks of the layer-2 network jumbo frame is greater than n1 and less than n2, the jumbo frame quality label of this jumbo frame is a medium-quality jumbo frame. If the number of cut blocks of the layer-2 network jumbo frame is greater than n2, the jumbo frame quality label of this jumbo frame is a high-quality jumbo frame.

[0046] S203, the step of selecting the optimal link is to find the corresponding link quality label according to the jumbo frame quality label.

[0047] Specifically, the jumbo frame and the link are made corresponding through labels. High-quality jumbo frames correspond to using the optimal quality link, medium-quality jumbo frames correspond to using the medium-quality link, and low-quality jumbo frames correspond to using the worst-quality link.

[0048] For example, let the jumbo frame block number thresholds n1 = 2 and n2 = 7, and the number of cut blocks of the layer-2 network jumbo frame be set as N; When N ≤ 2, add a low-quality jumbo frame label to this jumbo frame, and select the link with the worst quality link quality label for data packet transmission; When 2 < N ≤ 7, add a medium-quality jumbo frame label to this jumbo frame, and select the link with the medium-quality link quality label for data packet transmission; When N > 7, add a high-quality jumbo frame label to this jumbo frame, and select the link with the optimal quality link quality label for data packet transmission.

[0049] S204, the error-corrected data packet is to add a hash check value and a sequence number to the split data packets of the secondary network jumbo frame generated after cutting, which are used to detect transmission errors and identify the block order.

[0050] Specifically, the hash check value is used by the receiving end to judge whether the data block has been tampered with or damaged during transmission through the check value. Its check process is that the sending end calculates the hash for each data packet during packaging and attaches it to the data packet. Then, when the receiving end receives the data packet, it recalculates the hash and checks it against the hash value in the packet. If they match, it means the data is valid and can be processed continuously. If they don't match, it means the data is invalid, and the receiving end transmits a retransmission request for this data packet to the sending end.

[0051] In addition, the sequence number is allocated to the split data packets of the layer-2 network jumbo frame in an increasing manner, so that the receiving end can rearrange the out-of-order data packets according to the sequence number, and can judge whether there is a missing data packet during transmission according to the continuity of the sequence number. If a missing packet occurs, the receiving end transmits a retransmission request for this data packet to the sending end.

[0052] As an optional implementation of this application, optionally, in step S300, according to the received error-corrected data packet, perform verification and add it to the splicing list, check the integrity of the data packets in the list, and perform assembly or validity judgment, including: Specifically, the splicing and assembly process is as Figure 6 shown: After receiving a data packet at the receiving end, obtain the sequence number and hash check value in the data packet. Confirm the data validity of the data packet by matching the hash value in the data packet with the hash value calculated by the receiving end. Determine whether the data packet is the one required by the current splicing list or the missing data packet based on the sequence number. If any of these two checks fails, discard this data packet and request retransmission. If all checks pass, add the data packet to the splicing list.

[0053] Then the receiving end determines whether all the data packets have been collected in the splicing list. If all have been collected, immediately perform the data packet assembly work. If not, determine the situation of the missing data packets.

[0054] The determination of the situation of the missing data packets is that the receiving end determines whether the gap between the sequence number of the latest received data packet and the sequence number of the missing data packet is greater than the set threshold. In this embodiment, the threshold is set to 5. If the gap is less than or equal to the threshold, it is determined that the current round of assembly is still valid, and the receiving end continues to wait for the transmission of data packets. If the gap is greater than the threshold, it is determined that the current round of assembly is invalid, the receiving end clears all the data packets in the splicing list, and transmits a request to start over to the sending end.

[0055] As an alternative implementation of this application, optionally, when there is only one physical link between the sending end and the receiving end, it further includes: Specifically, in the only physical link, several forward error correction coding (FEC) tunnels with different redundancy levels are pre-created between the sending end and the receiving end. Send the detection packets in S100 to these FEC tunnels to obtain and arrange the quality of these FEC tunnels. Then, the layer 2 network jumbo frame can be transmitted according to the content described in steps S200 and S300.

[0056] Among them, the FEC technology can achieve error detection and correction functions during network transmission by adding redundant check information to data packets. Its core value lies in that when the link transmission quality deteriorates, the data packet loss rate can be significantly reduced and the transmission reliability can be improved by enhancing the error correction strength of FEC. However, the greater the error correction strength, the more redundant data is required, which will lead to a decrease in the utilization rate of the effective transmission bandwidth. At the same time, due to the increase in the encoding and decoding processing time, the transmission delay will also increase. Therefore, in practical applications, it is necessary to dynamically balance between the error correction ability and the transmission efficiency according to the link quality status to achieve the optimal transmission effect.

[0057] Embodiment 2 Based on the same principle as the foregoing method, an implementation device for carrying layer 2 jumbo frames over the Internet based on intelligent routing technology is also proposed. See Figure 8, an implementation device 100 for carrying a layer-2 jumbo frame over the Internet based on intelligent routing technology according to an embodiment of the present disclosure includes: 110, a link test module, configured to obtain all link quality data according to detection packets transmitted between a sending end and a receiving end, and generate a quality weight calculation result; 120, an intelligent routing module, configured to cut and divide an input layer-2 network jumbo frame according to a cut reference value of a link to obtain jumbo frame quality, select an optimal link, and generate and transmit an error correction data packet; 130, a receiving and assembling module, configured to perform verification on the received error correction data packet, add it to a splicing list, check the integrity of data packets in the list, and perform assembly or validity judgment.

[0058] As an alternative implementation of the present application, optionally, the intelligent routing module 120 includes: 121, a link selection module, configured to sort the quality of all links and select a link with corresponding quality according to the size of the cut of the layer-2 network jumbo frame; 122, a jumbo frame cutting module, configured to cut the layer-2 network jumbo frame and generate a data packet containing error correction information.

[0059] Obviously, those skilled in the art should 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 computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above control methods. The above modules or steps of the present invention can be implemented by a general computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. Optionally, they can be implemented by program codes executable by the computing device. Thus, they can be stored in a storage device and executed by the computing device, or they can be separately fabricated into individual integrated circuit modules, or multiple modules or steps among them can be fabricated into a single integrated circuit module to implement. In this way, the present invention is not limited to any specific combination of hardware and software.

[0060] Those skilled in the art can understand that to implement all or part of the processes in the above method embodiments, it can be completed by instructing relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above method embodiments of each control method. Among them, the storage medium can be a magnetic disk, an optical disc, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (HDD), or a solid-state drive (SSD), etc.; the storage medium can also include a combination of the above types of memories.

[0061] Embodiment 3 Furthermore, the present application proposes an electronic device, which is characterized in that it is used for any of the above implementation methods of carrying a layer-2 jumbo frame through the Internet based on intelligent routing technology, including: The first SD-WAN edge device is deployed at the sending end and is used for sending link detection packets, dynamically cutting the maximum transmission unit value, encapsulating data packets, and performing intelligent routing; The second SD-WAN edge device is deployed at the receiving end and is used for link quality feedback, verifying the received data packets, and performing data packet error correction and recombination.

[0062] It should be noted that the first SD-WAN edge device is deployed at the data sending end and has high-performance encapsulation and dynamic routing capabilities. The device uses UDP detection packets to detect the packet loss rate and delay of each link in real time, calculates the quality score by combining preset weights, and constructs a dynamic link quality table. For the input layer-2 jumbo frame (such as a 10-gigabit Ethernet frame), the device divides it into blocks based on the minimum value of the full-link MTU, and intelligently selects a link according to the number of blocks. And attach a hash check value and a sequence number to each block of data, and encapsulate it into an error-correcting data packet. The device supports multi-link parallel scheduling, maximizes throughput while ensuring a low packet loss rate, and adapts to high-demand scenarios such as financial synchronization and video surveillance.

[0063] The second SD-WAN edge device is deployed at the receiving end, focusing on data verification and recombination to ensure transmission reliability. The device feeds back link quality data to the sending end in real time to assist it in dynamically adjusting the routing strategy. When receiving data packets, the device first verifies the sequence number to identify out-of-order or lost packets, and verifies the data integrity through the hash value. If the verification fails, a retransmission request is immediately initiated, and only specific blocks need to be retransmitted instead of the entire frame. All valid data packets are cached in the splicing list according to the sequence number, and are automatically recombined into the original large frame when consecutive blocks are collected; if some blocks arrive late, the device judges the availability of the current data based on the service tolerance threshold. This design significantly reduces the recombination delay, especially suitable for scenarios sensitive to latency and integrity such as virtual machine migration and industrial Internet of Things.

[0064] Specifically, the first SD-WAN edge device is deployed at the sending end to send detection packets at a fixed frequency to test the quality of each link and sort the link quality. When receiving a large frame in the second-layer network, the first SD-WAN edge device cuts the large frame, judges the quality of the large frame according to the large frame block number threshold, enables the large frame to select the link with the corresponding quality for transmission, and generates data packets with an error correction mechanism, which are transmitted to the second SD-WAN edge device through the link.

[0065] The second SD-WAN edge device is deployed at the receiving end. After receiving the data packets sent by the first SD-WAN edge device, it determines the data validity of the data packets by verifying the hash value. If the data packets fail the verification, the second SD-WAN edge device sends feedback information to retransmit the data packets. Then, it confirms whether the data packets are expected or missing in the splicing list through the sequence number. If so, the data packets are added to the splicing list and wait for assembly. If not, the data packets are discarded. Next, the second SD-WAN edge device checks whether the data packets in the splicing list are complete. If so, it immediately assembles to complete this round of assembly. If not, it checks whether the gap between the latest sequence number and the missing sequence number is greater than the threshold. When it is less than the threshold, the second SD-WAN edge device considers that this round of assembly is still valid and continues to wait for the missing data packets. When it is greater than the threshold, the second SD-WAN edge device considers that this round of assembly is invalid, clears all the data packets in the splicing list, and sends a restart request to the first SD-WAN edge device.

[0066] The embodiments of the present application have been described above. The above description is exemplary and not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes are obvious to those of ordinary skill in the art in the technical field without departing from the scope and spirit of the described embodiments. The selection of the terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technologies in the market, or to enable other ordinary skill in the art in the technical field to understand the embodiments disclosed herein.

Claims

1. An implementation method for carrying a layer 2 jumbo frame over the Internet based on intelligent routing technology, characterized in that, Including: Based on the User Datagram Protocol, all link quality data is obtained according to the detection packets transmitted between the sender and the receiver, and a quality weight calculation result is generated. According to the cutting reference value of the link, the input Layer 2 network jumbo frame is cut into blocks to obtain the jumbo frame quality, the optimal link is selected, and an error correction data packet is generated for transmission. According to the received error correction data packet, verification is performed and it is added to the splicing list, the integrity of the data packets in the list is checked, and assembly or validity judgment is carried out.

2. The implementation method of carrying a layer-2 jumbo frame over the Internet based on the intelligent routing technology according to claim 1, wherein The step of "Based on the User Datagram Protocol, all link quality data is obtained according to the detection packets transmitted between the sender and the receiver, and a quality weight calculation result is generated" includes: The detection packet is a lightweight data packet containing a timestamp and a sequence number, which is sent at a fixed frequency in all links and is used to measure the packet loss rate and delay data of each link. The generation of the quality weight calculation result is the overall link quality score generated by weighted calculation according to the packet loss rate and delay data of each link, which is sorted and a link quality label is added.

3. The implementation method of carrying a layer-2 jumbo frame over the Internet based on intelligent routing technology as claimed in claim 1, wherein, The step of "According to the cutting reference value of the link, the input Layer 2 network jumbo frame is cut into blocks to obtain the jumbo frame quality, the optimal link is selected, and an error correction data packet is generated for transmission" includes: The cutting reference value of the link is the minimum value among the maximum transmission unit values of all links. The step of "cutting the input Layer 2 network jumbo frame into blocks to obtain the jumbo frame quality" is to cut the jumbo frame through the cutting reference value of the link, deduct the reserved bytes of the data packet, obtain the number of blocks by cutting the jumbo frame, and judge the quality of the current jumbo frame and add a jumbo frame quality label by setting a jumbo frame block number threshold. The selection of the optimal link is to find the corresponding link quality label according to the jumbo frame quality label. The error correction data packet is an additional hash check value and sequence number added to the split data packets of the secondary network jumbo frame generated after cutting, and is used to detect transmission errors and identify the block order.

4. The implementation method of carrying a layer 2 jumbo frame through the Internet based on the intelligent routing technology as described in claim 1, wherein, The step of "According to the received error correction data packet, verification is performed and it is added to the splicing list, the integrity of the data packets in the list is checked, and assembly or validity judgment is carried out" includes: For the step of "performing verification and adding to the splicing list", the receiver obtains the sequence number and hash check value in the error correction data packet for judgment. If the sequence number is the expected or missing sequence number and the hash check value is correctly verified, the data packet is added to the splicing list; otherwise, this data packet is discarded and a retransmission is requested. For the step of "checking the integrity of the data packets in the list", after the splicing list receives the data packet, it judges the receiving status of the data packet. If all are received, the data packet assembly starts; otherwise, the validity of the current round of assembly is judged.

5. The implementation method of carrying a layer 2 jumbo frame over the Internet based on intelligent routing technology according to claim 3, characterized in that, The selection of the optimal link is to find the corresponding link quality label according to the jumbo frame quality label, and includes: The jumbo frame block number threshold is set to n1 and n2, and n1 < n2, and the number of cut blocks of the Layer 2 network jumbo frame is set to N. When N ≤ n1, a low-quality jumbo frame label is added to this jumbo frame, and the link with the worst quality link quality label is selected for data packet transmission. When n1 < N ≤ n2, a medium-quality jumbo frame label is added to this jumbo frame, and the link with the middle-quality link quality label is selected for data packet transmission. When N > n2, add a high-quality superframe tag to the superframe, and select the link with the optimal quality as the link quality tag for data packet transmission.

6. The implementation method of carrying a layer-2 jumbo frame over the Internet based on intelligent routing technology as claimed in claim 4, wherein Check the integrity of the data packets in the checklist. After receiving the data packets in the splicing list, judge the receiving status of the data packets. If all are received, start data packet assembly; otherwise, judge the validity of the current round of assembly, including: The judgment of the validity of the current round of assembly is to judge the sequence gap between the missing data packets and the latest received data packets; When the gap is less than the set threshold, the current round of assembly is valid, and wait for the missing data packets to be resent; When the gap is greater than the set threshold, the current round of assembly is invalid, clear the received data packets and start over.

7. The implementation method of carrying a layer-2 jumbo frame over the Internet based on the intelligent routing technology according to claim 1, wherein When there is only one physical link between the sender and the receiver, it further includes: Create virtual links with different forward error correction coding strengths in the physical link, and distinguish the virtual link quality through detection packets to perform superframe transmission.

8. An implementation device for carrying a layer 2 jumbo frame over the Internet based on intelligent routing technology, characterized in that, The device includes: A link test module, configured to obtain all link quality data based on the detection packets transmitted between the sender and the receiver, and generate a quality weight calculation result; An intelligent routing module, configured to cut and divide the input layer-2 network superframe according to the link cutting reference value to obtain the superframe quality, select the optimal link, and generate error correction data packets for transmission; A receiving and assembly module, configured to check and add the received error correction data packets to the splicing list, check the integrity of the data packets in the checklist, and perform assembly or validity judgment.

9. The implementation device for carrying layer-2 superframes over the Internet based on intelligent routing technology according to claim 8, wherein the intelligent routing module includes: A link selection module, configured to sort the quality of all links and select the link with the corresponding quality according to the size of the layer-2 network superframe cut; A superframe cutting module, configured to cut the layer-2 network superframe and generate data packets containing error correction information.

10. An electronic device, characterized in that, The implementation method for carrying layer-2 superframes over the Internet based on intelligent routing technology according to any one of claims 1 to 7, including: The first SD-WAN edge device, deployed at the sender, is configured to send link detection packets, perform dynamic maximum transmission unit value cutting, encapsulate data packets, and perform intelligent routing; The second SD-WAN edge device, deployed at the receiver, is configured to provide link quality feedback, check the received data packets, and perform data packet error correction and recombination.

Images