SD-WAN dynamic routing method based on routing query

Abstract

The invention relates to the technical field of routing query, and particularly discloses an SD-WAN dynamic routing method based on routing query, which comprises the following steps: S1, acquiring the maximum throughput and the actual throughput of each router, calculating the residual throughput of the router, and taking the minimum value of the residual throughput in the same link as the path capacity of the link; s2, taking the link corresponding to the maximum path capacity as a main link, taking the remaining links as auxiliary links, obtaining the byte number of the transmission, presetting the minimum delay time, calculating the ideal byte number of the main link, judging whether the main link is crowded or not, and calculating an initial distribution proportion; and S3, transmitting the test information along each auxiliary link, obtaining the propagation delay and packet loss rate of each auxiliary link, correcting the initial distribution proportion to obtain a correction proportion, calculating a transmission task of the auxiliary link, and transmitting the transmission task. According to the invention, by adjusting the transmission strategy, the problem of congestion under burst traffic is solved.

Description

Technical Field

[0001] The present invention relates to the field of routing query technology, and in particular to an SD-WAN dynamic routing method based on routing query. Background Art

[0002] The core function of SD-WAN dynamic routing is to achieve intelligent path optimization. During actual network operation, it continuously monitors the network status through multiple protocols. These protocols collect detailed information about each node and link in the network, including key metrics such as bandwidth utilization, latency, and packet loss rate. Based on this rich data, the dynamic routing mechanism uses complex algorithms for in-depth analysis and calculation, automatically selecting the optimal path under current network conditions.

[0003] However, in practice, primary link congestion can occur. Due to the uncertainty of network traffic, under high load or bursty conditions, the primary link may reach or exceed its maximum throughput and become congested, impacting data transmission efficiency and stability. In traditional networks, traffic is often concentrated on the primary link, and secondary link resources are not effectively utilized, resulting in overall network performance degradation. Traditional static configuration methods lack flexibility in the face of rapidly changing network conditions, making it impossible to adjust transmission strategies in a timely manner to meet current needs. Summary of the Invention

[0004] The purpose of the present invention is to provide an SD-WAN dynamic routing method based on routing query to solve the above technical problems.

[0005] The purpose of the present invention can be achieved through the following technical solutions: A SD-WAN dynamic routing method based on routing query includes the following steps: S1: Obtain the maximum throughput MT and actual throughput CT of each router in the same link. The maximum throughput MT represents the maximum data flow processed by the router per unit time, and the actual throughput CT represents the data flow processed by the router per unit time in actual operation. Calculate the remaining throughput of the router ST = MT-CT, and obtain the minimum value ST of the remaining throughput on the same link min , let the path capacity of this link LT = ST min ; S2: Select the link with the maximum path capacity as the primary link and the remaining links as auxiliary links. Obtain the number of bytes B transmitted this time, pre-set the minimum delay time t, and calculate the ideal number of bytes ZB=LT for the primary link. main ×t, where LT main Represents the path capacity corresponding to the main link; If the number of bytes B ≥ 0.8 ZB, it is recorded as the main link congestion. When the main link is congested, the initial allocation ratio BL is calculated based on the path capacity of each auxiliary link; S3: Transmit the preset test information along each auxiliary link. The test information has the same content and the same number of bytes. Obtain the propagation delay Y and packet loss rate D of each auxiliary link at the end of the test information transmission. Modify the initial allocation ratio BL of the auxiliary link. The modified initial allocation ratio is recorded as the modified ratio XL. The modified ratio , where λ1 represents the preset first coefficient, and λ2 represents the preset second coefficient; Calculate the transmission task of the auxiliary link , let the main link transmit 0.8ZB bytes, and the auxiliary link transmits the bytes corresponding to the transmission task R, where XL all Represents the sum of the correction ratios.

[0006] As a further solution of the present invention: in the step S1, if the remaining throughput ST≤0.2MT, it is recorded as a busy link, and the busy link does not participate in the subsequent selection of the primary link and the auxiliary link.

[0007] As a further solution of the present invention: in the step S1, if there is a routing loop between the primary link and the auxiliary link, it is removed and does not participate in subsequent operations.

[0008] As a further solution of the present invention: in step S2, when there are links with equal path capacity, the link with fewer routers is recorded as the main link.

[0009] As a further solution of the present invention: in step S2, if the number of bytes B is less than 0.8 ZB, it is recorded that the main link is idle, and subsequent steps are stopped, and all data is transmitted through the main link.

[0010] As a further solution of the present invention: in step S2, the method for calculating the initial allocation ratio BL based on the path capacity of each auxiliary link includes: Get the total path capacity LT of all auxiliary links all , the initial allocation ratio BL of the i-th auxiliary link i =LT i / LT all , among which, LT i Represents the path capacity of the i-th auxiliary link.

[0011] As a further solution of the present invention: in the step S3, the header and tail of the test information each contain a segment of the same verification bit. When the test information is received, if the contents of the verification bits in the header and tail are the same, the subsequent steps are executed; if they are different, the test information is resent.

[0012] As a further solution of the present invention: in step S3, auxiliary links with propagation delay Y>t are eliminated and do not participate in subsequent operations.

[0013] Beneficial effects of the present invention: First, the remaining throughput in the same link needs to be calculated. It should be noted that although there are multiple routers on the same link, the router with the smallest remaining throughput on the link can determine the ultimate data transmission capacity of the link. Using the minimum value of the remaining throughput in the same link as the path capacity of the link can avoid the situation where congestion occurs due to exceeding its maximum throughput again. It should be noted here that the routers in the link do not only serve a certain data transmission. Therefore, it is necessary to reserve some data transmission space for data transmission in other links.

[0014] Then, based on the calculated path capacity, the link with the maximum path capacity is selected as the primary link. The path capacity here refers to the maximum throughput of the router minus the actual throughput. From this, we can see that the path capacity refers to the maximum data transmission capability that the current router can provide. The reason for selecting the link with the largest path capacity as the primary link is to reduce the use of auxiliary links for data transmission. Because, under the same conditions, the fewer links are used for transmission, the lower the probability of errors in the entire transmission process. Sometimes, when the primary link is sufficiently idle, the primary link can independently assume the data transmission task, without the need for auxiliary links for transmission, thereby improving transmission efficiency and stability.

[0015] Afterwards, the minimum delay time is set according to the specific delay requirements, and the ideal number of bytes within this period is calculated based on the product of the minimum delay time and the path capacity corresponding to the main link. When the number of bytes transmitted this time is greater than or equal to 80% of the ideal number of bytes, it is recorded as congestion of the main link. At this time, relying solely on the main link for transmission will result in too high a delay, so it is necessary to use the auxiliary link for data transmission. However, the need for data transmission through the auxiliary link requires regulations on the amount of data responsible for transmission on each auxiliary link, which also needs to be controlled within a certain propagation delay. Therefore, it is necessary to set test information here, and obtain the propagation delay and packet loss rate of each auxiliary link based on the test information. Then, the correction ratio is calculated through a weighted formula, and then the transmission task undertaken by each auxiliary link is calculated based on the proportion of the correction ratio of each auxiliary link. The present invention solves the problem of congestion under burst traffic by adjusting the transmission strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will be further described below with reference to the accompanying drawings.

[0017] Figure 1This is a flow chart of an SD-WAN dynamic routing method based on routing query in the present invention. DETAILED DESCRIPTION

[0018] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0019] See also Figure 1 As shown, the present invention is an SD-WAN dynamic routing method based on routing query, comprising the following steps: S1: Obtain the maximum throughput MT and actual throughput CT of each router in the same link. The maximum throughput MT represents the maximum data flow processed by the router per unit time, and the actual throughput CT represents the data flow processed by the router per unit time in actual operation. Calculate the remaining throughput of the router ST = MT-CT, and obtain the minimum value ST of the remaining throughput on the same link min , let the path capacity of this link LT = ST min ; S2: Select the link with the maximum path capacity as the primary link and the remaining links as auxiliary links. Obtain the number of bytes B transmitted this time, pre-set the minimum delay time t, and calculate the ideal number of bytes ZB=LT for the primary link. main ×t, where LT main Represents the path capacity corresponding to the main link; If the number of bytes B ≥ 0.8 ZB, it is recorded as the main link congestion. When the main link is congested, the initial allocation ratio BL is calculated based on the path capacity of each auxiliary link; S3: Transmit the preset test information along each auxiliary link. The test information has the same content and the same number of bytes. Obtain the propagation delay Y and packet loss rate D of each auxiliary link at the end of the test information transmission. Modify the initial allocation ratio BL of the auxiliary link. The modified initial allocation ratio is recorded as the modified ratio XL. The modified ratio , where λ1 represents the preset first coefficient, and λ2 represents the preset second coefficient; Calculate the transmission task of the auxiliary link , let the main link transmit 0.8ZB bytes, and the auxiliary link transmits the bytes corresponding to the transmission task R, where XL all Represents the sum of the correction ratios.

[0020] It's important to note that calculating the residual throughput on a link is crucial during network data transmission. In real-world networks, a link often contains multiple routers, working together to forward and transmit data. However, even though these routers are all on the same link, their data processing capabilities and resource availability may differ. Of particular note, even though multiple routers exist on the same link, the key factor determining the link's ultimate data transmission capacity lies in the router with the lowest residual throughput. This is because data transmission on a link, like water flowing through a pipe, is subject to limitations imposed by each node, or router. If a router has a low residual throughput, even if other routers have greater transmission capacity, the data transmission speed of the entire link will be limited by this bottleneck. Based on this principle, the minimum residual throughput of all routers on a link is used to determine the path capacity of that link. This approach is crucial because it effectively prevents congestion caused by exceeding the link's maximum throughput. When the amount of data transmitted exceeds the actual carrying capacity of the link, data will accumulate in the link and cannot be processed and forwarded in a timely manner, which will cause network congestion and seriously affect network performance and data transmission efficiency.

[0021] It's important to note that the routers in a link aren't solely responsible for a single data transmission task. In actual network operation, these routers must simultaneously handle large amounts of data traffic from multiple sources and destinations. They act like the dispatch center at a transportation hub, effectively allocating resources to ensure smooth data transmission in all directions.

[0022] Therefore, when calculating the remaining throughput and determining the path capacity, some space must be reserved for data transmission on other links. This reserved space is like an emergency lane or buffer zone on a road, ensuring the stability and reliability of the entire network in emergencies or when data traffic on other links suddenly increases.

[0023] After calculating the remaining throughput on a given link and determining the path capacity, the next step is to select an appropriate primary link based on the path capacity. The path capacity mentioned here refers to the router's maximum throughput minus its current actual throughput. This provides a clear understanding of each router's potential for data transmission under the current network conditions. Therefore, the path capacity accurately reflects the maximum data transmission capacity available from the current router, taking into account the router's own performance limits and the amount of data transmission currently being handled. One of the core goals of selecting the link with the largest path capacity as the primary link is to minimize the use of auxiliary links for data transmission. In a complex network environment, every link and router carries certain risks and uncertainties. When data needs to traverse multiple links for transmission, the probability of problems increases. Transmission delays on different links can vary, potentially leading to out-of-order arrival of data packets. Alternatively, a router failure on an auxiliary link can disrupt the smooth flow of data transmission.

[0024] However, in real-world applications, when the primary link is sufficiently idle, it is fully capable of independently handling data transmission. In such cases, rationally arranging data transmission entirely over the primary link can avoid the use of the auxiliary link, significantly improving transmission efficiency and stability. Overall, allowing the primary link to fully utilize its idle time not only fully utilizes network resources, but also reduces the various issues that may arise from multi-link transmission, ensuring that data reaches its destination quickly and stably.

[0025] After selecting the primary link, you need to set the minimum latency based on your specific latency requirements. This step directly impacts the real-time and efficiency of data transmission. Minimum latency is a comprehensive consideration of business needs and network performance. It represents the ideal minimum time required for data to travel from source to destination.

[0026] Once we have the minimum delay time, we can calculate the ideal number of bytes within that timeframe by multiplying this time by the path capacity corresponding to the primary link. This ideal number of bytes refers to the maximum amount of data that the primary link can theoretically transmit without blocking within the minimum delay time, providing an important indicator for measuring whether the current transmission task is approaching or exceeding the carrying capacity of the primary link. When the number of bytes transmitted is greater than or equal to 80% of the ideal number of bytes, the primary link is considered congested. This is because in real-world network environments, if the primary link is overloaded, its transmission delay will increase significantly, potentially even leading to packet loss or retransmissions, severely impacting transmission efficiency and stability. Therefore, relying solely on the primary link for transmission in this situation is unwise; a secondary link is needed to share the transmission workload and alleviate pressure on the primary link.

[0027] However, data transmission over auxiliary links is not simply about distributing data evenly across them. This is because different auxiliary links may have varying performance, including propagation delay and packet loss rate. To ensure efficient and stable data transmission, the amount of data transmitted on each auxiliary link must be appropriately specified. This also requires control within a certain propagation delay to prevent excessive delay on the auxiliary link from impacting overall transmission performance. To distribute transmission tasks across auxiliary links, test information must be set. This test information can be specific data packets or data streams, which are sent to each auxiliary link and key transmission metrics, such as propagation delay and packet loss rate, are recorded. By collecting this test information, a comprehensive understanding of the performance of each auxiliary link can be obtained.

[0028] We can then calculate the correction ratio using a weighted formula. This correction ratio is calculated based on metrics such as the auxiliary link's propagation delay and packet loss rate, reflecting the auxiliary link's actual transmission capacity under the current network conditions. A higher correction ratio means a higher transmission load for the auxiliary link, and vice versa.

[0029] Finally, the transmission tasks assigned to each auxiliary link are calculated based on the corrected ratios of each link. This allows us to dynamically adjust the transmission task allocation of each auxiliary link based on the actual network conditions to ensure efficient and stable data transmission. This flexible transmission strategy adjustment effectively addresses congestion issues caused by bursty traffic, improving overall network performance and reliability.

[0030] In another preferred embodiment of the present invention, if the remaining throughput ST≤0.2MT, it is recorded as a busy link, and the busy link does not participate in the subsequent selection of the primary link and the auxiliary link.

[0031] It's worth noting that before entering the primary link selection process, all links must be screened in advance. This step simplifies the subsequent selection of primary and secondary links. During the screening process, attention is paid to each link's remaining throughput. Remaining throughput, a key metric for measuring a link's currently available transmission capacity, directly reflects the link's potential to handle data transmission tasks under the current network environment. When a link's remaining throughput is less than 20% of its maximum throughput, it indicates that the link is already highly busy and its available transmission resources are limited. In this case, even if the link is included in the subsequent selection of primary or secondary links, its insufficient carrying capacity will likely lead to congestion, delays, or even data loss, thus affecting the stability and reliability of the entire data transmission.

[0032] To simplify operations, we hope to avoid unnecessary calculations and judgment complexity during the subsequent complex primary and secondary link selection process, thereby improving link selection efficiency. At the same time, to ensure data transmission stability, the impact of the link's actual carrying capacity on data transmission cannot be ignored. Therefore, busy links with a remaining throughput below 20% are marked as busy and removed from the subsequent primary and secondary link selection lists. This pre-screening mechanism can initially ensure that the selected links have relatively sufficient transmission capacity, building a stable and efficient data transmission channel and allowing data to flow smoothly within the network.

[0033] In another preferred embodiment of the present invention, if a routing loop exists between the primary link and the auxiliary link, it is removed and does not participate in subsequent operations.

[0034] It is understandable that if a routing loop exists in the set of primary links and auxiliary links, in order to ensure the efficiency of data transmission across the entire network, the link with the routing loop will be removed so that it does not participate in any subsequent operations. The existence of a routing loop will cause data to continuously circulate in the loop once it enters the loop, and it will not be able to reach its destination smoothly. This will not only greatly waste network resources, resulting in meaningless bandwidth occupation, affecting other normal data transmission tasks, causing freezes or even interruptions, but will also increase the delay of data transmission infinitely, seriously violating our requirements for real-time and efficient data transmission.

[0035] In another preferred embodiment of the present invention, when there are links with equal path capacity, the link with the smaller number of routers is recorded as the main link.

[0036] It's important to note that in complex network environments, path capacity is a crucial metric for measuring link transmission capabilities. When multiple links have the same path capacity, other factors must be considered to optimize the selection. Routers, as key nodes in the network, have a multifaceted impact on data transmission. Links with fewer routers require data to traverse fewer forwarding nodes during transmission, reducing data packet processing delays and improving the real-time nature of data transmission, allowing data to reach its destination more quickly.

[0037] On the other hand, fewer routers also means lower transmission complexity and greater stability. Each time a data packet passes through a router, it undergoes a series of processing operations, such as route lookup and queue management. These operations not only consume a certain amount of time but can also lead to packet loss or transmission errors due to excessive router load or other failures. Therefore, choosing a link with fewer routers as the primary link can effectively reduce these potential risks, mitigate potential data transmission issues, and improve the reliability and stability of the entire network.

[0038] In another preferred embodiment of the present invention, if the number of bytes B is less than 0.8 ZB, it is recorded that the main link is idle, and subsequent steps are stopped, and all data are transmitted through the main link.

[0039] It should be noted that the number of bytes B currently required to be transmitted is less than 0.8 times the ideal number of bytes ZB. In this case, since the remaining transmission capacity of the primary link is sufficient to easily handle the current data transmission task and no auxiliary link is needed to share the transmission pressure, all data is transmitted through the primary link to simplify the transmission process, improve transmission efficiency, and reduce unnecessary resource usage. As the path with strong transmission capacity and relative stability in the network, the primary link can complete data transmission tasks with higher efficiency and lower latency when idle.

[0040] In another preferred embodiment of the present invention, the method for calculating the initial allocation ratio BL based on the path capacity of each auxiliary link includes: Get the total path capacity LT of all auxiliary links all , the initial allocation ratio BL of the i-th auxiliary link i =LT i / LT all , among which, LT i Represents the path capacity of the i-th auxiliary link.

[0041] In another preferred embodiment of the present invention, the header and tail of the test information each contain a segment of the same verification bit. When the test information is received, if the contents of the verification bits in the header and tail are the same, the subsequent steps are executed; if they are different, the test information is resent.

[0042] It's worth noting that the test message contains identical verification bits in both its header and trailer. These unique and verifiable verification bits ensure the integrity and accuracy of the test message during transmission. When the receiving end successfully receives the test message, it immediately compares the verification bits in the header and trailer. If the verification bits in the header and trailer are identical, this indicates that the test message has not been tampered with, lost, or damaged during transmission, maintaining its original integrity and accuracy. Subsequent steps can then be performed.

[0043] However, if the verification bits in the header and tail differ during the comparison process, this indicates that some anomaly may have occurred during transmission, resulting in modification or partial loss of the test information. To obtain accurate and reliable test results, the potentially problematic test information cannot be used directly for subsequent operations. Instead, the test information must be resent. The purpose of resending the test information is to obtain a complete and accurate copy of the test data, ensuring that subsequent steps can proceed smoothly based on the correct data and thus ensuring the effectiveness of the entire testing process.

[0044] In another preferred embodiment of the present invention, auxiliary links with propagation delay Y>t are eliminated and do not participate in subsequent operations.

[0045] It's worth noting that when the propagation delay Y of a particular auxiliary link during network data transmission exceeds the preset minimum delay time t, there's a high probability that data packets will be delayed during transmission using this auxiliary link. Once this delay occurs, data will backlog on the transmission path, significantly increasing latency across the entire system.

[0046] In network systems, increased latency can have serious negative impacts on data real-time performance, system responsiveness, and user experience. To avoid these adverse consequences caused by using auxiliary links with excessive propagation delays and ensure efficient and stable operation of the entire network system, auxiliary links with propagation delays exceeding a preset minimum delay are removed. This eliminates potential latency risks introduced by these auxiliary links at the source, ensuring data transmission over the remaining high-quality links. This maintains the low latency characteristics of the entire network system and ensures smooth operation of various network applications.

[0047] The above is a detailed description of an embodiment of the present invention. However, the content is only a preferred embodiment of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the present invention.

Claims

1. A SD-WAN dynamic routing method based on routing query, characterized in that: The following steps are involved: S1: Obtain the maximum throughput MT and actual throughput CT of each router in the same link. The maximum throughput MT represents the maximum data flow processed by the router per unit time, and the actual throughput CT represents the data flow processed by the router per unit time in actual operation. Calculate the router's remaining throughput ST = MT - CT, and obtain the minimum value ST of the remaining throughput on the same link. min , let the path capacity of this link LT=ST min ; S2: Select the link with the maximum path capacity as the primary link and the remaining links as auxiliary links. Obtain the number of bytes B transmitted this time, pre-set the minimum delay time t, and calculate the ideal number of bytes ZB=LT for the primary link. main ×t, where LT main Represents the path capacity corresponding to the main link; If the number of bytes B ≥ 0.8 ZB, it is recorded as the main link congestion. When the main link is congested, the initial allocation ratio BL is calculated based on the path capacity of each auxiliary link; S3: Transmit the preset test information along each auxiliary link. The test information has the same content and the same number of bytes. Obtain the propagation delay Y and packet loss rate D of each auxiliary link at the end of the test information transmission. Modify the initial allocation ratio BL of the auxiliary link. The modified initial allocation ratio is recorded as the modified ratio XL. The modified ratio , where λ1 represents the preset first coefficient, and λ2 represents the preset second coefficient; Calculate the transmission task of the auxiliary link , let the main link transmit 0.8ZB bytes, and the auxiliary link transmits the bytes corresponding to the transmission task R, where XL all Represents the sum of the correction ratios.

2. The SD-WAN dynamic routing method based on routing query according to claim 1, characterized in that: In the step S1, if the remaining throughput ST≤0.2MT, it is recorded as a busy link, and the busy link does not participate in the subsequent selection of the primary link and the auxiliary link.

3. The SD-WAN dynamic routing method based on routing query according to claim 1, characterized in that: In the step S1, if there is a routing loop between the primary link and the auxiliary link, it is removed and does not participate in subsequent operations.

4. The SD-WAN dynamic routing method based on routing query according to claim 1, characterized in that: In step S2, when there are links with equal path capacity, the link with the smaller number of routers is recorded as the main link.

5. The SD-WAN dynamic routing method based on routing query according to claim 1, characterized in that: In step S2, if the number of bytes B is less than 0.8 ZB, it is recorded that the main link is idle, and subsequent steps are stopped, and all data is transmitted through the main link.

6. The SD-WAN dynamic routing method based on routing query according to claim 1, characterized in that: In step S2, the method for calculating the initial allocation ratio BL based on the path capacity of each auxiliary link includes: Get the total path capacity LT of all auxiliary links all , the initial allocation ratio BL of the i-th auxiliary link i =LT i / LT all , among which, LT i Represents the path capacity of the i-th auxiliary link.

7. The SD-WAN dynamic routing method based on routing query according to claim 1, characterized in that: In step S3, the header and tail of the test information each contain a same verification bit. When the test information is received, if the contents of the verification bits in the header and tail are the same, the subsequent steps are executed. If they are different, the test information is resent.

8. The SD-WAN dynamic routing method based on routing query according to claim 1, characterized in that: In step S3, auxiliary links with propagation delay Y>t are removed and do not participate in subsequent operations.

Images