Low-delay transmission control method and system of Ethernet switch

Abstract

The invention discloses a low-delay transmission control method and system for an Ethernet switch, relates to the technical field of network communication, and solves the problem that in the prior art, when a large number of data packets are transmitted at the same time, a preset dequeue rate is used for transmission, so that queue blockage is caused. If an efficient dequeue rate is always used for transmission, resource waste is caused, and certain flexibility is lacked. The method comprises the steps of obtaining a plurality of to-be-transmitted data packets; analyzing and processing the plurality of to-be-transmitted data packets to obtain a traffic type and an emergency degree; distributing the plurality of data packets to a plurality of corresponding message queues based on the traffic types; adjusting the transmission of the plurality of data packets based on the emergency degree; monitoring the plurality of message queues, and scheduling transmission of the message queues according to monitoring data; network fluctuation can be quickly responded, a transmission strategy is automatically adjusted, and stability is kept.

Description

Technical Field

[0001] This application belongs to the field of network communication and relates to low-latency transmission control technology, specifically a low-latency transmission control method and system for an Ethernet switch. Background Technology

[0002] An Ethernet switch is a network device based on Ethernet technology, used to enable efficient data transmission between multiple devices within a local area network (LAN). By parsing the MAC address in a data frame, it intelligently forwards data to the target device, making it a core component of modern network architecture. Low-latency transmission control by Ethernet switches significantly improves network performance, ensuring high-priority traffic is forwarded first, absorbing short-term traffic bursts, and avoiding increased latency caused by packet loss and retransmissions. Simultaneously, through time synchronization and traffic shaping technologies, it provides microsecond-level deterministic latency guarantees, allocates dedicated bandwidth for critical applications, and avoids interference from other traffic. Low-latency transmission control provides deterministic performance guarantees for critical services by reducing packet processing, forwarding, and transmission time, while optimizing network resource utilization.

[0003] In existing technologies, data packets are typically transmitted based on the time of receipt and the timeliness of the data packets, allowing for the priority transmission of urgent data packets. However, when a large number of data packets are being transmitted simultaneously, using a preset dequeue rate can lead to queue congestion. If a high dequeue rate is always used, it will result in a waste of resources and a lack of flexibility.

[0004] This application provides a low-latency transmission control method and system for Ethernet switches to solve the above-mentioned technical problems. Summary of the Invention

[0005] This application aims to solve at least one of the technical problems existing in the prior art; to this end, this application proposes a low-latency transmission control method and system for Ethernet switches to solve the technical problem that when a large number of data packets are transmitted simultaneously, using a preset dequeue rate for transmission will cause queue congestion, and if a high-efficiency dequeue rate is always used for transmission, it will result in waste of resources and lack of flexibility.

[0006] To achieve the above objectives, a first aspect of this application provides a low-latency transmission control method for an Ethernet switch, comprising: Obtain several data packets to be transmitted; Analyze and process several data packets to be transmitted to obtain the traffic type and urgency level; Distribute several data packets to several corresponding message queues based on traffic type; The transmission of several data packets is adjusted based on their urgency. Several message queues are monitored, and the transmission of message queues is scheduled based on the monitoring data.

[0007] Preferably, the step of analyzing and processing several data packets to be transmitted to obtain the traffic type and urgency level includes: Retrieve several data packets to be transmitted; extract features from these data packets to obtain basic features and traffic statistics features; among them, the basic features include: packet size, arrival time, protocol type, and port number; The basic characteristics of several data packets are integrated with traffic statistical characteristics to obtain a traffic analysis sequence; the traffic classification model is then called, and the traffic analysis sequence is input into the traffic classification model to obtain the traffic type corresponding to the data packets; the traffic classification model is constructed based on an artificial intelligence model; Obtain the transmission parameters of the data packets; analyze and calculate the urgency of the data packets based on the traffic type and transmission parameters.

[0008] Preferably, the step of extracting features from several data packets to obtain basic features and traffic statistics features includes: Retrieve several data packets; identify and analyze the fields of the data packets to obtain their basic characteristics; Based on formula Calculate the average packet size of data packets within a specified time period; based on the formula. Calculate the standard deviation of package size; where, This represents the size of the i-th data packet; This sets the total number of data packets within a given time period. Based on formula Calculate the average arrival interval of data packets within a specified time period; based on the formula Calculate the standard deviation of the arrival interval; where, This represents the arrival time of the i-th data packet; The average packet size, standard deviation of packet size, average arrival time, and standard deviation of arrival time are integrated into flow statistics features.

[0009] Preferably, the traffic classification model is constructed based on an artificial intelligence model, including: Obtain the standard dataset; the standard dataset includes standard input data consistent with the content attributes of the traffic analysis sequence, and standard output data consistent with the content attributes of the traffic type; Select a model framework and deep learning algorithm from the artificial intelligence library; build and improve the model framework based on the deep learning algorithm to obtain a deep learning model; The standard dataset is divided into a training set, a validation set, and a test set according to a preset ratio; the deep learning model is trained using the training set; the internal parameters of the deep learning model are adjusted using the validation set; and the deep learning model is tested using the test set to obtain test metrics. Obtain the indicator threshold; compare the test indicator with the indicator threshold; if all test indicators are greater than the indicator threshold, then mark the deep learning model as a traffic classification model; otherwise, rebuild and retrain the traffic classification model.

[0010] Preferably, the step of analyzing and calculating the urgency of data packets based on traffic type and transmission parameters includes: Obtain the basic urgency level of the traffic type; retrieve the transmission parameters of the data packets; among which, the transmission parameters include: packet loss rate and latency tolerance; The urgency level of a data packet is calculated based on an urgency analysis function; the urgency analysis function is: ; in, Based on the level of urgency; This is the performance sensitivity coefficient; The target latency for the corresponding traffic type; This represents the measured average latency of the most recent N packets; The base coefficient is affected by the packet loss rate; This is the response curvature control coefficient for packet loss rate; Packet loss rate; Sensitivity coefficient for low packet loss rate; The power-law decay intensity; For delay tolerance; For reference delay time; The decay index; It is a very small constant.

[0011] It should be noted that the performance sensitivity coefficient, the baseline coefficient for packet loss rate impact, the control coefficient for packet loss rate response curvature, the low packet loss rate sensitivity coefficient, the power law attenuation intensity, the reference delay time, the attenuation exponent, and the minimum constant are all set by professional technicians according to the actual application scenario.

[0012] Preferably, the step of allocating several data packets to several corresponding message queues based on traffic type includes: Retrieve the traffic type of data packets within a specified time period; filter and integrate data packets with the same traffic type to obtain several sets of data packet sequences; Obtain the traffic transmission library; match the traffic types corresponding to several sets of data packet sequences with the traffic transmission library to obtain the message queues for the transmission of several sets of data packet sequences.

[0013] Preferably, the adjustment of the transmission of several data packets based on urgency includes: Retrieve the urgency level of each data packet in several sets of data packet sequences; sort the data packets in the several sets of data packet sequences in descending order of urgency; Obtain the urgency threshold; queue data packets with an urgency level greater than the urgency threshold into the urgency message queue; allocate the remaining data packets to the corresponding message queues according to the sorting results.

[0014] Preferably, the step of scheduling message queue transmissions based on monitoring data includes: Retrieve monitoring data from the message queue; the monitoring data includes: message queue length and message queue network load; The transmission of messages in the message queue is adjusted based on the dequeue rate adjustment function; the expression of the dequeue rate adjustment function is: ;in, The baseline dequeue rate; This is the queue congestion sensitivity coefficient; To normalize queue congestion; Network load sensitivity coefficient; This is a normalized network load stress indicator. The expression for the normalized queue congestion degree is: ;in, This represents the current queue length. The optimal queue length; This represents the maximum capacity of the queue. This represents the growth rate of the queue length. The threshold for stable growth; The expression for the normalized network load pressure index is: ;in, ; For link utilization; Historical packet loss rate; Average queuing delay for messages within the queue; The target delay threshold; , as well as These are the weighting coefficients; This is the steepness coefficient of the Sigmoid function; This is the load-sensitive threshold.

[0015] A second aspect of this application provides a low-latency transmission control system for an Ethernet switch, comprising: a data acquisition module, an emergency analysis module, and a queue adjustment module; The data acquisition module is used to acquire several data packets to be transmitted; The emergency analysis module is used to analyze and process several data packets to be transmitted to obtain the traffic type and urgency level; and to allocate several data packets to several corresponding message queues based on the traffic type. The queue adjustment module is used to adjust the transmission of several message queues based on their urgency; monitor several message queues; and schedule the transmission of data packets based on the monitoring data.

[0016] A third aspect of this application provides a computer-readable storage medium storing instructions for performing method steps as described in the first aspect and any possible implementation thereof.

[0017] Compared with the prior art, the beneficial effects of this application are: 1. This application comprehensively characterizes the spatiotemporal distribution patterns of data packets by extracting basic and statistical features; its classification mechanism based on an artificial intelligence model can automatically learn traffic patterns and adapt to new business or attack traffic; it comprehensively evaluates the urgency level through a multi-parameter weighting function, covering multiple dimensions of requirements such as latency, packet loss, and business tolerance; it integrates data packets of the same traffic type into a dedicated queue to avoid mutual interference between different businesses and improve resource utilization; it adjusts the transmission order within the queue based on the urgency level value to ensure that highly urgent data packets are processed first; and it reduces manual intervention through intelligent scheduling while meeting users' needs for low latency and high reliability.

[0018] 2. This application ensures that high-priority data packets are always transmitted first by sorting them in descending order of urgency, avoiding service failures due to queue waiting; the urgency value calculated based on multiple parameters accurately reflects the urgency of data packets, avoiding the rigidity of traditional priority systems; the emergency queue is independent of the ordinary queue, preventing sudden high-urgency traffic from impacting ordinary services; the dequeue rate is dynamically adjusted through an exponential decay function to balance queue congestion and network load, avoiding resource waste or overload; and both queue and network conditions are considered to prevent erroneous adjustments caused by a single factor. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of the overall method steps of this application; Figure 2 This is a schematic diagram illustrating the urgency analysis and transmission allocation steps for this application type; Figure 3 This is a schematic diagram illustrating the transmission adjustment steps of this application; Figure 4 This is a schematic diagram of the system module structure connection of this application. Detailed Implementation

[0021] The technical solutions of this application will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0022] Please see Figure 1 The first aspect of this application provides a low-latency transmission control method for an Ethernet switch, comprising: S101. Obtain several data packets to be transmitted; S102. Analyze and process several data packets to be transmitted to obtain the traffic type and urgency level; S103. Distribute several data packets to several corresponding message queues based on traffic type; S104. Adjust the transmission of several data packets based on their urgency level; S105. Monitor several message queues and schedule the transmission of message queues based on the monitoring data.

[0023] Based on the above steps, by analyzing the traffic type of data packets, data packets for different services are allocated to dedicated message queues to avoid interference between different traffic types; the transmission order is dynamically adjusted based on the urgency level to ensure that critical data packets are processed first, reducing the risk of delay or loss; different traffic types use independent queues to prevent one type of traffic from occupying all bandwidth and causing other services to lag; by monitoring the queue status in real time, the transmission strategy is dynamically adjusted to balance the load and improve throughput; urgent data is transmitted first to reduce latency, queue isolation avoids resource contention, and dynamic scheduling alleviates congestion, thereby reducing packet loss rate and retransmission frequency; the real-time monitoring and scheduling mechanism enables the system to respond quickly to network fluctuations, automatically adjust the transmission strategy, and maintain stability.

[0024] In one possible implementation of the embodiments of this application, combined with Figure 1 ,like Figure 2 As shown, the above S102-S103 can be specifically implemented through the following S201-S203, which are explained in detail below: S201. Retrieve several data packets to be transmitted; identify and analyze the fields of the several data packets to obtain the basic characteristics of the several data packets.

[0025] The basic characteristics include: packet size, arrival time, protocol type, and port number.

[0026] S202, Based on Formula Calculate the average packet size of data packets within a specified time period; based on the formula. Calculate the standard deviation of package size.

[0027] in, This represents the size of the i-th data packet; This sets the total number of data packets within a given time period.

[0028] S203, Based on Formula Calculate the average arrival interval of data packets within a specified time period; based on the formula Calculate the standard deviation of arrival interval time; integrate the average package size, standard deviation of package size, average arrival interval time, and standard deviation of arrival interval time into flow statistics features.

[0029] in, This represents the arrival time of the i-th data packet.

[0030] S204. Integrate the basic characteristics of several data packets with the traffic statistics characteristics to obtain a traffic analysis sequence; call the traffic classification model, input the traffic analysis sequence into the traffic classification model, and obtain the traffic type corresponding to the data packet.

[0031] In one possible implementation, the traffic classification model is built upon an artificial intelligence model, including: Obtain the standard dataset; the standard dataset includes standard input data consistent with the content attributes of the traffic analysis sequence, and standard output data consistent with the content attributes of the traffic type; Select a model framework and deep learning algorithm from the artificial intelligence library; build and improve the model framework based on the deep learning algorithm to obtain a deep learning model; The standard dataset is divided into a training set, a validation set, and a test set according to a preset ratio; the deep learning model is trained using the training set; the internal parameters of the deep learning model are adjusted using the validation set; and the deep learning model is tested using the test set to obtain test metrics. Obtain the indicator threshold; compare the test indicator with the indicator threshold; if all test indicators are greater than the indicator threshold, then mark the deep learning model as a traffic classification model; otherwise, rebuild and retrain the traffic classification model.

[0032] Example: A company network is simultaneously running three types of services: video conferencing (VoIP), file transfer (FTP), and web browsing (HTTP). A sample of 100 data packets was collected during a certain time period. The specific distribution is as follows: Video conferencing: 30 packets, average packet size 1400 bytes, arrival interval 0.08 seconds, protocol UDP / 1234, basic urgency level U_b=0.9; File transfer: 40 packets, average packet size 2000 bytes, arrival interval 0.15 seconds, protocol TCP / 21, basic urgency U_b=0.5; Web browsing: 30 packets, average packet size 1000 bytes, arrival interval 0.05 seconds, protocol TCP / 80, basic urgency level U_b=0.6.

[0033] Video conferencing packet: packet size [1300-1500 bytes], arrival timestamp [0.00s-2.40s], protocol UDP, port 1234; File transfer packet: packet size [1800-2200 bytes], arrival timestamp [0.00s-6.00s], protocol TCP, port 21; Web browsing packet: packet size [900-1100 bytes], arrival timestamp [0.00s-1.50s], protocol TCP, port 80; Video conferencing: average packet size 1400 bytes, standard deviation 150 bytes; average arrival interval 0.08s, standard deviation 0.01s; File transfer: average packet size 2000 bytes, standard deviation 200 bytes; average arrival interval 0.15s, standard deviation 0.03s; Web browsing: Average packet size 1000 bytes, standard deviation 100 bytes; average arrival interval 0.05s, standard deviation 0.005s; Inputting the traffic analysis sequence into the deep learning model, the model achieves a recognition accuracy of 98%, accurately classifying data packets into three categories: video conferencing, file transfer, and web browsing.

[0034] S205. Obtain the basic urgency level of the traffic type; retrieve the transmission parameters of the data packet; calculate the urgency level of the data packet based on the urgency analysis function.

[0035] The transmission parameters include packet loss rate and latency tolerance; the urgency analysis function is: ; Based on the level of urgency; This is the performance sensitivity coefficient; The target latency for the corresponding traffic type; This represents the measured average latency of the most recent N packets; The base coefficient is affected by the packet loss rate; This is the response curvature control coefficient for packet loss rate; Packet loss rate; Sensitivity coefficient for low packet loss rate; The power-law decay intensity; For delay tolerance; For reference delay time; The decay index; It is a very small constant.

[0036] S206. Retrieve the traffic type of data packets within a set time period; filter and integrate data packets with the same traffic type to obtain several sets of data packet sequences.

[0037] S207. Obtain the traffic transmission library; match the traffic types corresponding to several sets of data packet sequences with the traffic transmission library to obtain the message queues for the transmission of several sets of data packet sequences.

[0038] Example: Taking a video conferencing package as an example: ; ; ; ; ; ; The parameters are set as follows: ; ; ; ; ; The calculated urgency level is 0.92.

[0039] Data packets of the same type are integrated to obtain three sets of data packet sequences; the three sets of data packet sequences are then assigned to the video transmission queue, the file transmission queue, and the web page transmission queue, respectively.

[0040] Based on the above steps, the spatiotemporal distribution patterns of data packets are comprehensively characterized by extracting basic and statistical features; the classification mechanism based on artificial intelligence models can automatically learn traffic patterns and adapt to new business or attack traffic; the urgency level is comprehensively evaluated through a multi-parameter weighting function, covering multiple dimensions of requirements such as latency, packet loss, and business tolerance; data packets of the same traffic type are integrated into a dedicated queue to avoid mutual interference between different businesses and improve resource utilization; the transmission order within the queue is adjusted according to the urgency level value to ensure that highly urgent data packets are processed first; and intelligent scheduling reduces manual intervention while meeting users' needs for low latency and high reliability.

[0041] In one possible implementation of the embodiments of this application, combined with Figure 1 ,like Figure 3 As shown, the above S104-S105 can be specifically implemented through the following S301-S303, which are explained in detail below: S301. Retrieve the urgency level of each data packet in several sets of data packet sequences; sort the data packets in the several sets of data packet sequences from highest to lowest urgency.

[0042] S302. Obtain the emergency threshold; queue data packets with an emergency level greater than the emergency threshold into the emergency message queue; allocate the remaining data packets to the corresponding message queues according to the sorting results.

[0043] S303, retrieve the monitoring data of the message queue; adjust the transmission of the message queue based on the dequeue rate adjustment function.

[0044] The monitoring data includes: message queue length and message queue network load; the expression for the dequeue rate adjustment function is: ; The baseline dequeue rate; This is the queue congestion sensitivity coefficient; To normalize queue congestion; Network load sensitivity coefficient; This is a normalized network load pressure indicator.

[0045] It should be noted that the expression for the normalized queue congestion degree is: ;in, This represents the current queue length. The optimal queue length; This represents the maximum capacity of the queue. This represents the growth rate of the queue length. The threshold for stable growth; The expression for the normalized network load pressure index is: ;in, ; For link utilization; Historical packet loss rate; Average queuing delay for messages within the queue; The target delay threshold; , as well as These are the weighting coefficients; This is the steepness coefficient of the Sigmoid function; This is the load-sensitive threshold.

[0046] Example: Taking video conferencing data packets as an example: If one of the 30 packets in a video conference has an urgency level of 0.92, which is greater than the urgency threshold of 0.85, then the corresponding packet is placed in the urgency message queue. The remaining packets are then placed in the video transmission queue according to their urgency level, from highest to lowest.

[0047] The current queue length is 25, the optimal length is 10, the maximum capacity is 50, the growth rate is 0.3, and the stable growth threshold is 0.2; therefore, the normalized queue congestion degree is calculated to be 0.6. The current queue has a link utilization rate of 80%, a historical packet loss rate of 0.2%, an average queuing latency of 15ms, and a target latency of 10ms. , as well as If we set the values ​​to 0.6, 0.3, and 0.1 respectively, the calculated normalized network load stress index is 0.53. ; Then the calculated departure rate is adjusted to 91% of the base departure rate.

[0048] Based on the above steps, by sorting data packets in descending order of urgency, high-priority data packets are always transmitted first, avoiding service failures due to queue waiting; the urgency value calculated based on multiple parameters accurately reflects the urgency of data packets, avoiding the rigidity of traditional priority systems; the emergency queue is independent of the ordinary queue, preventing sudden high-urgency traffic from impacting ordinary services; the dequeue rate is dynamically adjusted through an exponential decay function to balance queue congestion and network load, avoiding resource waste or overload; and both queue and network conditions are considered to prevent erroneous adjustments caused by a single factor.

[0049] Please see Figure 4 The second aspect of this application provides a low-latency transmission control system for an Ethernet switch, including: a data acquisition module, an emergency analysis module, and a queue adjustment module; The data acquisition module is used to acquire several data packets to be transmitted; The emergency analysis module is used to analyze and process several data packets to be transmitted to obtain the traffic type and urgency level; and to allocate several data packets to several corresponding message queues based on the traffic type. The queue adjustment module is used to adjust the transmission of several message queues based on their urgency; monitor several message queues; and schedule the transmission of data packets based on the monitoring data.

[0050] A third aspect of this application provides a computer-readable storage medium storing instructions for performing method steps as described in the first aspect and any possible implementation thereof.

[0051] Some of the data in the above formula are calculated by removing dimensions and taking their numerical values. The formula is the closest to the real situation obtained by software simulation of a large amount of collected data. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0052] The working principle of this application is as follows: This application acquires several data packets to be transmitted; analyzes and processes the several data packets to be transmitted to obtain the traffic type and urgency level; allocates the several data packets to several corresponding message queues based on the traffic type; adjusts the transmission of the several data packets based on the urgency level; monitors the several message queues, and schedules the transmission of the message queues according to the monitoring data.

[0053] The above embodiments are only used to illustrate the technical methods of this application and are not intended to limit it. Although this application has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of this application without departing from the spirit and scope of the technical methods of this application.

Claims

1. A low-latency transmission control method for an Ethernet switch, characterized in that, include: Obtain several data packets to be transmitted; Analyze and process several data packets to be transmitted to obtain the traffic type and urgency level; Distribute several data packets to several corresponding message queues based on traffic type; The transmission of several data packets is adjusted based on their urgency. Several message queues are monitored, and the transmission of message queues is scheduled based on the monitoring data.

2. The low-latency transmission control method for an Ethernet switch according to claim 1, characterized in that, The process of analyzing and processing several data packets to be transmitted to obtain the traffic type and urgency level includes: Retrieve several data packets to be transmitted; extract features from these data packets to obtain basic features and traffic statistics features; among them, the basic features include: packet size, arrival time, protocol type, and port number; The basic characteristics of several data packets are integrated with traffic statistical characteristics to obtain a traffic analysis sequence; the traffic classification model is then called, and the traffic analysis sequence is input into the traffic classification model to obtain the traffic type corresponding to the data packet; the traffic classification model is constructed based on an artificial intelligence model; Obtain the transmission parameters of the data packets; analyze and calculate the urgency of the data packets based on the traffic type and transmission parameters.

3. The low-latency transmission control method for an Ethernet switch according to claim 2, characterized in that, The process of extracting features from several data packets to obtain basic features and traffic statistics features includes: Retrieve several data packets; identify and analyze the fields of the data packets to obtain their basic characteristics; Based on formula Calculate the average packet size of data packets within a specified time period; based on the formula. Calculate the standard deviation of package size; where, This represents the size of the i-th data packet; This sets the total number of data packets within a given time period. Based on formula Calculate the average arrival interval of data packets within a specified time period; based on the formula Calculate the standard deviation of the arrival interval; where, This represents the arrival time of the i-th data packet; The average packet size, standard deviation of packet size, average arrival time, and standard deviation of arrival time are integrated into flow statistics features.

4. The low-latency transmission control method for an Ethernet switch according to claim 2, characterized in that, The traffic classification model is built based on an artificial intelligence model, including: Obtain the standard dataset; the standard dataset includes standard input data consistent with the content attributes of the traffic analysis sequence, and standard output data consistent with the content attributes of the traffic type; Select a model framework and deep learning algorithm from the artificial intelligence library; build and improve the model framework based on the deep learning algorithm to obtain a deep learning model; The standard dataset is divided into a training set, a validation set, and a test set according to a preset ratio; the deep learning model is trained using the training set; the internal parameters of the deep learning model are adjusted using the validation set; and the deep learning model is tested using the test set to obtain test metrics. Obtain the indicator threshold; compare the test indicator with the indicator threshold; if all test indicators are greater than the indicator threshold, then mark the deep learning model as a traffic classification model; otherwise, rebuild and retrain the traffic classification model.

5. The low-latency transmission control method for an Ethernet switch according to claim 2, characterized in that, The method of analyzing and calculating the urgency of data packets based on traffic type and transmission parameters includes: Obtain the basic urgency level of the traffic type; retrieve the transmission parameters of the data packets; among which, the transmission parameters include: packet loss rate and latency tolerance; The urgency level of a data packet is calculated based on an urgency analysis function; the urgency analysis function is: ; in, Based on the level of urgency; This is the performance sensitivity coefficient; The target latency for the corresponding traffic type; This represents the measured average latency of the most recent N packets; The base coefficient is affected by the packet loss rate; This is the response curvature control coefficient for packet loss rate; Packet loss rate; Sensitivity coefficient for low packet loss rate; The power-law decay intensity; For delay tolerance; For reference delay time; The decay index; It is a very small constant.

6. The low-latency transmission control method for an Ethernet switch according to claim 1, characterized in that, The method of allocating several data packets to several corresponding message queues based on traffic type includes: Retrieve the traffic type of data packets within a specified time period; filter and integrate data packets with the same traffic type to obtain several sets of data packet sequences; Obtain the traffic transmission library; match the traffic types corresponding to several sets of data packet sequences with the traffic transmission library to obtain the message queues for the transmission of several sets of data packet sequences.

7. The low-latency transmission control method for an Ethernet switch according to claim 1, characterized in that, The adjustment of the transmission of several data packets based on their urgency includes: Retrieve the urgency level of each data packet in several sets of data packet sequences; sort the data packets in the several sets of data packet sequences in descending order of urgency; Obtain the urgency threshold; queue data packets with an urgency level greater than the urgency threshold into the urgency message queue; allocate the remaining data packets to the corresponding message queues according to the sorting results.

8. The low-latency transmission control method for an Ethernet switch according to claim 1, characterized in that, The scheduling of message queue transmissions based on monitoring data includes: Retrieve monitoring data from the message queue; the monitoring data includes: message queue length and message queue network load; The transmission of messages in the message queue is adjusted based on the dequeue rate adjustment function; the expression of the dequeue rate adjustment function is: ;in, The baseline dequeue rate; This is the queue congestion sensitivity coefficient; To normalize queue congestion; Network load sensitivity coefficient; This is a normalized network load stress indicator. The expression for the normalized queue congestion degree is: ;in, This represents the current queue length. The optimal queue length; This represents the maximum capacity of the queue. This represents the growth rate of the queue length. The threshold for stable growth; The expression for the normalized network load pressure index is: ;in, ; For link utilization; Historical packet loss rate; Average queuing delay for messages within the queue; The target delay threshold; , as well as These are the weighting coefficients; This is the steepness coefficient of the Sigmoid function; This is the load-sensitive threshold.

9. A low-latency transmission control system for an Ethernet switch, applied to the low-latency transmission control method for an Ethernet switch as described in any one of claims 1-8, characterized in that, include: Data acquisition module, emergency analysis module, and queue adjustment module; The data acquisition module is used to acquire several data packets to be transmitted; The emergency analysis module is used to analyze and process several data packets to be transmitted to obtain the traffic type and urgency level. Distribute several data packets to several corresponding message queues based on traffic type; The queue adjustment module is used to adjust the transmission of several message queues based on their urgency; monitor several message queues; and schedule the transmission of data packets based on the monitoring data.

10. A computer-readable storage medium, characterized in that... The computer-readable storage medium stores instructions for performing the method steps of any one of claims 1-8.

Images